PaperlayerView exact revision on arXiv

Source-linked AI summary

Applications of blockchain and artificial intelligence technologies for enabling prosumers in smart grids: A review

Weiqi Hua, Ying Chen, Meysam Qadrdan, Jing Jiang, Hongjian Sun, Jianzhong Wu

arXiv:2202.10098v1cs.ITcs.AIcs.LGmath.IT

TL;DR

Prosumers are increasingly active in smart grids, but existing market structures and power-system resources create barriers to their integration. This review examines policy, blockchain-supported markets, and AI for power-system operations, and identifies a lack of dynamic and decentralized policy measures as a vital barrier.

  • Problem

    Existing energy markets do not accommodate prosumer exchanges well, and small or medium prosumers face limited control budgets for optimal scheduling.

    Source: §1 · p. 2

  • Method

    The paper reviews regulation, energy markets, and power-system operations, focusing on blockchain for decentralized trading and AI for operational decision-making.

    Source: §1 · p. 3

  • Results

    The review identifies the lack of dynamic and decentralized policy measures as a vital barrier to prosumer engagement.

    Source: §5 · p. 16

  • Takeaways & Limitations

    The review calls for future research and practical regulatory design to identify key responsibilities, assets, and roles.

    Source: §5 · p. 16

    Blockchain faces physical-network constraints on trading distance, lower throughput, higher latency, and transaction costs that can hinder prosumer participation.

    Source: §3.2.1 · p. 9, §3.3 · p. 11

Abstract

from arXiv · show

Governments' net zero emission target aims at increasing the share of renewable energy sources as well as influencing the behaviours of consumers to support the cost-effective balancing of energy supply and demand. These will be achieved by the advanced information and control infrastructures of smart grids which allow the interoperability among various stakeholders. Under this circumstance, increasing number of consumers produce, store, and consume energy, giving them a new role of prosumers. The integration of prosumers and accommodation of incurred bidirectional flows of energy and information rely on two key factors: flexible structures of energy markets and intelligent operations of power systems. The blockchain and artificial intelligence (AI) are innovative technologies to fulfil these two factors, by which the blockchain provides decentralised trading platforms for energy markets and the AI supports the optimal operational control of power systems. This paper attempts to address how to incorporate the blockchain and AI in the smart grids for facilitating prosumers to participate in energy markets. To achieve this objective, first, this paper reviews how policy designs price carbon emissions caused by the fossil-fuel based generation so as to facilitate the integration of prosumers with renewable energy sources. Second, the potential structures of energy markets with the support of the blockchain technologies are discussed. Last, how to apply the AI for enhancing the state monitoring and decision making during the operations of power systems is introduced.

1. Introduction

Smart grids support prosumers who produce, store, and consume energy, but current markets, information infrastructures, and control resources pose challenges to their participation. The review examines blockchain for decentralized energy markets and AI for power-system operations.

1. Introduction

  • Prosumers produce, consume, and store energy, and are expected to support demand-side management and reduce dependence on fossil-fuel generation.They may exchange energy with utilities or other prosumers to meet demand or make profits from energy arbitrage.Source: §1 · p. 2
  • Current markets and power systems struggle to accommodate prosumers because pricing and balancing omit their exchanges, information flows are growing, and smaller users have limited control-system budgets and uncertain prosumption.The review identifies challenges in both market structures and power-system operations.Source: §1 · p. 2
  • Blockchain offers decentralized energy-market trading platforms, while AI supports operational control and decision-making in power systems.The review frames flexible markets and intelligent operations as key factors for integrating prosumers.Source: §1 · p. 2
  • The review covers carbon-emissions policy, blockchain-supported energy-market structures, and AI applications for power-system monitoring and decision-making.Its motivating question is how blockchain and AI applications in smart grids could enable prosumer integration to decarbonize power systems.Source: §1 · p. 3

2. Carbon emissions tracing and pricing

This section reviews carbon-emissions tracing as a foundation for policy design, then considers regulations and research on pricing emissions from fossil-fuel generation.

2. Carbon emissions tracing and pricing

  • The section reviews carbon-emissions tracing to inform policy design and discusses how regulations and research price emissions from fossil-fuel generation.These are the section's two stated components.Source: §2 · p. 3

2.1. Carbon emissions tracing in power systems

The section describes carbon intensities and carbon emission flows as complementary ways to account for emissions in power systems. Carbon intensities measure emissions per unit of generation, while emission flows trace emissions through transmission and consumption.

2.1. Carbon emissions tracing in power systems

  • Carbon intensities quantify emissions per unit of energy generation, whereas carbon emission flows (CEFs) trace emissions from generation through transmission and consumption.The review uses carbon emissions to mean carbon dioxide equivalent, converted from greenhouse-gas amounts according to their global-warming potentials.Source: §2.1 · p. 3

2.1.1. Carbon intensities

  • Research on carbon intensities distinguishes average, marginal displacement, and marginal emission factors to assess part-load effects on annual, marginal-generator, and marginal-demand emissions.Long-term average carbon intensities from life-cycle analysis are also used to audit emissions associated with renewable energy sources.Source: §2.1.1 · p. 3

2.1.2. Carbon emission flows

  • CEFs are virtual flows concurrent with power flows, used to allocate emissions responsibility across network components; studies apply them to consumption-related and power-delivery emissions.The reviewed work includes analysis of regional variation in indirect emissions and emissions from power delivery.Source: §2.1.2 · pp. 3–4

2.2. Policy design for pricing carbon emissions

The review describes carbon taxes and emissions trading as market-based ways to price emissions, with carbon costs linking generation to energy markets and consumers. It also compares their design trade-offs and discusses how carbon prices and scenarios relate to decarbonization targets.

2.2. Policy design for pricing carbon emissions

  • Carbon pricing seeks to make fossil-fuel emitters pay for environmental damage; by the end of 2019, 46 countries had implemented it, including 25 with carbon taxes and 21 with emissions trading schemes.The review presents carbon pricing as a way to increase fossil-fuel costs and encourage prosumers using distributed renewables.Source: §2.2 · p. 4

2.2.1. Carbon tax

  • A carbon tax levies a fixed rate on fossil-fuel carbon content, with the rate determined by the social cost of carbon.The social cost of carbon quantifies the marginal damage costs of emissions to society.Source: §2.2.1 · p. 4

2.2.2. Emissions trading scheme

  • Emissions trading allocates carbon allowances for a period; emitters must cover their emissions, and surplus or deficient allowances can be traded.The scheme links to energy markets through generators’ carbon emissions and incurred carbon costs; generators can receive, auction, or trade allowances and report emissions coverage.Source: §2.2.2 · p. 4
  • An inappropriate emissions-trading price may weaken emissions-reduction incentives or cause carbon leakage when producers shift emissions to a lower-price region.Price floors and ceilings are used to constrain carbon prices; the UK price floor followed a period when a lower EU price had failed to incentivize the coal-to-gas transition.Source: §2.2.2 · p. 4
  • A generator’s carbon cost depends on its carbon price, generation carbon intensity, and output; fossil-fuel technologies have higher life-cycle carbon intensities than renewables, and costs pass to consumers through electricity bills.The review describes coupling energy and carbon markets and discusses peer-to-peer trading of energy and carbon allowances.Source: §2.2.2 · pp. 5–6
  • The four UK scenarios project different carbon-cost paths: Leading the Way is the most progressive and yields the highest cost, while steady progression would fail to meet the 2050 net-zero target.System transformation and consumer transformation have the same carbon-cost trend and are projected to achieve the target.Source: §2.2.2 · pp. 5–6, Figure 3 · p. 6

2.2.3. Comparison remark

  • Both policies price emissions and can raise public revenue, but a carbon tax fixes the price while emissions trading fixes the allowance quantity and allows greater flexibility for financial innovations.The review notes that tax relies on established tax systems, whereas emissions trading can embed peer-to-peer trading and options.Source: §2.2.3 · p. 6

2.3. Remark of research challenges

Existing carbon-pricing policies leave a gap in targeting emissions at the level of real-time power profiles. As prosumer energy patterns shape emissions, tracing them remains challenging, particularly for assigning personal carbon allowances.

2.3. Remark of research challenges

  • Long-term policies for overall power systems cannot specifically target real-time power profiles and their incurred carbon emissions.The paper identifies this as an opportunity for dynamic, decentralised policy measures aimed at high-carbon generators and consumers.Source: §2.3 · p. 7
  • Tracing emissions from individual prosumer energy patterns is challenging as those patterns reflect responses to pricing incentives.Patterns include decisions about on-site generation, consumption, and energy exchange.Source: §2.3 · p. 7
  • Prosumer-centric emissions tracing is particularly important when assigning personal carbon allowances to individual prosumers.Source: §2.3 · p. 7

3. Energy markets transition with prosumers’ integration

The section reviews decentralised market structures that accommodate prosumers and blockchain approaches to energy trading. It compares peer-to-peer, intermediary-based, and microgrid-based arrangements, then surveys blockchain applications and contrasts decentralised with conventional trading.

3.1. Potential structures of decentralised energy markets

  • The three market structures are peer-to-peer, intermediary-based, and microgrid-based, distinguished by control-unit functions and information exchange.The comparison is presented in Table 2.Source: §3.1 · p. 7, Table 2 · p. 7

3.1.1. Peer-to-peer trading markets

  • Peer-to-peer markets let prosumers trade energy and services directly, without a central control authority, increasing their control and information-flow burdens.The distribution system operator manages the trading platform and power distribution.Source: §3.1.1 · p. 7

3.1.2. Intermediary-based trading markets

  • Intermediary-based markets pool a community’s generation, loads, and storage for collective coordination by an aggregator or other intermediary.The intermediary can provide services such as demand-side management and earn bonuses from regulators or utilities.Source: §3.1.2 · p. 8

3.1.3. Microgrid-based trading markets

  • Microgrid-based markets connect prosumers to microgrids, which can connect to utility grids or operate islanded, shaping whether surplus is sold, stored, or used for load shifting.Utility-grid connection can reward surplus generation, while islanded operation incentivises scheduling for local energy balance.Source: §3.1.3 · p. 8

3.2.1. Blockchain Technologies

  • Blockchain can provide an open, decentralised trading ledger, protect private information, and secure transactions through ownership accounting and network consensus.Its application also faces physical-grid constraints and lower throughput and higher latency than existing trading technologies; the cited examples report Ethereum at 15 transactions per second and 3 minutes, Bitcoin at 7 and 10 minutes, and Visa at 2000 and 0.05 minutes.Source: §3.2.1 · p. 9

3.2.2. Smart Contracts

  • Smart contracts automate standardised, self-enforcing settlement: monitored energy or services can trigger payment from buyer to seller.Trust depends on reliable smart meters and executable programs, while their interaction with meters or controllers requires new communication protocols and interfaces.Source: §3.2.2 · p. 9

3.2.3. Research and implementations

  • Reviewed implementations apply blockchain and smart contracts to energy exchange, demand response, voltage regulation, and carbon-allowance or credit trading.Examples include virtual-currency incentives for demand response and energy balance, and blockchain carbon trading and accounting systems [81].Source: §3.2.3 · pp. 9–10

3.2.4. Comparison between the centralised trading and blockchain based decentralised trading

  • Centralised trading relies mainly on large fossil-fuel generators, market-wide pricing, negotiated contracts, legal enforcement, and third-party trust; decentralised trading shifts toward prosumers, individual supply-demand pricing, standardised smart contracts, and blockchain consensus.Smart-contract settlement can check meter data to prevent contract violations before delivery, while trust also depends on the interface with smart meters.Source: §3.2.4 · p. 10

3.3. Remark of research challenges

The transition to decentralised energy markets presents operational, system-governance, and cost challenges. These include managing grid effects of distributed generation, aligning prosumer incentives with system-wide benefits, and reducing blockchain transaction costs.

3.3. Remark of research challenges

  • Feeding distributed generation into utility grids can challenge current market and grid operations through negative prices, voltage spikes, power imbalance, and harmonic distortion.Source: §3.3 · p. 11
  • Decentralised markets without central authorities need rules, incentives, and pricing schemes to align individual prosumer behaviour with system-wide resilience and carbon-mitigation benefits.Source: §3.3 · p. 11
  • Current blockchain transaction costs exceed those of conventional IT trading systems and can deter prosumer participation in peer-to-peer energy trading.Fig. 5 presents average transaction fees for Bitcoin and Ethereum [89].Source: §3.3 · p. 11, Figure 5 · p. 11

4. Artificial intelligence supporting operations of power systems

The section reviews AI approaches for using smart-grid metering data to improve power-system situational awareness and operability. It highlights the relevance of AI for prosumers with independent decisions and limited control budgets.

4. Artificial intelligence supporting operations of power systems

  • Smart-grid metering data could be used by AI to improve power-system situational awareness and operability.Source: §4 · p. 11
  • AI use is especially relevant when small or medium-sized prosumers make independent operational decisions with limited control-system budgets.Source: §4 · p. 11
  • Household-level control solutions cost on the order of 57 USD per year, and the section reviews state-of-the-art AI approaches for power-system operations.Source: §4 · p. 11

4.1. Analysing and optimising operations of power systems

This section reviews game-theoretic and optimisation approaches for analysing power-system operations, alongside statistical methods for modelling uncertainty from renewable generation and flexible demand.

4.1.1. Game theory

  • Cournot models simultaneous competition over quantities, whereas Stackelberg models sequential decisions by leaders and followers.Stackelberg interactions may extend across multiple decision levels.Source: §4.1.1 · pp. 11–12

4.1.2. Optimisation approaches

  • Optimisation includes programming techniques and heuristics; heuristics search feasible spaces when nonlinear problems are difficult to solve analytically or optimize globally.The reviewed heuristics include particle swarm, genetic, and artificial immune algorithms.Source: §4.1.2 · p. 12

4.1.3. Research and implementations

  • Game-theoretic applications address pricing, decarbonisation, prosumer exchange, and regional balancing, while identifying unresolved carbon-revenue allocation and network constraints.Reported extensions include imperfect information, RES uncertainty, and power losses.Source: §4.1.3 · pp. 12–13

4.1.4. Analysis of uncertainties in power systems

  • Monte Carlo, Latin hypercube, and stochastic-interval approaches model operational uncertainty, with reported benefits including computational efficiency, accurate estimates, and industrial applicability.The reviewed studies also identify opportunities to improve efficiency, include risk criteria, and determine interval probability degrees.Source: §4.1.4 · p. 13
  • Scenario methods still face limits: mixed electric-vehicle energy-pattern probabilities need further study, and typical-scenario selection may not scale across power networks.The cited work also identifies online probabilistic analysis as a possible extension.Source: §4.1.4 · pp. 13–14

4.2. Data-driven machine learning

This section reviews machine learning as a data-driven approach to power-system operations, outlining learning paradigms and applications that report operational benefits alongside unresolved constraints.

4.2. Data-driven machine learning

  • Machine learning can use historical data to capture typical actor characteristics and improve scalability and computational efficiency relative to optimisation approaches.Source: §4.2 · p. 14

4.2.1. Learning approaches

  • Supervised learning uses labelled data, unsupervised learning identifies hidden features without labels, and reinforcement learning adapts through interaction with an environment.Source: §4.2.1 · p. 14

4.2.2. Research and implementations

  • Applications span scheduling, residential storage, EV exchange, demand-response bidding, and load control, with reported efficiency, cost, or load-balance benefits and remaining physical or participation challenges.The reviewed studies also identify concerns including grid congestion, vehicle-to-grid constraints, synchronized demand-response peaks, and load changes between set-points.Source: §4.2.2 · pp. 14–15

4.3. Remark of practicalities, benefits, and challenges

The section describes four levels of AI implementation in power-system operations, summarizes potential benefits, and identifies data and physical-constraint challenges.

4.3.1. Practical implementations

  • AI implementation progresses from responsive assistance through predictive and prescriptive functions toward future autonomous operation without system-operator intervention.The four levels are illustrated in Fig. 6.Source: §4.3.1 · p. 15, Figure 6 · p. 15

4.3.2. Benefits

  • AI may automate decisions for generators, operators, and consumers, improve computational efficiency for real-time operations, and coordinate system actors.Source: §4.3.2 · p. 16
  • AI models may scale from households to regions using historical data and adapt to both common low-impact issues and rare high-impact events.Source: §4.3.2 · p. 16

4.3.3. Challenges

  • When historical data are too sparse to train an accurate model, avoiding overfitting or underfitting remains a challenge.Source: §4.3.3 · p. 16
  • AI-generated control decisions must align with the physical constraints of power systems, which remains an implementation challenge.Source: §4.3.3 · p. 16

5. Conclusion

The review examines regulatory, market, and operational requirements for integrating prosumers into smart grids through blockchain and AI. It concludes that these technologies can support prosumer trading, control, and policy functions only if barriers in all three areas are overcome.

5. Conclusion

  • The review identifies regulatory, market, and operational perspectives as central to integrating prosumers and decarbonising power systems.Source: §5 · p. 16
  • Regulatory progress requires dynamic, decentralised measures defining prosumers’ responsibilities, assets, roles, and models.Source: §5 · p. 16
  • Local market design must align individual profits with system benefits through rules, pricing, transactions, trading platforms, and auctions.Source: §5 · p. 16
  • Operational integration requires fitting AI models to physical power-system constraints and developing digitalised, interoperable systems linking digital and physical layers.Source: §5 · p. 16
Loading 2202.10098v1…