Network Security Assessments for Integrating Large ... - IEEE Xplore

2 downloads 440 Views 5MB Size Report
Mar 15, 2013 - Network Security Assessments for Integrating Large-Scale. Tidal Current and Ocean. Wave Resources Into. Future Electrical Grids. This paper ...
INVITED PAPER

Network Security Assessments for Integrating Large-Scale Tidal Current and Ocean Wave Resources Into Future Electrical Grids This paper provides an insight on how the novel schemes of marine power generation can be analyzed under conventional network planning exercises using generic information. By Jahangir Khan, Member IEEE , Daniel Leon, Member IEEE , Ali Moshref, Senior Member IEEE , Saeed Arabi, and Gouri Bhuyan

ABSTRACT | Marine energy, especially tidal current and ocean

South Korea for the years 2017 and 2022. The primary objective

wave resources, bear immense potential for generating renew-

of these high-level grid scenario analysis is to identify the

able power toward meeting global electricity needs. A number

practical potential for longer term large-scale wave and/or tidal

of conversion technologies have been successfully demonstrat-

power generation, particularly focusing on the transmission

ed worldwide and precommercial/commercial deployments

networks. Steady-state (overloading and voltage deviations/

are expected to appear in the near future. While electric power

collapses), time-domain (angular stability and dynamic voltage

utilities foresee renewable technologies as a viable alternative

recovery characteristics), and small-signal (eigenvalue) stability

to fossil fuels, marine energy technologies are generally excluded in their energy planning processes. Lack of techno-

analysis are carried out for these systems. Subsequent to establishing a number of study scenarios, N-1 contingencies and

logical preparedness and unavailability of device information

various suitable violation criteria are prepared, which are used

are two major obstacles in that regard. This article provides an

in identifying the underlying network bottlenecks, especially

insight on how such novel schemes of power generation can be

within the coastal areas. Simplified generic dynamic models of

analyzed under conventional network planning exercises using

tidal current and ocean wave devices are developed for the

generic information. The first study focuses on wave power

time-domain analysis. It is expected that similar studies, once

integration along the coasts of Oregon, USA for the year 2019

conducted by the interested utilities, will equip them better

and analyzes the Northwest electrical network. The second study considers both tidal current and ocean wave resources in

toward considering marine energy in their future portfolios. In addition, device manufacturers and project developers will have greater confidence in the emerging marine energy market.

Manuscript received August 17, 2011; revised March 3, 2012; accepted January 3, 2013. Date of publication March 12, 2013; date of current version March 15, 2013. J. Khan and S. Arabi are with Powertech Labs Inc., Surrey, BC V3W7R7 Canada (e-mail: [email protected]; [email protected]). D. Leon is with the Smart Grid Division, Siemens Canada Ltd., Richmond, BC V6V2K9 Canada (e-mail: [email protected]). A. Moshref is with Power Systems and Testing, BBA Inc., Vancouver, BC V6E3S7 Canada (e-mail: [email protected]). G. Bhuyan is with Bhuyan Consulting, Delta, BC V4C 4G6 Canada (e-mail: [email protected]). Digital Object Identifier: 10.1109/JPROC.2013.2238491

956

Proceedings of the IEEE | Vol. 101, No. 4, April 2013

KEYWORDS | Grid integration; long-term planning; network security; ocean wave; tidal current

I . INTRODUCTION The immensity of the ocean energy resources available throughout various coastal areas in the world bring realistic and justifiable potentials for generating electrical power 0018-9219/$31.00  2013 IEEE

Khan et al.: Network Security Assessments for Integrating Large-Scale Tidal Current and Ocean Wave Resources

Table 1 Highlights of Case Studies

in a sustainable manner. In order to achieve this potential, ocean energy conversion technologies need to advance from concept to commercial deployment. At the current state of these technologies, a plethora of concepts are being tested: some in laboratories, others in open seas. It is expected that multiunit tidal current and ocean wave plants will constitute the norm of the future. Tidal barrage, ocean thermal gradient technologies as well as shore-based large wave power plants will also be a part of this class of energy harvesting [1], [2]. With rapid advancement of these technologies in recent times, it is perceived that the grid integration issues for this variable power generation resource will require due attention in the near future. This is particularly true for largescale deployments where network impact and planning studies need to be conducted as a prerequisite for considering such alternative resources in the future energy mix. The first study focuses on wave power integration along the coasts of Oregon, USA, for the year 2019 and studies the Northwest electrical network. The second study considers both tidal current and ocean wave resources in South Korea for the years 2011 and 2022 (Table 1). The primary objective of these high-level grid scenario analysis is to identify the practical potential for longer term largescale wave and/or tidal power generation, particularly focusing on the transmission networks. Steady-state (overloading and voltage deviations/collapses), time-domain (angular stability and dynamic voltage recovery characteristics), and small-signal (eigenvalue) stability analysis are carried out for these systems. Details of dynamic models of tidal current and ocean wave devices are withheld in this article, which can be found in [3]–[5]. This article provides an insight on how such novel schemes of power generation can be considered under conventional network planning exercises using a set of generic device information. It is expected that this work will be treated as a catalyst toward instigating necessary discussions within the realms of wave power and electrical networks. Considering its exemplary nature and reliance on generic models, this work and produced results should not be treated as a planning study endorsed by associated authorities. Being a high-level analytical study, the underlying assumptions, criteria, scope, and approach need also be considered alongside the study findings. Considering the scope, objective, and nature of this work, large-scale wind energy related studies (such as

[6]–[8]) have been found to be very useful. From wave device modeling point of view, relevant publications (such as [9]–[11]) were reviewed and incorporated accordingly, particularly for induction-generator-type machines. Technology evaluation, wave energy resources, and network highlights for the state of Oregon can be found in [12]–[14]. Prospects and projects related to tidal, wave, and wind power in Korea are discussed in [15] and [16]. General overviews of the USA/Canada Northwest and Korean electrical system can be found in [17]–[20]. Articles discussing broader issues such as cost, economics, and resource mix are discussed in [21] and [22]. Prior to discussing these case studies in Sections IV and V, a brief highlight of various marine energy conversion schemes, technologies, and projects is given in Section II. Also, a generic outline for carrying out similar case studies is given in Section III. It is expected that similar studies, once conducted by the interested utilities, will equip them better toward considering marine energy in their future portfolios. In addition, device manufacturers, project developers, and energy marketers will have greater confidence in the emerging marine energy market.

II. ENERGY CONVERSION AND GRI D INT E GRAT ION The energy contained in the ocean waves is a form of concentrated solar energy that is transferred through complex wind–wave interactions. The effects of Earth’s temperature variation due to solar heating combined with a multitude of atmospheric phenomena generate wind currents at a global scale. Ocean wave generation, propagation, and direction are directly related to these wind currents. On the other hand, ocean tides are cyclic variations in seawater elevation and flow velocity as a direct result of the Earth’s motion with respect to the moon and the sun, and the interaction of their gravitational forces. A number of phenomena relating to Earth rotational tilt, rate of spinning, and interaction among gravitational and rotational forces cause the tides to vary significantly over time and space. Tidal variations are more obvious in coastal areas where constrained channels augment the water flow and increase the energy density. The ocean energy conversion processes can be broadly categorized into: 1) tidal barrage; 2) ocean wave; 3) tidal current; 4) ocean thermal gradient; and 5) salinity Vol. 101, No. 4, April 2013 | Proceedings of the IEEE

957

Khan et al.: Network Security Assessments for Integrating Large-Scale Tidal Current and Ocean Wave Resources

Fig. 1. Tidal current device architecture.

gradient. With the advent of various novel concepts and reported success of several deployments, the ocean energy sector, especially the field of tidal current and wave energy conversion technology, has gained significant attention throughout the world.

A. Marine Energy Converters The interactions between fluid motion (tidal current or wave formation) and front–end conversion unit result in primary energy transfer from the resource field to the device itself. For most tidal current devices, this is realized through blade spinning, which is directly coupled to the rotary electrical machines. In contrast, ocean wave devices might utilize more than one degree of freedom in capturing kinetic and/or potential energy of the ocean water. Each of these degrees of motion may contribute significantly in overall power conversion. Similar to wind energy converters, the power captured by a tidal current turbine is proportional to the rotor swept area, water density, velocity cubed, and rotor efficiency. The conversion efficiency of the turbine is termed ‘‘coefficient of performance Cp ’’ and is formulated through a normalized expression involving tip speed ratio  and blade pitch angle  (Figs. 1 and 2; [23]).

Fig. 2. Energy conversion and control of a tidal current device.

958

Proceedings of the IEEE | Vol. 101, No. 4, April 2013

The power captured by a wave energy device (under specific wave regimes) is traditionally given kilowatt per meter of wave crest. The capture width ratio Rw stands as a measure of the wave energy device’s conversion efficiency. This quantity can be interpreted as the dimension of effective wave field compared to the physical length of the converter. In other words, the capture width ratio can be used to indicate how much energy a physical wave device can absorb. A ratio of unity implies that a wave device harnesses all the energy contained in a wave front having the same width as that of the device itself. When the capture width ratio is multiplied with device length, overall power output in kilowatt is realized (Figs. 3 and 4; [23]). The capture width ratio Rw is a function of normalized system frequency (ratio of natural frequency of oscillation to wave frequency) and device control variable. The latter quantity can be either mass, damping, or stiffness (Fig. 4). The primary objective of the secondary converter is to translate all front–end motions into one rotary motion, which provides torque input to the electrical machines placed at the tertiary/final stage. In addition, short-term energy storage and dynamic control schemes can be enforced through this conversion stage.

Khan et al.: Network Security Assessments for Integrating Large-Scale Tidal Current and Ocean Wave Resources

Fig. 3. Ocean wave device architecture.

In most tidal current, system such intermediate conversion schemes are not considered. Except for a few selected wave energy devices, most of the wave technologies use some form of intermediate converters. Use of compressed air, sea water, or pressurized hydraulics is being widely considered. In a typical system, the motion induced in the primary stage drives the fluid into a short-term storage chamber. This fluid is released in a controlled/ uncontrolled fashion to drive a turbine or hydraulic motor, which is coupled to an electrical machine. In addition to the electromechanical systems, components such as power electronic devices, transformers, transfer switches, and protection equipment facilitate the interconnection between a device and the network. In most wave and tidal current devices, the input mechanical torque drives the rotary machines. However, some direct drive systems do not use any intermediate stage and linear movement of the front–end converter is directly coupled to a permanent-magnet-based linear generators. Strictly from an electrical machines’ point of view, all marine energy systems can be categorized under the following four classes (Fig. 5): • induction machines (no power electronic converters; directly connects to the grid with/without transformer; operates at super-synchronous speed

for positive power generation; consumes reactive power); • synchronous machines (no power electronic converters; directly connects to the grid with/without transformer; operates at synchronous speed for positive power generation, produces/consumes reactive power); • full power electronics (employs either induction, synchronous, permanent magnetVlinear/rotarydirect/geared machines, converter acts as the grid interface with/without transformers, operates at synchronous frequency at the grid side and asynchronous frequency at the generator side, produces/consumes reactive power); • doubly-fed induction machines (employs induction machines with partially rated converter on the rotor side, stator connects directly to the grid with/ without transformers, operates at synchronous frequency at the grid side and asynchronous frequency at the generator side, produces/consumes reactive power). It is expected that a future wave/tidal power plant will consist of arrays of individual devices similar to wind farms. However, unlike wind turbines, where each turbine contains only one unit of electromechanical machinery, a

Fig. 4. Energy conversion and control of an ocean wave device.

Vol. 101, No. 4, April 2013 | Proceedings of the IEEE

959

Khan et al.: Network Security Assessments for Integrating Large-Scale Tidal Current and Ocean Wave Resources

(Fig. 7). The topology of an offshore collection network is heavily dependent on a number of factors, including: size of the plant, cable length/cost, subsea components, need for redundancy, power loss minimization as well as installation/retrieval options. In general, four network topologies can be identified with this regard: 1) point-topoint; 2) radial; 3) ring; and 4) star topology (Fig. 8; [23]). The point-to-point topology is most suitable for smallscale plants where each conversion unit would have a dedicated cable and associated accessories. While this option provides the highest redundancy, the length of the cable as well as procurement, installation, and retrieval costs are highest compared to the other alternatives. The radial topology implements a ‘‘daisy chain’’ approach where subsequent to each unit, cable capacity needs to be increased. Terminals from two or more units need to be interfaced through hubs/junctions/joints or low-voltage or medium voltage (LV/MV) transformers. The ring topology incorporates looping of cables (consequently incurring higher cost) using additional switches. On the other hand, the star topology groups a number of units using cables of similar rating, and transfers the collected power using higher rated cables to the shore.

Fig. 5. Electrical machines and power converters for marine energy systems.

single wave/tidal unit may contain multiple core converters (machines). This analogy of ‘‘machine–unit–plant’’ is shown in Fig. 6. It should be mentioned that a reverse concept is also possible, where a number of front–end and corresponding intermediate converters provide power to drive a single-grid-interfaced machine.

B. Offshore Collection Network For tidal current and ocean wave devices, it is expected that an offshore electrical system consisting of a subsea electrical network and associated components will connect the tidal/wave power plants to the onshore substations

C. Technological Progression Similar to any other area of energy engineering, development of ocean power devices is a cost-intensive multidisciplinary iterative process. A typical development cycle starts with conceptual design and small-scale testing in a laboratory or at sea. Subsequent intermediate designs are further refined before realizing precommercial or commercial full-scale systems. While most of the marine energy devices are at concept, prototype, or part-scale test levels, some devices are currently going through fullscale demonstration process with expectations that precommercial deployments will be realized in the near future. It is reported that at the end of 2010, the total installed marine power capacity is around 6 MW (ocean wave: 2 MW; tidal current 4 MW), mostly in the form of demonstration projects [24]. Highlights of a number of such projects are given in Table 2. A further review of marine energy devices, projects, and deployments can be found in [24]–[26].

I II . CASE STUDY: A GENERIC OUTLINE

Fig. 6. Outline of a marine energy plant deploying multiple units.

960

Proceedings of the IEEE | Vol. 101, No. 4, April 2013

Each network security analysis can be driven by different objectives, based upon diverse assumptions and carried out using multiple sets of techniques. In the context of the large-scale longer term marine power generation, such objectives may include: • identifying the ‘‘baseline marine power capacity,’’ i.e., the amount of marine energy that can be added into the electrical system without requiring any significant transmission resource addition;

Khan et al.: Network Security Assessments for Integrating Large-Scale Tidal Current and Ocean Wave Resources

Fig. 7. Outline of offshore networks for marine energy plant.





determining the ‘‘network bottlenecks,’’ i.e., the constraining factors that may pose restrictions on further ocean wave and tidal current power generation beyond the baseline capacity; indicating the ‘‘suitable points of interconnections (POIs),’’ i.e., the target areas, substations, and buses that have significant capacity for marine power addition (from an electrical system point of view).

The underlying assumptions for such studies, including the ones presented here, are as follows. • It is assumed that the marine energy technologies, in general, will be commercially available for the studied time period. For numerical model development purposes, only one class of generic ocean wave and tidal current device will be considered. • It is assumed that the powerflow cases and associated dynamic data are representative of the actual systems that are expected to be in place for the given time horizon. This implies that projected load growth, generation growth, demand side management targets, and network expansion/ reinforcement plans are embedded within the base cases. • The criteria violations existing in the base cases (without any new modification, i.e., addition of a new generation) are assumed to be subject to further scrutiny and mitigation (by means of reinforcement or protection schemes) by the relevant authorities. Once addressed, the electrical network is expected to be more robust (allowing greater share of marine power addition), which makes the outcome of this study to be of conservative nature. The network security studies are expected to provide input to a broad range of audience and may need to be interpreted on a broader holistic scale. Keeping this in mind, a two-tiered study approach is generally followed: Consultation: At the onset of the study, a scenario team consisting of representatives from relevant utilities and authorities is formed. The scenario team facilitates data exchange, aids in defining the study scenarios, and provides necessary oversight toward conducting the study.

Fig. 8. Network topologies for offshore collection systems.

Assessment: The technical investigation commences with sanity checking of powerflow base cases and dynamic data files. Also, a set of suitable POI buses are identified for addition of marine power plants. Subsequently, transfer scenarios, contingencies, and criteria are established. In a typical study, the technical investigation falls within three broad classes (Fig. 9). • Steady-state analysis: This part of the study primarily aims at identifying the effects of marine Vol. 101, No. 4, April 2013 | Proceedings of the IEEE

961

Khan et al.: Network Security Assessments for Integrating Large-Scale Tidal Current and Ocean Wave Resources

Table 2 Selected Test Facilities and Demonstration Projects

power addition in the forms of overloading and voltage deviations/collapses that may occur in neighboring network components. • Time-domain analysis: The time-domain study focuses on the angular stability and dynamic voltage recovery characteristics under various transfer conditions and contingencies. The development of the dynamic numerical model of marine power devices is also a prerequisite for this analysis. • Small-signal stability analysis: Being a special type of study, this exercise can effectively identify potential stability issues of a large power system owing to insufficient damping of electromechanical oscillations. In Fig. 9, a generic outline of such network security assessment methodology is presented. Depending on the objectives and needs (especially for small-scale integration), voltage-security assessment and small-signal analysis may be excluded. Each step of the process is associated with various criteria and relevant observations, which constitute the final set of results.

IV. CASE STUDY 1: WAVE ENERGY IN OREGON, USA Ocean wave energy resources along the coast of Oregon bear tremendous potential for generation of electricity in a clean and environmentally friendly manner. In the context of the state of Oregon (as well as the Northwest electrical system), a ten-year time horizon (from year 2009) has been agreed to be the time frame of interest. Even though the powerflow solutions are obtained by solving the 962

Proceedings of the IEEE | Vol. 101, No. 4, April 2013

western electrical system as a whole (i.e., no reduction conducted) [17], relevant parameters are monitored only for the Northwest area and the state of Oregon, in particular.

A. Base Case Description 1) General Description: A set of two powerflow base cases (heavy summer and heavy winter 2009) were used throughout this study. In general, North American electricity consumption patterns are exhibited through these dominant peaks, which coincide with seasonal variations. Given that the electrical systems and associated components experience relatively higher stress during these conditions, network models reflective of summer/winter conditions are widely used for similar system planning studies. The base cases used here are of Western Electricity Coordinating Council (WECC) year 2019 genre [17] and reflect the projected load, generation, and network conditions. In addition, two dynamic data sets were used for the transient-security analysis. Highlights of these cases are given in Table 3. The WECC 2019 heavy summer approved base case (June 10, 2009) and 2018–2019 heavy winter base case (August 7, 2009) consist of 21 areas, 415 zones, and 284 owners. For the purposes of this study where the coastal region in Oregon is of interest, area #40 (Northwest) and zones 401 (Portland), 402 (Western Oregon), and 411 (PacifiCorp) are of the highest relevance. The Northwest area is neighbored by six other areas, which are: BC Hydro, Idaho, Montana, Pacific Gas & Electric, LA Department of Water & Power, and Sierra Pacific Power. The WECC base

Khan et al.: Network Security Assessments for Integrating Large-Scale Tidal Current and Ocean Wave Resources

models consist of both alternating current (ac) and direct current (dc) systems. For the Northwest area, the projected loads for summer and winter peaking periods are around 29 and 35 GW, respectively. Corresponding generation capacities within this region are around 35 and 43 GW. 2) Coastal Region: As reflected by the 2019 base case models, the coastal regions are characterized by little or no generation sources. In other words, there is no expected or planned new generation from the coastal regions within a ten-year time horizon. The existing coastal transmission network is shown in Fig. 10(a). The load centers along the coast of Oregon are primarily supplied by Bonneville Power Administration’s (BPA’s) 230- and 115-kV transmission network. Starting from the North–South 500-kV BPA backbone, the power flow direction is toward the West. Along the coastline this flow is generally directed to the South. On a broader scale, generating stations in the North and in the Northeast supply the major load centers in Oregon, whereas the flow is primarily through the I5 corridor (North–South) and cross-cascade South interfaces (East–West). It has been observed that the projected accumulated coastal load is in the range of 600 MW for the summer peak condition, and 850 MW for the heavy winter case. With the addition of wave-power-based generating stations along the coastline [target areas are indicated in Fig. 10(b)], it was anticipated that the general direction of power flow would face a reversal (within the coastal network).

Fig. 9. A generic outline of the network security assessment.

Table 3 Summary of Summer and Winter Powerflow Base Cases

3) Significant Changes: Considering the study focus on the Northwest region, significant additions from 2007 to 2019 period for this area are listed below (inclusive but not limited to): • wind plant addition: Saddle Back 70 MW, Hey Canyon 200 MW, Miller Ranch 100 MW; • thermal plant addition: Cherry Point combined cycle combustion turbine (CCCT) 560 MW; • transmission line (230 kV): Connection of existing power substations at Walla Walla, Wallula, and McNary; Covington-Berrydale; Sedro WoolleyHorse Ranch #2; IP line converted from 115 to 230 kV; • transformer addition (230/115 kV): North King County (Novelty), Pierce County (Alderton), Thurston County (St Clair), Lake Tradition. Additional key changes for the neighboring Balancing Authorities include: • Gateway West (Idaho Power & PacifiCorp); • Hemingway Boardman (Idaho to Northwest); • Hemingway to Captain Jack (PacifiCorp); • Montana Alberta Intertie. Further details of the base cases can be found through relevant authorities such as WECC [17] and BPA [18]. Vol. 101, No. 4, April 2013 | Proceedings of the IEEE

963

Khan et al.: Network Security Assessments for Integrating Large-Scale Tidal Current and Ocean Wave Resources

Fig. 10. Oregon coast and the transmission network.

B. Scenario Setup 1) POI and Transfers: The process of POI identification reflects the collective views of the project and scenario team, contemporary wave power projects being proposed (such as [12] and [13]), as well as characteristics of the electrical network (expected weak/strong points). In many cases, POIs closer to the shoreline are owned by the host utilities, whereas higher capacity connection points are generally farther inland and owned by BPA. Subsequently, a set of six geographical target areas (consisting of 12 POIs) have been analyzed. These are shown in Fig. 10 and also listed in Table 5. At any given instance, the total power generation must equal the total consumption (load and losses). In order to add a new generation in certain locations, it is therefore necessary to decrease the power production in other locations (or to increase the loads) in order to maintain the power balance. In this study (with load levels predetermined in the base cases), the transfers are established such that gradual increase in ocean power generation along the coastal POIs is balanced against a similar decrease in the conventional generation in a set of remotely located plants. These plants are selected based on various criteria, such as plant location (relative to the intended transfer directions) and provisions for 964

Proceedings of the IEEE | Vol. 101, No. 4, April 2013

eliminating greenhouse emissions. Also, it is customary to select larger power plants for ease of defining the transfers. Based on these selection methods, a set of coaland natural-gas-based power plants have been chosen (Table 4 and Fig. 11). Typically, the coal fired plants are expensive to operate, emit greenhouse gases, and are older in the operational age. Therefore, these units are selected as priority units to be scheduled as shown in Table 4. The ‘‘East–West (EW)’’ transfer essentially represents the flow along the ‘‘West of Slatt’’ flow gate, whereas the ‘‘North–South (NS)’’ transfer represents that of the ‘‘South of Allston’’ interface. The ‘‘Combined North & East (CNE)’’ interface consists of large coal powered units belonging to both north and eastern locations (Fig. 11). Under all transfer scenarios, the increase of power generation is introduced through the coastal ocean power plants. The selected units under each transfer are dispatched using a predefined order (as against sharing the generation reduction among the units equally) as indicated in Table 4. An initial estimate indicated that the total wave power generation would be around 2000 MW, and these units are selected such that this bulk power can be adequately consumed by the overall system.

Khan et al.: Network Security Assessments for Integrating Large-Scale Tidal Current and Ocean Wave Resources

Table 4 Transfer Description

2) Contingencies: In addition to the transfer scenarios discussed above, a set of contingencies (single element outage, i.e., N-1 contingencies) are used. From the POI buses, N-1 contingencies are considered up to five tiers (i.e., five buses away from the POIs, total 132 branch outage and 57 generator tripping contingencies) of the system. The contingencies include three-phase faults at buses cleared by single circuit tripping, and single generator

tripping without fault. The contingencies were applied for lines rated at 69 kV and above, and for generators rated at 50 MVA and above. Details of these contingencies can be found in [4]. In the transient-security analysis, fault clearing times for the line outage contingencies are specified using their kV ratings (below 100 kV: 14 cycles; 100– 138 kV: nine cycles; 161–230 kV: seven cycles; above 345 kV: four cycles), as found in BPA’s guidelines on technical requirements for interconnection [18]. 3) Criteria: Under the transfer scenarios and contingencies described earlier, the performance of the network elements needs to be analyzed against a set of criteria. The criteria applied on steady-state and time-domain analysis are as follows. a) Steady-state/voltage-security analysis criteria: • voltage deviation (decline or rise): for single contingency, not to exceed 7% at any bus; • branch overload: for the heavy summer case, 100% continuous rating for precontingency and 100% emergency rating for postcontingency conditions; for the heavy winter case, similar ratings are used, except that different rate tables are followed. The branch overloading criteria are given higher priority than the voltage deviations, especially if the latter violation takes place at 69- and 115-kV buses. However, sufficient attention is paid toward evaluating the cause, effect, and extent of such voltage deviations. b) Time-domain/transient-security analysis: • transient stability: system remains stable for all the specified contingencies; • transient voltage dip: for single contingencies, not to exceed 20% for more than 20 cycles at load buses. These criteria are in-line with standard practices as specified by relevant authorities such as WECC and BPA.

Fig. 11. Transfers and POIs.

4) Monitors: For all the scenarios and contingencies analyzed in this study, all the branches and units included Vol. 101, No. 4, April 2013 | Proceedings of the IEEE

965

Khan et al.: Network Security Assessments for Integrating Large-Scale Tidal Current and Ocean Wave Resources

in the nearby system to the POIs (zone 401: Portland; zone 402: Western Oregon; and zone 411: PacifiCorp) are monitored in order to identify any violation of the aforementioned criteria. 5) Study Method: In order to achieve the study objectives outlined earlier, the following approach was taken. Step 1) POI evaluation: Identify individual POI capacities through steady-state analysis under N-1 contingencies, defined criteria (voltage rise/decline and branch overload), and one transfer scenario (East–West transfer, Table 4 ). Step 2) Aggregated capacity evaluation (preliminary): Identify aggregated coastal generation capacity from all POIs through steady-state analysis, using the limits found in Step 1) under N-1 contingencies, defined criteria (voltage rise/decline and branch overload), and all transfers (EW, NS, and CNE transfers). Step 3) Aggregated capacity evaluation (final): Further evaluate the aggregated coastal generation capacity from all POIs through dynamic analysis, using the limits found in Step 2) under N-1 contingencies, defined criteria (transient stability and voltage dip), and one transfer (EW transfer). As a part of the powerflow base case analysis and sanity checking (data quality, convergence, etc.), the above set of criteria was used to identify the inherent violations existing in the 2019 heavy summer and winter files. These violations are independent of the issues related to ocean power integration. From a system reliability perspective, it is expected that these violations will be addressed by means of reinforcement, reactive compensation, and/or protection schemes. This implies that the 2019 electrical network, in fact, will be more robust than the base cases being studied, allowing more wave power resources to be added into the system. In other words, the results of this assessment, albeit realistic, will be of conservative nature.

C. Steady-State/Voltage-Security Analysis The voltage-security assessment is conducted in multiple steps under a number of scenarios (using various transfer conditions and individual/aggregated POIs). At first, powerflow base cases are modified to accommodate wave power generators. A total of 12 POIs (ten originally suggested and two considered as an outcome of preliminary analysis) have been implemented. Each OWEC device is modeled as a combination of a generator (with Qmax ¼ 0 MVAr, Qmin ¼ 60% of MVA rating) and a shunt capacitor (with continuous control capability up to 60% of the generator’s MVA rating). The OWEC output is at 0.6 kV, and is connected to the transmission bus using a 33-kV collector bus. Also, the 966

Proceedings of the IEEE | Vol. 101, No. 4, April 2013

contingencies, criteria, and monitor parameters are defined in accordance with the aforementioned setup. 1) POI Evaluation: For the purposes of identifying the favorability and capacity of various POIs, at first, the EW transfer (Table 4) is considered. It should be mentioned that as long as the investigation is focused on the POIs themselves, the transfer option is immaterial. Maximum generation capacity for the OWEC generators is set preliminarily around 300 MVA (exceptions: 500 MVA for Toledo & Tahkenitch and 700 MVA for Wendson POI). During the transfer analysis, this generation is increased with steps of 20 MW and corresponding impacts on the neighboring system are analyzed (with and without contingencies). It has been established that even though line overloading and bus voltage violations were taken as limiting criteria, only overload conditions were reported in most cases. Also, in spite of having higher level of coastal load in the heavy winter case (Fig. 10), the heavy summer case appeared to be more constraining. Primarily, this is due to the fact that power flow directions are different and are more limiting to an addition of a new generation, because summer ratings are lower than winter ratings (owing to higher ambient temperature). 2) Aggregated Capacity Evaluation (Preliminary): A direct summation of wave power generation capacities (for each of the POIs, as identified in the previous step) does not necessarily determine the aggregated capacity from the coastal region, as a whole. This arises from the fact that with the addition of multiple plants throughout the coast, the power flow direction and magnitude are altered in unique ways. This relaxes or tightens the associated constraints on the network elements. Therefore, a separate step needs to be undertaken where aggregated capacity limits can be evaluated (to be further scrutinized under a dynamic study platform). Under this step, the maximum capacity of ocean power plant at each of the POIs is set according to their respective limits, as identified in the previous step (POI evaluation). The transfer scenario is defined such that generation increase along the coastal POIs is incremented with steps of 20 MW, each POI reflecting a fraction of this generation based on its maximum allowed capacity. Also, generation decrease is scheduled for all three transfer scenarios (EW, NS, and CNE transfers, separately). An inspection of the results indicated that the first bottleneck in the heavy summer case is exhibited in the form of line overloading of ‘‘Glasgow 115 to Hauser 115’’ line under the contingency: ‘‘Outage Branch ¼ Alvey 500 Dixonvle 500’’ [4]. Corresponding maximum wave power generation capacity is 430 MW. Under the precontingency condition, the same element is affected/overloaded. However, the maximum transfer limit is 200 MW higher.

Khan et al.: Network Security Assessments for Integrating Large-Scale Tidal Current and Ocean Wave Resources

Similar observations can be made for the heavy winter case. It was also observed that the transfer scenarios do not affect the underlying findings and the bottlenecks are primarily localized within the coastal region (between ‘‘Sumner C115’’ and ‘‘Hauser 115’’).

D. Time-Domain/Transient-Security Analysis 1) Case Preparation: At first, the base cases were analyzed with an intention to identify existing transientsecurity problems (irrespective of addition of newer/wave power plants) in the neighborhood of the selected POIs. Subsequently, newer cases were set up accommodating the maximum transfer capacity identified in the steady-state study (430 MW considering wave generation at all POIs simultaneously). This scenario analysis was performed with an ability to identify any new potential contingency that could cause transient-security criteria violation (as a result of addition of the wave power plants). Dynamic models of the ocean wave devices were considered in this step. The selected transfer for the transient stability study is the CNE transfer (considering only coal fired plants). 2) Aggregated Capacity Evaluation (Final): Based on the results of the time-domain analysis, it was concluded that the heavy summer case did not exhibit any criteria violation with addition of wave power plants along the coastal POIs (maximum aggregated capacity 430 MW). On the other hand, the winter case indicated only one contingency that caused voltage criteria violation. Also, it was observed that this voltage criteria violation could be solved by adding additional shunt compensation at the critical buses. This contingency is further analyzed in the following discussions and reflects observations only in the winter case, considering a maximum generation capacity of 430 MW from ocean power plants. Fig. 12 shows the bus voltage criteria violation for the contingency: ‘‘3PHBF Fairview230VRogue230’’ and the weakest voltage recovery is characterized by a rise up to

Fig. 12. Bus voltage criteria violations observed at selected buses.

Fig. 13. Ocean wave generator speed observed at all POIs.

0.8 p.u. in 23 cycles (considered criteria ¼ 20 cycles). Also, the postcontingency steady-state voltages magnitudes are below 0.9 p.u. for the selected buses, as shown in this figure. Additional shunt compensation on one or more of these critical buses can be an option to solve these voltage violations. With regard to the implementation of wave device models, a set of generator speed curves are presented in Fig. 13, which indicates successful model performance (60-Hz rotor speed). Fig. 14 shows the generator relative rotor angles for a set of units located in the neighborhood of the wave power units under the contingency ‘‘3PHBF Fairview230V Rogue230’’ and ‘‘3PHBF Santiam 230,’’ which also indicates good transient stability characteristics. Considering these results, it can be concluded that from a transientsecurity perspective, addition of 430-MW wave power (as established in the steady-state study) is within the prescribed criteria.

E. Results According to this grid scenario analysis, we have the following conclusions. • Baseline wave power capacity: Considering simultaneous wave energy power generation from selected target areas along the coast of Oregon, the aggregated capacity transfer limit from West to East is found to be approximately 430 MW. This threshold of capacity addition is a conservative estimate. Further evaluation (refined/relaxed criteria and contingencies), wave resource specific POI selection (as against considering all the POIs simultaneously), and addressing the inherent network bottlenecks (as embedded within the 2019 network models) would undoubtedly indicate higher capacity for wave energy resource additions. • Network bottlenecks: Under the scope of this study, with its underlying assumptions and criteria, it has been identified that the primary limiting factor is line overloading. In order to address these limits, several transmission lines Vol. 101, No. 4, April 2013 | Proceedings of the IEEE

967

Khan et al.: Network Security Assessments for Integrating Large-Scale Tidal Current and Ocean Wave Resources

Fig. 14. Generator relative rotor angles for units located near the wave power plants.



and/or transformers near several POIs may need reinforcement/addition. Also, addition of reactive compensation may be deemed necessary at several substations in order to address possible voltage rise/decline issues. Such requirements would be identified during the interconnection study and/or transmission service study process associated with a specific wave resource addition.It is noteworthy that issues of line overloading were experienced in the base case analysis prior to the addition of any wave energy resources. These results indicate that localized system upgrades will be needed to address already anticipated changes in load and resources. Although it was not assumed in the studies, local transmission owners and distribution utilities will make system upgrades necessary to address these issues. Those changes may help eliminate some of the limiting factors identified in this study and thereby increase the individual POI and aggregated capacity transfer limits. Evaluation of POIs: A set of 12 POIs were evaluated and the capacity levels outlined in Table 5 are representative of each of the POIs when considered separately. (Note: The resulting simultaneous

Table 5 POI Capacities for Added New Wave Power

968

Proceedings of the IEEE | Vol. 101, No. 4, April 2013

capacity transfer limit is discussed under the ‘‘baseline wave power capacity’’ heading.) The preceding analysis estimates the amount of wave power that can be interconnected to the electrical system at specific points without requiring any significant transmission infrastructure additions. Though local reactive compensation resource additions would likely be needed and are not trivial in cost, they do not rise to the level of transmission line additions in either scope or cost. The study also estimates the transfer capability of the electric system eastward from the coastal areas. However, depending on the amount of wave power being injected into the system at any given location, local system upgrades would be likely and interconnection facilities would be required to integrate the project. This baseline capacity limit accounts for the effects on the electrical network as a result of wave power generation at multiple locations throughout the Oregon coastline.

V. CASE STUD Y 2 : TI D AL A N D WA VE EN ERG Y I N K ORE A In the context of an ever-increasing demand for electrical energy, the Korean electricity system is undergoing a number of restructures and enhancements. The need for clean, renewable, and reliable power is being addressed in conjunction with ambitious pursuits for economic development and industrial growth. The promise of renewable energy, especially tidal current and wave energy, is also being heavily underscored. With the advent of a number of technologies, developed either in Korea or internationally, the long-term prospects appear to be immense. For the Korean case study, near-term and long-term time horizons are agreed to be projections for 7 and 12 years from the year when this study was conducted, i.e., year 2010, respectively. Details of this study can be found in [27].

A. Base Case Description 1) General Description: Two sets of powerflow base cases for the years 2017 and 2022 (both under light and peak loading conditions) were provided for the Korean system study. Initially, these cases contained two more sets for the Jeju Island and for the mainland, which were subsequently combined into one interconnected system. The major highlights of these cases are summarized in Table 6. The Korean electrical power system can be characterized by (Fig. 15) the following. • A large island system without any interconnection with neighboring countries, constrained with limited land usage, right-of-way, and environmental considerations. • Year when this study was conducted, i.e., year 2010, peak demand, average demand, and generation capacity are in the order of 55, 65, and 75 GW, respectively.

Khan et al.: Network Security Assessments for Integrating Large-Scale Tidal Current and Ocean Wave Resources

Table 6 Power Flow Summaries of the Combined Korean Base Cases









The bulk share of the power is generated by nuclear power plants and coal power fired steam plants having capacities in the order of 1000 and 500 MW, respectively. Load centers primarily concentrated in metropolitan areas whereas generation stations are dispersed over the coastal regions throughout the country. The major transmission backbone is primarily rated at 345 kV with recent addition of several 765-kV lines. The local transmission systems are at 154 kV whereas most 66-kV lines are either being removed or upgraded to higher voltages. The Jeju Island is connected to the mainland through 180-kV dc link (double circuit). Traditionally, transmission problems in the Korean network have been associated with ther-

mal limits, voltage stability, or transient stability. In particular, line overloading at 154 kV and voltage regulation at some 354-kV substations are of concern. • A meshed electrical network with significant power flows taking place through the South–North interconnection. The Korean generation, transmission, and distribution resources are traditionally grouped into seven geographical areas (Gyeongin North–South, Jingu, Honam, Yeongong, Yeongnam, and Jeju Island). The Gyeonging area in the Northwest accounts for more than 40% of the national demand (includes load centers such as Seoul and Inchon) whereas the Yeongnam areas in the Southeast typically exhibit the second highest consumption (more than 30%, includes cities such as Busan and Taeku).

Fig. 15. Korean power system: Generation and load centers.

Vol. 101, No. 4, April 2013 | Proceedings of the IEEE

969

Khan et al.: Network Security Assessments for Integrating Large-Scale Tidal Current and Ocean Wave Resources

Table 7 Significant Changes in the Future Korean System

2) Coastal Region: The Korean peninsula is rich with the potential for various types of renewable energy resources. With a commercially operating tidal barrage plant in Lake Sihwa (250 kW), the pursuit for additional barrages is on, especially in the northwestern part of the country. Wind resources are spread throughout the coastal region with significant potential in the Juju Island areas as well as in the Southeast. Currently, about 200 MW of wind power is connected to the Korean grid [20]. The potential for tidal current and ocean wave is mostly concentrated around the southwestern part of Korea. With regard to tidal current power, various target areas within this region include: Daebang, Uldolmok, Wando, Samchonpo, Changjuk, and Maeggol. For wave energy utilization, various locations around North and West of the Jeju Island have been identified. The theoretical tidal current and coastal wave energy potentials account for about 1000 and 650 MW (þ offshore 5000 MW), respectively [16]. However, it remains to be seen how the technological advancements fulfill the desired targets. 3) Significant Changes: In addition to a number of addition/upgrades with regard to generation stations, substations, and transmission lines (especially 345-kV lines in the Seoul area) the following notable changes (compared to current/2009 network) are incorporated in the base cases (Table 7 and Fig. 16): In addition to the major network changes listed above, a number of equipment/substation upgrades, line reenforcements, and component additions are expected for the future Korean network [19].

B. Scenario Setup 1) POI and Transfers: A total of 1620 MW of the new renewable generation is added at six new buses, as specified in Table 8, which are connected to the grid by 154-kV double circuit lines. In order to assess the voltage stability margin of the system, the additional generation is increased in steps of 20 MW , first in the island and then in the mainland, which is offset by scaling up the load in areas 1 through 5 (i.e., Seoul, South Seoul, Incheon, North 970

Proceedings of the IEEE | Vol. 101, No. 4, April 2013

Geonggi, Geonggi). In transient and small-signal simulations, the final power flows (i.e., after addition of all renewable generations, if applicable) are used. Note that in the light load cases, no more than 900 MW of new generation can be added in Jeju Island due to reaching the 700-MW limit of the high-voltage direct current (HVDC) capacity, resulting in total addition of 1520 MW (Fig. 17 and Table 8). In addition to the above renewables, existing/planned wind power at POI buses 120 and 122 (150 and 80 MW, respectively) in Jeju Island is also considered. 2) Contingencies: In this study, the system is scanned for single contingencies. The resulting contingency numbers applicable to various types of studies are as specified in Table 9. 3) Criteria: The following criteria were within the steady-state and time-domain analyses. a) Steady-state/voltage-security analysis criteria: • branch overload: 120% of rating A for precontingency and 120% of rating B for postcontingency situations, applied to branches bigger than 100-kV buses; • voltage magnitude (min/max): not to violate 0.95/ 1.1 pu at precontingency and 0.9/1.1 pu after single contingency, applied to branches bigger than 100-kV buses; • voltage change (decline or rise): not to exceed 6% for single contingency, applied to branches bigger than 100-kV buses; • voltage stability (collapse) margin: not less than 5% for single contingency (i.e., 81 MW for 1620-MW new injection). Switched shunt and under-load tap changer (ULTC) controls are activated in both precontingency and postcontingency situations. b) Time-domain/transient-security analysis: • Transient stability: system remains stable for the specified contingencies having three-phase faults cleared after five cycles; • transient voltage dip (TVD): for single contingency not to exceed 20% for more than 20 cycles at load buses.

Khan et al.: Network Security Assessments for Integrating Large-Scale Tidal Current and Ocean Wave Resources

Fig. 16. Changes in the Korean system by 2017 and 2022, compared to 2009.

For small-signal stability, a minimum damping ratio of 3% is considered. 4) Study Methodology: This study accommodates voltagesecurity, transient-security, and small-signal stability analysis on the Korean interconnected network for years 2017 and 2022, under both peak and light loading conditions. It is assumed that, before 2017, Jeju Island will be connected to the Korean mainland through two HVDC submarine transmission links, totaling a maximum capacity of 700 MW in either direction. Two locations of ocean wave

energy generation in the island and four locations of tidal current flow generation in the mainland are considered. The maximum new generation injections in the island and mainland are 1000 and 620 MW, respectively, to be dispatched against forecasted load increases in certain areas of the mainland.

C. Steady-State/Voltage-Security Analysis The steady-state situation is analyzed from voltagesecurity point of view, which consists of branch overload, bus voltage magnitude (min/max), bus voltage change

Table 8 New Renewable Generation Interconnection

Vol. 101, No. 4, April 2013 | Proceedings of the IEEE

971

Khan et al.: Network Security Assessments for Integrating Large-Scale Tidal Current and Ocean Wave Resources

Fig. 17. POIs and transfer scenarios for wave and tidal power addition.

(both decline and rise), and voltage stability (collapse) analysis. System loads are represented by constant power models for both active and reactive components. The voltage stability margin is applied through PV (voltage versus active power) analysis, namely by increasing the applied transfer in precontingency and postcontingency situations, until a converged powerflow solution

cannot be obtained. This is not dependent on any particular bus and can be seen on the corresponding curves of any bus. With all the equipment in service, the total of 1520 MW of the renewable resources can be added to the 2017 light case without any voltage violation. With single contingencies, however, the security limit is 700 MW. The

Table 9 Number of Applied Contingencies to the Four Cases in Various Types of Studies

972

Proceedings of the IEEE | Vol. 101, No. 4, April 2013

Khan et al.: Network Security Assessments for Integrating Large-Scale Tidal Current and Ocean Wave Resources

Fig. 18. PV curves for the limiting contingency.

limiting contingency is ‘‘Sinanseong 7765. [4010]V Singapyeon 7765. [1020]-1,’’ which causes voltage collapse at 780 MW of the added renewable generation in the island. The divergence occurs around the contingency buses. The corresponding PV curves are presented in Fig. 18. In addition to the above observation, other limiting factors are listed in Table 10.

D. Time-Domain/Transient-Security Analysis Transient-security studies are performed using threephase faults cleared after five cycles. System loads are represented according to the models of Table 11, as spe-

cified by the relevant Korean authority [Korean Electric Power Company (KEPCO)]. Time-domain simulations further showed a locally unstable situation under all loading conditions even before adding any new generation. That is, the clearance of either ‘‘Yeonggw NP 3345. [7152] Singimje 3 345. [6450]2’’ or ‘‘Yeonggw NP 3345. [7152] Sinnomwon 3 345. [7100]2,’’ after a three-phase fault, resulted in the rotor angle separation of ‘‘Yeonggwa 5G 22.0 [27155] 1’’ and ‘‘Yeonggwa 6G 22.0 [27156] 1’’ units. A special protection system (SPS), such as generation reduction/shedding of these units upon such contingencies, is recommended. The

Table 10 Summary of Voltage-Security Assessment

Table 11 Load Models for Dynamic Simulations

Vol. 101, No. 4, April 2013 | Proceedings of the IEEE

973

Khan et al.: Network Security Assessments for Integrating Large-Scale Tidal Current and Ocean Wave Resources

Table 12 Summary of Transient-Security Assessment

summary of transient-security assessment is given in Table 12. In the 2022 peak case, the total of 1340 MW of the new generation, which is the voltage-security limit, can be added without any transient-security violations. A snapshot of the ocean power plant terminal voltages for the 2022 peak load case under the worst contingency is shown in Fig. 19.

E. Eigenvalue/Small-Signal Stability Analysis Under small-signal stability analysis, interarea modes of the system were analyzed for single contingency screening. A relevant mode that is somewhat affected by the new generation addition is presented in Table 13 for all four cases. These results are for the worst contingency, namely, ‘‘Yeonggw NP 3345. [7152] Singimje 3 345. [6450]-2.’’ The most dominant unit of the mode is either ‘‘Sinkori 3G 24.0 [29013] 1’’ or ‘‘Sinkori 4G 24.0 [29014] 1.’’

Fig. 19. Ocean power plant terminal voltage for the 2022 peak load case under the worst contingency.

974

Proceedings of the IEEE | Vol. 101, No. 4, April 2013

Under light loading conditions of 2022, a 0.55-Hz mode was found to have lower than suggested criterion of 3% damping, namely, about 2% damping before adding the new generation, which could deteriorate to less than 1.5% after addition of the renewable resources. The above recommended SPS, as well as the new 765-kV circuit, would improve the situation significantly. Generating units with the highest participation factors were also identified for addition of new power system stabilizers (PSSs) and/or retuning of the existing PSS, if further enhancement is desired.

F. Results With regard to the addition of ocean renewables into the future Korean grid, voltage-security, transient-security, and small-signal stability analysis on Korean interconnected network for years 2017 and 2022 (under both peak and light loading conditions) have been conducted under this task. It was assumed that, before 2017, Jeju Island would be connected to the Korean mainland through two HVDC submarine transmission links, totaling a maximum capacity of 700 MW in either direction. Two locations of ocean wave energy generation in the island and four locations of tidal current flow generation in the mainland were considered. The maximum new generation injections in the island and mainland were 1000 and 620 MW, respectively, to be dispatched against forecasted load increases in certain areas of the mainland. Voltage-security assessments revealed that with all the equipment in service, the new resources could be added to the system without any voltage violation. With single contingencies, however, the security limit was about 700 MW determined by 2017 light loading condition. The limiting contingency was ‘‘Sinanseong 7765. [4010]V Singapyeon 7765. [1020]-1,’’ which could cause voltage

Khan et al.: Network Security Assessments for Integrating Large-Scale Tidal Current and Ocean Wave Resources

Table 13 Relevant Interarea Mode for the Worst Contingency Before and After Renewable Resources

collapse around the contingency buses. Branch overload violations were detected, some of which existed before new generation additions or at precontingency, although at lower percentages. The limiting contingency of the voltage-security assessments also caused transient instability under both 2017 and 2022 light loading conditions, even before adding any new generation. Implementing a second 765-kV circuit in parallel with ‘‘Sinanseong 7765. [4010]VSingapyeon 7765. [1020]-1’’ line is expected to remove the corresponding voltage and transient-security limitations. Time-domain simulations further showed a locally unstable situation under all loading conditions even before adding any new generation. That is, the clearance of either ‘‘Yeonggw NP 3345. [7152]VSingimje 3 345. [6450]-2’’ or ‘‘Yeonggw NP 3345. [7152]VSinnomwon 3 345. [7100]-2,’’ after a three-phase fault, resulted in rotor angle separation of ‘‘Yeonggwa 5G 22.0 [27155] 1’’ and ‘‘Yeonggwa 6G 22.0 [27156] 1’’ units. An SPS, such as generation reduction/ shedding of these units upon such contingencies, is recommended. The key results of this evaluation are given below. • Baseline tidal and wave power capacity: Considering simultaneous wave and tidal power generation from selected target areas along the Korean peninsula, it was observed that: / by 2017, maximum ocean power addition to the Korean network may range between 700 and 1620 MW (prescribed maximum) for light and peak loading seasons, respectively; / by 2022, this resource addition may range between 1340 and 1520 MW for peak and light loading seasons, respectively. • Limiting factors: With regard to identifying the limiting factors toward addition of ocean renewables, it was observed that: / the most significant technical issue, which is largely independent of any new generation addition, is associated with the faults at the Sinyongin–Singapoyeong 765-kV line; this contingency can be attributed to both voltage and transient instability, and a second circuit parallel to this line would remove the underlying limitations;

/

additional technical issues include: local instability at the Yonggwang nuclear plant due to nearby faults, and interarea oscillations between Daegu-Gyeong and Busan area, both of which can be effectively addressed by developing an appropriate SPS and the addition of new lines stated above. In general, the Korean electrical system is undergoing a number of timely and state-of-the-art upgrades which will allow a significant addition of various renewable resources as well as enhance the overall system reliability and security.

VI . CONCLUSION Marine energy technologies, especially tidal current and ocean wave systems, are at a nascent stage of technological development. Considering the current trends in advancements and future prospects for generating bulk renewable power, the future electrical systems need to prepare for their possible addition. This work primarily focuses on exhibiting how such novel technologies can be analyzed under conventional power system scenario analysis exercises. Two case studies discussing tidal current and ocean wave resource integration were presented and subtle observations were highlighted. It can be seen that for both cases, ocean power resources are located in areas distant from load centers, often generation-deprived areas. For the Oregon case study, the limiting factor is line/transformer overloading, whereas for the Korean case study, contingency in a major transmission line is attributed to voltage and transient stability issues, which limits the addition of ocean power to a baseline value. The models used in both studies are of generic nature, and further model development, tuning, and benchmarking need to be conducted before commissioning any detailed planning study. In general, the following observations can be made for tidal current and ocean wave systems that are being considered for other jurisdictions. • In many cases, the uptake of marine energy resources is concentrated in coastal locations which are away from load or generation centers. Both localized (subtransmission lines/equipment) and Vol. 101, No. 4, April 2013 | Proceedings of the IEEE

975

Khan et al.: Network Security Assessments for Integrating Large-Scale Tidal Current and Ocean Wave Resources

system-level bottlenecks may arise. These can be identified through powerflow analysis and shortcircuit calculations. • The dynamics associated with marine devices performance and resource variation are not adequately captured in conventional transient stability simulations, as these dynamics are exhibited in longer time scales. Also, the models of such devices need also be refined and established. At lower and midpenetration levels, the marine devices are not expected to contribute to system instability. However, for high-penetration systems, further analysis may be required to ensure availability of sufficient reserve. • Further, for high-level planning, detailed interconnection studies and impact assessments are needed for each project in order to understand the unique attributes of such systems, and to mitigate them. Any similar future work should be complemented with: 1) detailed and verified device models (dynamic characteristics); 2) analysis of longer term resource variability

REFERENCES [1] A. Lewis, S. Estefen, J. Huckerby, W. Musial, T. Pontes, and J. Torres-Martinez, ‘‘Ocean energy,’’ Cambridge Univ. Press, Cambridge, U.K., IPCC Special Rep. Renewable Energy Sources and Climate Change Mitigation, 2011. [2] M. Santos, F. Salcedo, D. Ben Haim, J. L. Mendia, P. Ricci, J. L. Villate, J. Khan, D. Leon, S. Arabi, A. Moshref, G. Bhuyan, A. Blavette, D. O’Sullivan, and R. Alcorn, ‘‘Integrating wave and tidal current power: Case studies through modelling and simulation,’’ Tech. Rep. T0331-A. [Online]. Available: http://www.ocean-energy-systems. org/library/annex_iii_reports/ [3] J. Khan, A. Moshref, and G. Bhuyan., ‘‘A generic outline for dynamic modeling of ocean wave and tidal current energy conversion systems,’’ in Proc. IEEE Power Energy Soc. General Meeting, Calgary, AB, Canada, Jul. 2009, DOI: 10.1109/PES.2009. 5275177. [4] J. Khan, D. Leon, A. Moshref, and G. Bhuyan, ‘‘A scenario analysis of northwest electrical system toward determining the level of wave power that can be integrated by 2019 in Oregon,’’ Powertech Report for PEV/OWET. [Online]. Available: http://ir.library. oregonstate.edu/xmlui/bitstream/handle/ 1957/19037/Task-4.2-Integrated-SystemsAnalysis-1.pdf?sequence=25 [5] Powertech Labs Inc., ‘‘Development of dynamic model of tidal current conversion systems,’’ Tech. Rep. 19629-21-00, 2011, prepared by Powertech Labs Inc. for Environment Canada. [6] D. Andersson, A. Petersson, E. Agneholm, and D. Karlsson, ‘‘Kriegers flak 640 MW off-shore wind power grid connectionVA real project case study,’’ IEEE Trans. Energy Conv., vol. 22, no. 1, pp. 79–85, Mar. 2007. [7] S. Grijalva, S. R. Dahman, K. J. Patten, and A. M. Visnesky, Jr., ‘‘Large-scale integration of wind generation including network temporal

976

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

(minutes to hours); 3) multimachine/plant equivalencing; and 4) lessons learned from contemporary field trials. It is important to note that the assessments presented here are of illustrative nature and further technical and economic analysis needs to be carried out prior to initializing any design or deployment initiative. Being founded on highlevel technical grounds, these studies can be treated as tools for enabling longer term marine power deployment targets and associated policy mechanisms. h

Acknowledgment The authors would like to thank the Oregon-study scenario team (BPA, PacifiCorp, Tillamook PUD, Central Lincoln PUD, PNGC Power, and other utilities) for facilitating this study. Guidance from J. Klure and T. Hampton (Pacific Energy Venture), and Oregon Wave Energy Trust (OWET), as well as technical contributions by F. Howell (Principal Engineer, Powertech Labs) are greatly appreciated. For the Korean case study, S.-H. Kim (KEPCO) and K. W. Lee (DAEHWA Power Engineering Co.) are sincerely thanked for their support.

security analysis,’’ IEEE Trans. Energy Conv., vol. 22, no. 1, pp. 181–188, Mar. 2007. C.-I. Chai, W.-J. Lee, P. Fuangfoo, M. Williams, and J. Liao, ‘‘System impact study for the interconnection of wind generation and utility system,’’ IEEE Trans. Ind. Appl., vol. 41, no. 1, pp. 163–168, Jan./Feb. 2005. L. Wang and Z.-J. Chen, ‘‘Stability analysis of a wave-energy conversion system containing a grid-connected induction generator driven by a Wells turbine,’’ IEEE Trans. Energy Conv., vol. 25, no. 2, pp. 555–563, Jun. 2010. M. Behnke, A. Ellis, Y. Kazachkov, T. McCoy, E. Muljadi, W. Price, and J. Sanchez-Gasca, ‘‘Development and validation of WECC variable speed wind turbine dynamic models for grid integration studies,’’ in Proc. AWEA WindPower Conf., Los Angeles, CA, USA, Jun. 2007, pp. 1–5. D. R. Kiran, A. Palani, S. Muthukumar, and V. Jayashankar, ‘‘Steady grid power from wave energy,’’ IEEE Trans. Energy Conv., vol. 22, no. 2, pp. 539–540, Jun. 2007. G. Hagerman, R. Bedard, and M. Previsic, ‘‘EPRI survey and characterization of potential offshore wave energy sites in Oregon,’’ Tech. Rep. E21 EPRI WP-OR-003, May 17, 2004. R. Bedard, G. Hagerman, and O. Siddiqui, ‘‘System level design, performance and costsVOregon State Offshore Wave Power Plant,’’ Tech. Rep. E21 EPRI Global WP 006 OR-Rev1, Nov. 30, 2004. M. J. Khan, G. Bhuyan, A. Moshref, and K. Morison, ‘‘An assessment of variable characteristics of the Pacific Northwest Region’s wave and tidal current power resources, their interaction with electricity demand & implications for large scale development scenarios for the region,’’ Tech. Rep. 17485-21-00 (Rep 3), Jan. 2008. C. H. Jo, K. H. Lee, and Y. H. Rho, ‘‘Recent TCP (tidal current power) projects

Proceedings of the IEEE | Vol. 101, No. 4, April 2013

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

in Korea,’’ Sci. China Ser. E: Technol. Sci., vol. 53, no. 1, pp. 57–61, 2010. K. Hong, S.-H. Shin, and S.-W. Hong, ‘‘Current status and future perspectives of marine renewable energy development in Korea,’’ presented at the PICES Annu. Meeting, Portland, OR, USA, Oct. 2010. WECC, Western Electricity Coordinating Council information summary, Jun. 2012. [Online]. Available: http://www.wecc.biz/ About/Documents/WECC_FactSheet.pdf. Technical Requirements for Interconnection to the BPA Transmission Grid, STD-N-000001-00-01, Bonneville Power Administration (BPA), Jun. 2012. [Online]. Available: http://www.transmission.bpa.gov/ business/generation_interconnection. Korea Power Exchange; Ministry of Knowledge Economy, The 4th basic plan of long-term electricity supply and demand (2008 2022), Tech. Rep., 2008. J. K. Park and Y. J. Kim, ‘‘Status and perspective of electric power industry in Korea,’’ in Proc. IEEE Power Eng. Soc. General Meeting, San Francisco, CA, USA, Jun. 2005, vol. 3, pp. 2896–2899. J. G. Vining and A. Muetze, ‘‘Economic factors and incentives for ocean wave energy conversion,’’ IEEE Trans. Ind. Appl., vol. 45, no. 2, pp. 756–763, Mar./Apr. 2010. R. Doherty, H. Outhred, and M. O’Malley, ‘‘Establishing the role that wind generation may have in future generation portfolios,’’ IEEE Trans. Power Syst., vol. 21, no. 3, pp. 1415–1422, Aug. 2006. J. Khan, C. Morton, and A. Rao, ‘‘Electrical power connectors for marine energy systems,’’ Tech. Rep. 20490-21-00 Rep-1, 2011, prepared by Powertech Labs Inc. for Natural Resources Canada (NRCan). A. Brito-Melo and J. Huckerby, Eds., ‘‘Annual report 2010 implementing agreement on ocean energy systems,’’ 2011. [Online]. Available: http://www.ocean-energy-systems. org/library/annual_reports/

Khan et al.: Network Security Assessments for Integrating Large-Scale Tidal Current and Ocean Wave Resources

[25] M. J. Khan, G. Bhuyan, M. T. Iqbal, and J. E. Quaicoe, ‘‘Hydrokinetic energy conversion systems and assessment of horizontal and vertical axis turbines for river and tidal applications: A technology status review,’’ Appl. Energy, vol. 86, no. 10, pp. 1823–1835, 2009.

[26] A. F. de O. Falca˜o, ‘‘Wave energy utilization: A review of the technologies,’’ Renew. Sustain. Energy Rev., vol. 14, no. 3, pp. 899–918, 2010. [27] Powertech Labs Inc., Enabling accelerated deployment of tidal current energy conversion systems through open-sea grid-connected

demonstration, Tech. Rep. APP REDGTF Canada 012, 2011, prepared by Powertech Labs (an Asia-Pacific Partnership (APP) on Clean Development and Climate Renewable Energy and Distributed Generation Task Force-REDGTF).

ABOUT THE AUTHORS Jahangir Khan (Member, IEEE) received the B.Sc. degree from Bangladesh University of Engineering and Technology (BUET), Dhaka, Bangladesh, in 2001, and the M.Eng. and Ph.D. degrees from Memorial University of Newfoundland, St. John’s, NL, Canada, in 2004 and 2010, respectively, all in electrical engineering. Since 2007, he has been working as a Power Systems Engineer at Powertech Labs, Inc., Surrey, BC, Canada. His technical expertise falls within various crosscutting areas of power systems and renewable technologies. This includes off-grid/microgrid system design and analysis, transmission system planning, and grid integration of large-scale alternative energy resources, particularly wave, tidal, and hydrokinetic energy. Dr. Khan is currently the Chair of the IEEE Vancouver’s Joint IAS/IES Chapter.

Daniel Leon (Member, IEEE) received the B.Sc. and M.Sc. degrees in electrical engineering and power systems from the National Polytechnic Institute, Mexico City, Mexico, in 1993 and 2001, respectively. From 1993 to 2007, he worked in the Unit of Specialized Engineering, Comision Federal de Electricidad (CFE), the government electrical utility in Mexico. From early 2008 to May 2012, he worked as a Senior Engineer in the Power System Studies group, Powertech Labs, Inc., Surrey, BC, Canada. Currently, he is at Siemens PTIVNetwork Consulting, Richmond, BC, Canada. His main interests are power system stability and dynamic security assessment.

Ali Moshref (Senior Member, IEEE) received the B.Sc. degree from Iran University of Science and Technology, Tehran, Iran, in 1977, the M.Sc. degree from The George Washington University, Washington, DC, USA, in 1979, and the Ph.D. degree from McGill University, Montreal, QC, Canada, in 1983, all in electrical engineering. He has over 30 years of experience in power system planning, system operation, asset management, alternate energy resources for power generation, and software development for the analysis of power system. He most recently was at Powertech Labs, Inc., Surrey, BC, Canada, serving as a Manager of Power System Studies. Before that, he was at CYME International, Inc., where he played a critical role in the formation of the company, as he was one of the original founders. He is now with BBA Inc., Vancouver, BC, Canada, as the Department Manager for Power System and Testing.

Saeed Arabi received the B.Sc. degree in electrical engineering from Sharif University of Technology, Tehran, Iran, in 1974 and the M.Sc. and Ph.D. degrees in electrical engineering from the University of Manitoba, Winnipeg, MB, Canada, in 1981 and 1985, respectively. He worked for both government and private companies in Iran before coming to Canada in 1979. From 1985 to 1987, he was with the Department of Electrical Engineering, Concordia University, Montreal, QC, Canada, as a Visiting Assistant Professor. In 1987, he joined the Power System Planning Division, Ontario Hydro. Since October 1993, he has been with Powertech Labs, Inc., Surrey, BC, Canada, where he is currently a Principal Engineer. His interests include power system stability and control, flexible AC transmission systems (FACTS), and high voltage direct current (HVDC) transmission, and he has authored numerous publications on the subjects.

Gouri Bhuyan received the M.Tech. degree in ocean engineering from the Indian Institute of Technology, Chennai, India, in 1982 and the Ph.D. degree in ocean engineering from Memorial University of Newfoundland, St. John’s, NL, Canada, in 1986. From late 1980s until mid 2012, he has worked with Powertech Labs, Inc., Surrey, BC, Canada. He is a former Chair of the Executive Committee of the Ocean Energy Systems (OES) Implementing AgreementVan Intergovernmental and multinational organization operates under a framework established by the International Energy Agency (IEA) and the Operating Agent for an OES collaborative Annex on integration of ocean energy plants to electrical grids. He was a Lead Author for the Intergovernmental Panel on Climate Change (IPCC) Special Report on Renewable Energy Sources and Climate Change Mitigation (SRREN). Dr. Bhuyan is a Fellow of the American Society of Mechanical Engineers (ASME) and the Canadian Academy of Engineering (CAE), and a professional engineer in British Columbia.

Vol. 101, No. 4, April 2013 | Proceedings of the IEEE

977