Can game theory give a more realistic understanding of transactive energy and demand response efforts?

By Soheil Mohseni and Alan Brent

What makes people cooperate or not? The answer is simple: they generally do not cooperate with anybody on any task they can do themselves. That is, we tend to make cooperative commitments to achieve gains we perceive are worth the cost of commitment and that we cannot achieve by ourselves. A useful analytical tool to study the value bases that govern the behaviour of the social structure is game theory.

That said, the ability to harness consumers’ flexibility is opening new services to energy cooperatives, from the use of peer-to-peer and transactive community energy markets to dispatchable load aggregation that drives economies of scale — supporting cost-effective energy democracy and self-sufficiency.

New research into the applications of game theory in designing local flexibility energy markets offers key insights into the dynamics and patterns of social convergence into the uptake of incentive-based demand response (DR) programs, with associate financial implications for designing renewable energy systems.

But first, let’s learn a little bit more about demand-side management and game theory.

Demand-side management and game theory

Interruptible DR programs have traditionally been targeted at large-scale industrial and commercial end-users for emergency periods when the utility cost of service exceeds a pre-defined threshold. However, now with new advances in communications and control technologies, smaller incentive-responsive demand-side flexibility resources are set to form a core part of the utilities’ overall resource portfolios — to enable non-dispatchable distributed energy resources (DER) to provide cost-efficient electricity services at scale.

This has brought to light the importance of third-party DR aggregators, who enlist end-users of the same load segment and give them enough scale to participate in DR curtailment services and sell the bundled load reduction to utilities.

In this context, market-based aggregator-mediated flexibility procurement interventions are increasingly considered as an effective approach to ensure that system-level dispatch of DR capacity is aware of the value of flexibility to all actors. That is, market-driven flexibility procurement solutions with non-discriminatory access at the wholesale and retail levels are proven to maximize the associated social welfare, liquidity of customer-supplied DR capacity, fiscal transparency, as well as the robustness of DR provision in the long run — necessary to improve the security and resilience of DR-integrated power systems of the future.

To this end, market designs for DR aggregation explore the conditions under which competitive market equilibrium is simultaneously reached at the upper utility level and the lower distribution network level.

Schematic showing a conceptual design of a typical aggregator-mediated transactive energy framework (two-sided market).  

In this setting, borrowing ideas from game theory is seen as a key prerequisite to project the process of negotiations towards minimizing the overall operational system costs, whilst additionally optimizing the individual payoffs (numeric representations of the decision-makers’ preferences over all possible outcomes of the game). More specifically, game theory provides a platform to realistically characterize the strategic interactions among the involved instrumentally rational decision-makers — who pursue their own self-interest — in market-based, aggregator-activated DR programs.

Crucially, game theory, a subdiscipline of behavioral economics, provides tools for dealing with:

  • undefined sets of the associated agents’ strategies (alternatives among which each decision-maker chooses),
  • decision-makers’ bounded rationality, which refers to the fact that humans are faced with limited cognitive abilities that constrain their problem-solving abilities, and more generally,
  • the uncertainty in conjectures about possible errors in the choices of other players due to incomplete information.

Put simply, cooperative game theory studies how agents cooperate as coalitions in unstructured interactions to capture value, whereas non-cooperative game theory focuses on modeling the actions of agents maximizing their utilities in a defined procedure.

Now let’s see how renewable energy businesses can benefit from having greater visibility over potentially dispatchable demand-side flexibility resources in the long run in the light of stable interactive strategies envisioned using game theory — and how that can drive the deployment of community-level clean energy projects.

Game-theoretic DR in the long run

In addition to enabling the customers to reap the full benefits of their demand flexibility potential, as well as providing aggregators with a consistent revenue stream, applying insights from game theory to the study of small- to medium-scale interruptible DR provision has important long-term financial consequences for utilities in terms of sourcing costs. Specifically, it helps utilities better forecast long-term load power demand, which results in narrowing the required system adequacy margins. That is, over the long-term, well-coordinated and well-forecasted DR programs reduce aggregate generation capacity requirements, allowing utilities to (1) build less new peaking generation capacity, (2) defer or deter excessive costly investments in transmission network capacity expansions, and (3) increase the penetration of non-dispatchable renewables.

Additionally, using consumer load response as an effective means of system control through modeling the normative behavior of active economic agents based on game theory, is able to save money and resources, minimize environmental impacts, and help move towards a robust and equitable allocation of the costs and benefits of third-party DR procurement among the players. Moreover, game-theoretic DR management systems leave no room for free-riding actions of aggregators (resulting from under-payments to enrolled customers), whilst allowing for the consideration of the indirect effects of DR (for example, the so-called rebound effect, which refers to increased electricity usage required to compensate for the earlier load reduction above the expected range by system operators). This has important implications for improving the accuracy of renewable energy pre-feasibility and business case analyses during the planning phases of both greenfield and brownfield renewable and sustainable energy systems.

A real-world problem

In our research, we used a game-theoretic model to better forecast the availability of smaller demand-side flexibility resources across various customer segments — residential, commercial, industrial, agricultural, and electric vehicle-charging power loads — while optimally designing grid-connected 100%-renewable energy systems. To this end, we have developed a two-stage demand-side management market framework that implements a uniform price (non-discriminatory) DR auction at the wholesale and retail levels, and is aware of the sectoral price elasticity of DR supply. Accordingly, we tested the performance of both cooperative and non-cooperative model variants by applying them to a hydrogen-based micro-grid specifically conceptualized for a 1,000-strong community — that swells to 8,500 people during skiing season — in the town of Ohakune, New Zealand.

Crucially, we found that capturing the essential concept of strategic interdependence in cooperative aggregator-mediated DR scheduling games clearly makes everyone better off over the long term — and improves the welfare of the society as a whole. More specifically, we demonstrated that taking a cooperative approach to demand-side management — in that no agent has any incentive to deviate from truth-telling — is able to reduce the system’s expected life-cycle costs by as much as 21% (equating to a saving of NZ$5.5m) and 32% (NZ$10m) respectively compared to the non-cooperative approach and non-DR-integrated case — by smoothing the demand profile to reduce peaks.

Summary of comparative results obtained from the application of cooperative and non-cooperative game theory to the procurement of small-scale DR in a semi-urban New Zealand case example.

Most of all, game theory is revealing how significant the financial consequences are of a fair allocation of the overall benefits that accrue from activating smaller sources of DR among the market participants — and how that can be a key to unlocking universal access to affordable, reliable, sustainable energy sources for all.

About the Authors

Soheil Mohseni is currently pursuing a PhD at Victoria University of Wellington, New Zealand. His PhD thesis revolves around developing a demand response-centered method to optimally size on-/off-grid micro-grids using AI-based meta-heuristic optimization algorithms considering model-inherent parametric uncertainties. In his project, he engages directly with the public- and private-sector energy stakeholders, as well as remote and rural communities in New Zealand, to understand how energy planning optimization can drive the deployment of stand-alone and grid-connected renewable energy systems. His research interests include: behavioral demand response; cost-optimal integration of distributed energy resources, green vehicles, and electrified heating technologies into renewable and sustainable energy systems; quantification of the uncertainties associated with the forecasted electrical/thermal loads, climatic data, green vehicles’ power consumption and driving patterns, as well as electricity price; application of evolutionary algorithms to the optimal operational scheduling and planning of renewable energy systems; and demand response-integrated techno-economic analysis of poly-generation energy networks. He also holds a BSc (with First-Class Honors) in power engineering from Kermanshah University of Technology, Iran, as well an MSc (with First-Class Honors) in power engineering, and power systems from the University of Guilan, Iran. Contact him at

Since 1995, Professor Alan Brent has consulted to a variety of industry and public sectors, in a number of countries, in the fields of environmental engineering and management. His research focus now revolves around sustainable technology management, with an emphasis on the energy sector.

In 2017, he joined the School of Engineering and Computer Science at Victoria University of Wellington as the inaugural holder of the Chair in Sustainable Energy Systems. The Chair supports the transition to sustainable energy of the New Zealand and regional economy, and society. It is aligned with the strategic sustainability focus at Victoria, to meet current and future challenges, by taking an inter-disciplinary approach, and engaging with partners across society in trans-disciplinary ways. Contact him at

Previous articleSustainability games: A solution for residential demand response
Next articleTransforming utility customer service: Helping decarbonize the residential sector

No posts to display