The green blockchain - IEEE Xplore

14 downloads 0 Views 332KB Size Report
The green blockchain. Managing decentralized energy production and consumption. F. Imbault, M. Swiatek. Evolution Energie. 8 Passage Brulon Paris, France.
The green blockchain Managing decentralized energy production and consumption

F. Imbault, M. Swiatek Evolution Energie 8 Passage Brulon Paris, France [email protected]

R. de Beaufort GE Grid Solutions Paris, France [email protected]

Abstract — Energy systems are evolving towards a more decentralized model accommodate with heterogeneous but competitive energy sources and energy storage systems (ESS). This will enable peer to peer energy transactions through microgrids architectures. This paper explores the use of blockchain technology implemented on an Industrial operating system (Predix) for a use case of green certificates, demonstrated within an eco-district. Keywords—decentralised energy systems ; microgrid; green certificates; blockchain

I. INTRODUCTION The environmental impact of energy generation has been extensively scrutinized to identify new technologies that rely more on renewable sources and less on fossil fuel resources. Integrating intermittent power generated from wind or solar plants into the power distribution network is a challenge. The decentralized nature of those physical assets leads to rethinking our architecture of energy management. One option under discussion to accommodate with this holistic approach of future energy systems that could be envisioned as an “honey comb” of microgrids all connected. These new architectures define energy sources and ESS as “Agents” or “Assets” being consumer or prosumer. The security, transparency and traceability of the transactions will necessitate new approaches that will be discussed in this paper. It is understood that blockchain technology will be a promising approach to ensure traceable and secure transactions. However, blockchain necessitates elastic computing resources, big data management and a professional environment to develop the applications under consideration (preventive maintenance, Energy Management Optimisation, Demand Response applications). In this paper, we propose to explore the implementation of a blockchain technology into an industrial operating system (Predix) to develop new applications for optimizing the future distributed energy systems. II. STATE OF THE ART A. Microgrids A microgrid is a local energy system consisting of distributed energy sources and loads capable of operating in

R. Plana GE Digital Paris, France [email protected]

parallel with, or independently from, the main power grid. A compromise with existing power sources made available by distribution system operators (DSO) is critical to maintain the stability of the power system [1]. The cost of electricity generated by photovoltaic (PV) systems has become “competitive for peak power production, for generation in grid-constrained areas, and for off-grid applications” [2]. Major improvements have been achieved in the recent years in energy storage capabilities [3] and a “range of mechanisms have been developed to encourage the adoption of microgeneration technologies” [4]. A new business model is emerging for local energy markets [5], with a triptych PV – Storage - Microgeneration as in Fig. 1.

Fig. 1. Microgrid

In this figure, it is seen that connectivity will be crucial to ensure fluidity and efficiency as well as traceability of the energy sources and energy transactions. In this paper, we propose to articulate the use case around four main items essentials for energy management systems (Energy markets, CO2 emissions, Green certificates and maintenance). B. Registries for energy management The registries for energy management may be used in the following cases: 1) Energy markets

This work is supported by the BPIFrance Aixpé Program and the DigitalIndustry Open Innovation Program powered by GE Digital and NUMA.

978-1-5386-3917-7/17/$31.00 ©2017 IEEE

The energy markets are interested in the registries for settlement and invoicing purposes. 2) Emissions CO2 emissions can also be managed with registries, but the security problem also has to be taken into account, as demonstrated by VAT frauds or quota theft in the past. 3) Green certificates A green certificate is a new kind of commodity product that serves as a proof that the electricity has been generated from renewable energy sources, such as solar or wind. Following the EU guideline 2009/28/EC, certificates are stored in national registries for Guarantees of Origin (GoO) that provide all information related to the production facility [6]. Standardized trading of certificates supports the market integration of renewable energies, increases the transparency and liquidity of European energy markets and creates new opportunities for plant operators [7]. Yet those certificates do not cover the new requirements of microgrids, including decentralized assets of smaller volumes (below 1MWh) dispatched in many places (not aggregated at national level) and the need for higher frequency monitoring for operational purposes, beyond a yearly statement. Aside those technical limitations, remains the underlying issue that operators can trust the full chain of certificates, in particular that the same volumes are taken into account only once. 4) Maintenance There is a need for a particular kind of registry that monitors and archives maintenance actions in order to perform predictive maintenance and be able to calculate health index of equipment and criticality of the maintenance operations. This would be helpful to check who has produced the energy gains and enable a transparent remuneration of energy efficiency services. III. BLOCKCHAIN TECHNOLOGY A. A distributed approach to data management Blockchain is a new technology to implement distributed registries. The first blockchain was conceptualized in 2008 as a core component of the “bitcoin” cryptocurrency where it serves as the public ledger for all transactions [8], including smart contracts which are digitally signed agreements among parties [9].

The current article explains how this approach can be extended to manage environmental certificates, emission allowances and other energy efficiency schemes to develop the concept of green blockchains. A focus should also be paid to the computational cost of the blockchain scheme itself, to make it as energy efficient as possible. B. The need for a multi-asset blockchain The blockchain is designed as a process of adding blocks to a chain of cryptographically signed data to form perpetual and immutable records. A block consists of a pointer to the previous block, some consensus information data, some transaction or state, a signature and some other useful information (version, timestamp, etc…). A blockchain scheme implements a peer to peer network protocol, so that each node can download one or multiple chains. Each node verifies the transaction or state by performing cryptographic calculation on the data in the record and notifies the other nodes each time a new block of transaction is verified as legitimate. Different consensus protocols may solve the problem of which canonical chain is trusted and shared by the majority. Despite the distributed nature of the technology, many realworld scenarios still require a rich permissioning to ensure a clear separation of duties and enforce Chinese walls, implemented as private chains. The permissioning role can be managed for instance by a consortia, a central authority, an energy supplier or a demand-response aggregators, who also bring other energy related services to the users. Some applications for energy management, such as demand-response for primary reserves, also require a low latency decision making, even for geographically distributed clusters. A private chain is more scalable as the trade-off between consensus and availability can be fine-tuned. Yet, those private chains present two related issues: • Security depends on the size of the network, as well as on the technical ability of private chain developers. • The assets become silo-ed into many different chains. Those issues can be resolved via communication between ledgers, such as Interledger protocol [13] and a pegged token to a commonly used blockchain [14]. The chosen architecture follows:

The coordination issues and the limited expressiveness of Bitcoin’s transaction language has pushed smart contracts onto altchains [10] to enable further innovation. Blockchains are not restricted to financial transactions, they can be used for any kind of digital asset. A few use cases for energy management have been implemented as proof-of-concepts, by startups and large energy companies. The goal is to establish a fully decentralized energy system for transaction management, such as buy and sell contracts between consumers and producers [11] [12].

Fig. 2. Blockchain architecture

To validate the user requirements as well as the technological framework, a real case has been implemented within the eco-district of Rueil-Malmaison in France.

been arbitrarily decided that a production certificate would be attributed for a production volume of 5 kWh, and that – for incentivizing purposes – the same certificate would cover a consumption of 10 kWh.

A. The use case The demonstration site is the Jean Jaurès elementary school with a wood-fired boiler that is used as a heat generator.

1) User interfaces Three user interfaces have been developed and pushed on the Predix platform. They correspond to three different cases of user profile, with different vision:

IV. THE ECO-DISTRICT CASE

Beyond the monitoring of the local production and consumption, users and maintenance operators would be able to follow-up their improvement actions to improve energy efficiency. A particular challenge is to make the application easy enough to understand for the general public and even for children, since the goal is to spread the application to any kind of local production facility. For instance, we translate the crypto-conditions into a human readable information for motivational purposes. Virtual nodes are added to the framework (taking form of two other buildings of the eco-district) in order to simulate the behavior of additional agents and to assess the scaling potential for an eco-district. The green certificates can also be allocated to local producers. These certificates can be bought by energy consumers, and the transaction is certified by the blockchain controller node, which also plays a role of central memory of certificate transactions.



A building user vision, which can be for instance a child of the school used for the experimentation case, or the user of the town library included in the ecodistrict. This UI shows very simple information about the building real time energy consumption, for the user to see and understand at first glance the level of green or locally produced energy consumption for the building.



A building manager or technical administrator vision allowing the user to access a more detailed view of the green certificates obtained by the building production, as well as the overall energy consumption of the building for monitoring purposes.



An eco-district manager vision, allowing the access to an overview and detailed information about the ecodistrict buildings consumption and production, as well as the certificates obtained, bought or sold. The dashboard of the developed UI for this user story is shown in Figure 3.

B. Implementation within the Predix framework The implementation has been carried out and integrated within the Predix framework [15]: •

A private chain with extendible physical asset definition.



A scalable framework to enable edge computing on the measurement devices and equipment.



Three user-friendly and intuitive interfaces to monitor the energy and environmental impact and achieve user involvement according to the user profile.

C. Results Results have been achieved throughout several aspects: user interfaces have been developed and pushed on the Predix platform, giving widespread possibility of access; Corda has been tested for the required application and has showed its limits for the use cases envisioned; and several market design cases have been defined. In the scope of this Proof of Concept (PoC), several functional hypotheses have been stated. In terms of frequency for attribution and collection of certificates, the server does these operations on a daily basis. It also balances the certificates to cover all the buildings’ consumption with production certificates. It means that if there is a surplus of certificate due to a greater amount of production and / or a lesser amount of consumption, then certificates are sold; and if there is a lack of certificates to cover the consumption, then they are bought. Concerning the covering thresholds, it has

Fig. 3. User interface pushed on Predix designed for an eco-district manager

It can be envisioned that such user interfaces, or others, could be developed for other kinds of users for other applications, depending on the type of network and the topology of building and population accessing the UI. 2) Corda testing At first, this green blockchain project was supposed to be implemented in partnership with XNotes, but due to circumstances, they withdrawn their contribution. Another technology solution to meet their specifications had to be found, and the closer seemed to be Corda. In the scope of this PoC, we used the application demo provided with Corda up to the version 8.2. We modified the app to adapt the transactions to our needs for the management of Green Certificates. Using Corda to store energy consumption and production has also been tested, but not retained for the PoC as will be discussed below. We deployed nodes corresponding to the 3 sites used in the PoC that

represent the buildings that consume and/or produce energy (the Jean Jaurès elementary school test site and two simulated sites). We used another node as notary service to validate the transactions sent on Corda. This node might represent the micro grid storage facility (controller node) and / or the gateway to the external power grid, depending on the scenario. While running our corda app, we encountered multiple limitations. The first one was that the current development state of Corda restricted deployment of nodes to be on the same subnetwork. Whereas this issue is not limiting for the scope of the PoC, it is a major lock for any production app consideration. The implementation was secondly limited by the use of h2 databases in Corda. The latter is not a major issue but more commonly used database frameworks might help with performances and interoperability. Nonetheless this choice can be understood for node portability. Third, the main issue encountered was on the memory usage and stability. A node from the modified Corda application demo required an important amount of memory from the start. Furthermore, the memory usage tends to increase progressively when the node is running. In some cases, we reached out of memory exceptions that brought the node down. More aggressive garbage collection settings made the things better but more tests would be required to investigate and fix the real issue. The last limitation we encountered consists in the instability of the nodes to large numbers of transactions. In particular, the problem occurred when we tried to store not only the certificates but also the consumption and production of energy on the node. The increased frequency of transactions and their fast-growing number lead to time out of the node to process the transactions or the monitoring queries. For these reasons, the authors envision the development of their own blockchain system adapted to the case of green certificate management and energy monitoring in the case of a decentralized production and management. 3) Market design cases In the scope of this project, the possible evolutions of the energy market were to be assessed and several market design cases should have been studied to try to see the results of various scenarios. However, due to a lack of time these results are too incomplete to be displayed for the time being. Indeed, this work was meant to evaluate the test hypothesis of the threshold for the attribution and collection of certificates, which was stated for the PoC in terms of data flux for technical reasons, rather than values corresponding to a viable market use. For this purpose, the approach envisioned would have been to try and reconcile a reasonable number of certificates for a production volume made by renewable energy systems at the scale of a building in an eco-district with a viable cost in term of management and value for the certificate “market place” and authentication / validation of all certificates for the organization responsible for these tasks. In addition, the models were also supposed to simulate changes in the regulation for taking into account the taxes relative to the fact that such eco districts would run on a micro-grid system with partial or no supply from the main and centralized power production.

V. CONCLUSION The blockchain is a promising technology for a trusted measurement and monitoring of energy related assets, which are expected to be more decentralized as self-generation and micro grid business scenarii spread, both in the public and private sectors. Much research remains to be done both on the theoretical foundations of the blockchain and on the relevant business applications for energy management. More experimental settings are needed to fully envision how the direct use of realtime measurement data from “Internet of Things (IOT)” sensors or from other embedded instrumentation can be achieved. Besides, other hypotheses could be tested or envisioned to ensure a flexible use of blockchain certificates, as for instance the way the certificates are used and the threshold at which they are attributed. For example, instead of an incentive method consisting of covering more consumption than the value of the production to get the certificate, the alternative would be that a certificate covers the same amount in consumption and production, but each building would have a “coverage” contract up to a given percentage. Such scenarios could be studied on other use cases and challenged to see which ones would better correspond to the use of building managers, city councilors, or grid operators. ACKNOWLEDGMENT The authors are grateful to the city of Rueil-Malmaison for letting us use this building as a case study and for allowing us to share data and information on this project. REFERENCES [1]

M. Parol, L. Rokicki, “Voltage stability in low voltage microgrids in aspects of active and reactive power demand”, Archives of Electrical Engineering, vol. 65, 2016 [2] Irena secretariat, “Renewable energy technologies: cost analysis series”, Irena working paper, vol. 1(4), 2012 [3] N. DiOria, A. Dobos, and S. Janzou, “Economic Analysis Case Studies of Battery Energy Storage with SAM”, NREL report, 2015 [4] A. Hawkes, E. Entchev, and P. Tzscheutschler, “A Comparative Review of Microgeneration Policy Instruments in OECD Countries”, IEA EBC Annex 54, 2014 [5] J.F. Sanz, G. Matute, H. Bludszuweit, E. Laporta, “Microgrids, a new business model for the energy market”, International Conference on Renewable Energies and Power Quality, 2014 [6] EEX, Garantees of Origin, Available online at https://www.eex.com/en/goo [7] J. Jansen, E. Drabik, and C. Egenhofer, “The Disclosure of Guarantees of Origin: Interactions with the 2030 Climate and Energy Framework”, CEPS Special Report, N° 149, 2016 [8] D. Awasthi, “Barter to bitcoin: the changing visage of transactions”, Elk Asia Pacific Journal of Finance and Risk Management, vol. 6, issue 4, 2015 [9] PwC, “Blockchain and smart contract automation: How smart contracts automate digital business”, 2016, Available online at https://www.pwc.com/us/en/technology-forecast/2016/blockchain/pwcsmart-contract-automation-digital-business.pdf [10] L.M. Goodman, “Tezos: A Self-Amending Crypto-Ledger Position Paper”, 2014, Available online at https://www.tezos.com/pdf/position_paper.pdf

[11] M. Merz, “Einsatzpotenziale der Blockchain im Energiehandel”, published in Daniel Burgwinkel et al.: “Blockchain Technology Einführung für Business- und IT Manager”, de Gruyter, 2016 [12] PwC, “Blockchain – an opportunity for energy producers and consumers?”, 2016, Available online at https://www.pwc.fr/fr/assets/files/pdf/2016/12/blockchain_opportunity_f or_energy_producers_and_consumers.pdf

[13] Interledger, Available online at https://interledger.org/interledger.pdf [14] Cosmos, Available online at https://cosmos.network/whitepaper/en-US [15] Predix, Available online at https://www.predix.io/