Contemporary Maintenance Management of Power

technics technologies education management

Contemporary Maintenance Management of Power Plant Life Exhaustion Components Savremeni koncept održavanja komponenti starih termoenergetskih postrojenja Vera Sijacki Zeravcic1, Gordana Bakic1, Milos Đukic1, Dragomir Markovic2, Bratislav Rajicic1 1 2

University of Belgrade, Faculty of Mechanical Engineering, Serbia Public Enterprise “Electric Power Industry of Serbia - EPS”, Serbia

Abstract Reliability centered maintenance (RCM) has a key role in the quality system of thermal power plants. Electricity production is definitely not only production of a product, but also becomes a customer service. Reliability is the top quality tier of this service. Incorporation in the Integrated energy system means fast transformations of organization on all levels. The reliability of modern generation components basically influences the efficiency and capacity of a power plant as a whole. This paper presents the results of RCM practice of one 620 MW thermal power plant. RCM implementation on current problems during exploitation emphasized problems with the boiler tubing system. The methodological approach included outages analyses, marking of the most significant ones, determination of main causes of damage (determination of the material exhaustion state based on expert analyses), application of adequate remaining life assessment methodology and finally the recommendations. Based on these analyses, certain technical solutions were performed and as a result, the mean time between failures of the plant increased, as well as its reliability. According to cost efficiency, RCM methodology presented the most effective cost-benefit method for reliability upgrade. 2010

1. Introduction The specific aspects and the complexity of electricity production power require a well thought out approach to enhancing the quality of the production process itself. In view of the deregulation and introduction of the energy market, the crux of quality enhancement shifts to the field of services. It is there where the fundamental parameter resides, the one which must be observed over time, which expresses the quality of the thermal power plant as a whole, the number of customer complaints. This is followed by pricing, availability, reliability, rationality, employment, and so forth. This was further promoted by demands proposed by envirnomentalists, insurance companies, responisibility issues, etc, all of which are playing ever more important roles. Such requirements have promoted preventive engineering and maintenance systems into the forefront, which because of their inherent complexity require teams of experts capable of timely reaction in crisis situations, and reliability, a category which ultimately encompasses all influences upon any plant and its realistic operation results /1/. Satisfying the market requirements, the needs of the consumers, becomes the task with the highest priority, which requires investments into making the production more reliable, while attaining greater reliability with economically optimized maintenance 431


become the challenges because they must at the same time also mean competitive price of the generated power. Maintenance of complex systems could be viewed as a very complex sequence of activities which can and must lead to the desired objective – for any plant to be in full function, performing as it should, within required limits. Although the importance of maintenance is well known, it is perhaps not understood well enough just how large are the possibilities of well designed and performed maintenance as per a clearly defined, made in advance concept with a defined objective. Proper selection of the maintenance concept is of particular importance for plants nearing the end of their nominal service life not only regarding quality of operation, but also in view of safety. The RCM (Reliability Centered Maintenance) concept has a significantly more general meaning in relation to common practice and encompasses the processes of maintenance, job execution, technical state of the equipment and analysis of plant indicators. The outcome of the RCM principle implementation includes: –– reduced plant breakdowns –– precise identification and stocking of spare parts –– more efficient planning of PM (Preventive Maintenance) measures –– reduction of corrective maintenance costs. This paper will deal with a maintenance system centerd to reliability and implemented on one of the 620 MW units of the thermal power plant, which has clocked about 160,000 operating hours, and on the effects which soon became appearent. 2. Data gathering and system selection The relevant and necessary data regarding the RCM system of observed thermal power plant, was obtained by examining and systematizing of all relevant documents, notes and reports from its exploitation history and the archives of operating orders issued for corrective maintenance. The next step was to systematize all gathered data, with special attention being paid to block outage analisys, arranged as per the following criteria: 432

–– location where the failure occurred, with as close identification of system-subsystemcomponent as possible; –– frequency of failure/outage occurrence; –– outage duration time interval; –– costs caused by individual outage. From the aspect of planning preventive measure to upkeep full functionality, it has been shown that it is best to perform an RCM analysis for the entire system. In most plants, systems are usually well defined. After concluding that the system is practically speaking the best level of compound systems to which an RCM analysis is applicable to, a question follows: “Which systems need to be focused on, and in which sequence of priority”? One of the possible approaches could be to review all plant systems. Nevertheless, it has been found that this approach is inefficient (least effective in terms of its costs), since there are systems with an insignificant number of failures, or those whose maintenance costs are so small that they do not merit being included in any cost savings scheme. With this in minds, criteria were formed to select an RCM approach yielding best overall results, such as /1/: 1. Systems with a large share in overall preventive maintenance (PM) costs, and/or overall PM costs; 2. Systems with large numbers of activities in corrective maintenance (CM) in past periods; 3. A combination of 1. and 2. above; 4. Systems with high corrective maintenance (CM) costs (the number of activities of corrective maintenance do not always correspond to cost effects); 5. Systems with a large share in overall number of complete or partial failures as per exploitation history data; 6. Systems with pronounced safety and environmental aspects. Using the above criteria, from the entire unit of thermal power plant, the boiler tubing system with hot reheated steam pipeline system was selected for RCM approach implementation. Why this system? The basic postulate of the RCM methodology is to “preserve system func2010


tioning”, while available data showed that parts of this system caused: –– the largest number of outages (boiler tubing system with a 22% share in total number of failures and a 31% share in forced outages); –– the largest maintenance/repair costs (hot reheated steam pipeline) in order to provide for system functioning and to satisfy the safety criteria. 3. Boiler tubing system description The boiler is the once through tower type, single-draught with membrane tube walls. Primary steam heating is performed in four stages. Regulating primary steam temperature uses three injections: injection 1 between superheaters 1 and 2, injection 2 between superheaters 2 and 3, and injection 3 between superheaters 3 and 4. Reheated steam is produced in three reheaters with biflux between first and second stages. Regulating reheated steam temperature uses two injections: injection 1 between biflux and reheater 2, and injection 2 between reheater 2 and 3. The thermal power plant has an average availability of 82,93%. Overhaul and reserves account for 62% and 10% respectively of the total duration of all outages, while on the equipment side, the boiler tube system accounts for 9% of all outages. This type of data is most relevant for any economic analysis of improving the plant operation. 4. Determination of critical components Figures 1 and 2 illustrate the Pareto analysis of the number and duration of the unit, respectively. As a cause of unit function losses, the boiler takes the first place both by number and duration.

Fig. 1 Share by number of outages of individual systems of unit However, it is important to note that the first three indicators (boiler, testing, coal) related to the analysis of duration are mutually related and in correlation, according to Figure 2.

Fig. 2 Share by duration of outages of individual systems in unit In view of the fact that the quality of the coal /2, 3/ used is unsatisfactory, this is primarily related to the following indicators: –– coal ® mills, burner system, air mixture ducts, …. –– testing ® constantly required control and testing to determine whether the system is functioning within its operating limits, –– boiler ® due to erosion, there is too much damage on the boiler tubing system, especially in the economizer tubes. Figure 3 shows the bath tube curve for the boiler tubing system, where the boiler system has

2010

433


been set apart by the Paretto analysis as critical for the entire installation operation for the whole unit functioning. By analyzing Figure 3, one can note infant mortality stage as well as the steade state stage. If we analyzing the diagram up to the 14-15 interoverhaul period it can be concluded that the unit began to entered to final, aging stage. However, by RCM approach applying the main cause of failure rate increasing is eliminated, which means that the failure rate narrow again to the level which corresponded to the second stage of exploitation /4/. According to the analysis, function losses of the system were mostly caused by economizer tubes cracking, despite the fact that their share in the overall number of outages has been greatly reduced since the very beginning. As previously mentioned, the cause for these outages was very intensive tubes erosion caused by poor coal quality, but also due to economizer construction, with far too small steps between tubes in view of the used coal type and quality. The next Pareto analysis was made for the boiler boiler system, Figure 4.

Fig. 3 Bath tube curve for boiler tubing system of unit /4/ Solving the problems related to economizer tubes, meaning control and reducing O&M costs and increasing installation operating time, was one of the first successes of the RCM - “shields” were installed to protect the tube bends from abrasion /5/.

434

Fig. 4 Share by number of outages of individual heating surfaces of the boiler Of the remaining heating surfaces in the boiler, by number and duration of outages, reheater 2 takes the significant second place. However, the reasons for pipes cracking in reheater 2 are different to those related to the economizer, and are caused by corrosion of tubes from the waterside. After the first more frequent cracks, an expert analysis of the cracking causes was performed /57/, which clearly showed that the cracking zone was localized to the part of the reheater made from material PN 15HM (class 15Mo3-DIN). As immediate causes of failure from the comprehensive case study analysis, the following were listed: –– inappropriate chemical composition of the tube materials; –– inappropriate mechanical characteristics of the tube materials; –– inappropriate microstructure of the tube materials, with numerous manufacturing faults and shortfalls (banded structure, numerous MnS inclusions, uneven distribution of grains by size and shape, decarburisation, segregation, uneven perlite distribution in tube wall thickness), All of which by mutual activity influence a pronounced decrease of this material’s resistance to corrosion. From the aspect of the classic approach to maintenance, the only way to prevent outages is to change parts of the reheater 2, made of this class of steel. However, the objectively hostile circumstances under which the power generation industry operated over the last decade forced finding of alternative 2010


modes to overcome problems which in their core directly implemented the RCM philosophy - use extremely short overhoul periods to enable the desired system functioning with as little cost as possible. How was the problem solved? By using the NDT device to detect the presence, magnitude and degree by corrosion caused damages on the waterside of tubes. What did this device make possible? To make the number of outages caused by tube cracking of the reheater 2 truly minimal, to have the time set aside for planning of purchases and preparing the tubes for exchanges, and to reduce the costs of corrective maintenance. How? During each overhaul, tubes were examined by this device to determine the corrosion attack level related to metal of reheater 2 tubes, thus indicating which tubes should be changed at once, and which according to their degree of corrosion can and should be exchanged in the next service round. Why? This device allows for a very short time interval, requiring no special preparations or prolonging of overhaul, to examine an extremely large number of tubes and to obtain excellent insight into the corrosion state of metals along the entire heating path. By using the above method, the number of system outages cause end by sudden tubes cracking in the reheater 2 has, over the last period, been reduced to zero /8/. The next problem is related to the hot reheat steam pipeline, which in the overall outages of unit participates with only 1%, and that only till 1991. This pipeline system was constructed using 0.5Cr0.5Mo0.25V, and has four main branches with auxiliary lines (dimensions: Æ558x32, Æ813x45, Æ457x30, Æ1016x56, Æ660x38). At the time the thermal power plant was being build up, world manufacturers of associated equipment were aware that this material did not have such characteristics which would completely qualify it as suitable for making high temperature loaded components due to its unsatisfactory creep characteristics and poor welding properties when welding higher thickness and diameters of pipes. Based on the engineering maintenance report of 1991, an entire group of problems was noted which were related to hot reheated steam pipe2010

lines, the solution to which required extensive works and unplanned expenditures for repairs, as well as prolonged overhaul works. Problems appeared at the first mostly with factory welded joints (first cracks appeared after only two years of operation), and were related to: –– appearance of cracks on factory, but also on mounting welded joints; –– appearance of cracks after welding or repairing of joints; –– undefined prestraining during welding; –– undefined technologies of welding, repair and thermal treatment; –– imprecise defining of repair and thermal treatment criteria; –– undefined criteria of operation monitoring, regarding the intervals of examining and anticipating certain repair works. By 1991, some welded joints had been repaired using classic methods two times already, which is the maximum allowable number of repairs on any one and the same welded joint. What was to be done? One had to find an alternative solution which would surely deviate from standard recommendations of equipment manufacturers, but would not only be usable, but would also have to provide for a selection of PM measures to enable reliable and anticipated system functioning. In the aim of problem solving the austenite electrodes for repairing of welded joints was used. The process itself is significantly simplified in comparison with classic methods - no preheating or annealing are required, there is no heat affected zone on the base material side to speak of, the fusion zone is narrow and has good strength characteristic, while the procedure itself can be used on the same place more than twice. Beside the shorter time interval required for repairs, the procedure itself is significantly cheaper than the alternatives, as shown by the following simplified analysis: –– in case of the pipeline leaking at the welded joint, e.g. for medium caliber size pipe, the classic method requires 5 days of overhaul time. If one accepts the minimum purchase price of power of $ 0.03/kWh which must replace the halted local production, one quickly arrives at the sum of $ 1,95 million for five days for a 620 MW plant; 435


–– for the same joint, when using austenite electrodes for repairing welds one needs no more than 2 days, or $ 0,78 million. If this analysis is now enlarged to a overhaul period of 45 days, which assumes examining 50% of all welded joints (and there is a total of 284 such joints), then the examining itself lasts about 25 days, with the provision that after 12 days one can initiate preparations of some welded joints. If one has planned for parallel repair of two welded joints, then this means that in 33 days one has available, using classic methods one can repair 13 joints. However, if the found number of welded joints needing repair is greater than 13, the overhaul time interval will have to be prolonged. Every two additional welded joints needing repair work cause additional 5 days of overhaul time to be required, and therefore additional expenditures of $ 1.95 million. When one uses austenite electrodes, the planned repair time can support repairs of significant number of welded joints. Also, as already noted with plants operating in their base regime, use of austenite electrodes will cause savings which can be taken as direct savings, since the lacking amount of electrical power can be replaced by simple purchasing alone. The most important aspect of RCM approach in this case is preventive preserving of system functioning. Because we know that welded joint of hot reheated pipeline present the potential weak point of the unit, by applying of planned NDT testing and timely weld repairing pipeline outages are completely avoided. 6. Conclusion In the base of RCM approach always lies the cost-effective preventive measure which provide reliable system functioning. In the case of presented critical boiler system, the resolved factor for RCM methodology applying was large number of outages (boiler tubing system), potential large number of outages (hot reheated steam pipeline welded joints), generaly large cost of repairing/ maintenance and finaly safety factor. Specific solution of preventive maintenance applyed on power plant present cumulative and sucsessive effect of RCM and experts approaches. 436

References 1. A. M. Smith, Reliability-Centered Maintenance, McGraw-Hill.Inc., New York, 1993 2. D. Markovic, Coal Quality Influence on Availability and Reliability of Two 620 MW Lignite Fired Units After 100.000 Operating Hours, VDI, Braunschweig, FRG, 1997, pp 247-260 3. D. Markovic, Tehnika, Vol. LV, N0 4/5, 2000, pp 22-26 4. G. Bakic, V. Sijacki Zeravcic, D. Milanovic, M. Djukic, Power Plant Reliability Enhance on the Bases of Expert Analysis of Outage Statistic, Proc. of the 4th DQM Conf. – Maintenance Management 2001, Vrnjacka Banja, Serbia, 2001, pp 381387 5. Internal Report 9/90, TENT-B Power Plant, Obrenovac, Serbia, 1990 6. Report 12-04-12.04/1995, Fac. of Mech. Eng., Belgrade, Serbia, 1995 7. Report 12-11-12.04/1998, Fac. of Mech. Eng, Belgrade, Serbia, 1998 8. V. Sijacki Zeravcic, G. Bakic, M. Djukic, B. Andjelic, D. Milanovic, Experience in NDT Unit EMF3-MI Application on Corrosion Damages Detection of Boiler Piping System, Proc. of the 3th Conf. – Enegetika Srpske 2001, Teslic, Republic of Serpska, 2001, pp 156-161

Corresponding author: Vera Sijacki Zeravcic University of Belgrade, Faculty of Mechanical Engineering, Serbia E-mail: [email protected]

2010