Development of control technology for virtual power plants that reflects the diverse values of different communities in real time

Aiming to achieve flexible energy deployment that reflects local needs, such as CO2 reduction and cost reduction

Summary

Hitachi, in partnership with Kyoto University, has developed a new system control technology that enables the stable operation of a virtual power plant (VPP)*1 while reflecting the diverse values of different cities, regions, and other communities in real time (Figure 1). This technology allows for more flexible system control based not only on conventional fixed indicators, such as economic efficiency, but also on value-based indicators that vary depending on the circumstances, such as CO2 reduction and convenience. To make this advanced VPP control a reality, a series of technologies have been developed: model predictive control (MPC),*2 which allows users to dynamically distribute energy resources; preference learning,*3 which automatically reflects the values of the recipient communities; and robust control,*4 which achieves stable operation. During simulated validation, when emphasis was placed on CO2 emissions, a reduction of up to approximately 20% was achieved, and when emphasis was placed on cost, a reduction of up to approximately 16% was achieved. These results confirm the stable operation that reflects different values of different communities. The plan going forward is to collaborate with customers and partners to make advances in the technology and validate its effectiveness in real-world applications, while repeatedly carrying out verification with community participation. By leveraging AI, as one of the technologies that support Lumada 3.0, the new system control technology will promote energy operations that reflect the diverse values of local communities and contribute to the realization of a sustainable society and the enrichment of people’s lives.

*1 VPP: A system that connects and controls distributed energy resources (solar power generation, storage batteries, electric vehicles, etc.) through the use of information technology to operate as if they constitute a single large-scale power plant.
*2 Model predictive control: A method to control a system by predicting the future state through the use of mathematical models, etc. and computing the optimal set of operations.
*3 Preference learning: A technology that models and predicts preferences and evaluation criteria through the machine learning of information concerning selections made by users and decision makers.
*4 Robust control: A control method that ensures the stable operation of a system despite any unforeseen changes in the surrounding environment or the presence of any modeling errors in the system.

Background and issues

As the transformation into a decarbonized society and a distributed energy society progresses, there is a need for control technology that secures the stability and efficiency of the energy system as a whole, while reflecting the diverse values of the recipient regions and residents. Yet, because conventional VPP control is optimized based on fixed target indicators, such as the reduction in operating cost, it was difficult to flexibly reflect the values of recipient regions and residents in real time without limiting them to economic values. Against such a backdrop, the development of a new control technology based on a community participation model can flexibly accommodate a diverse set of values, and is becoming critical to attaining a sustainable society.

Technologies and solutions developed to solve issues

To address these issues, Hitachi, in partnership with Kyoto University, has developed a technology to optimize the control of VPP while reflecting the diverse values of different communities in real time (Figure 2). The key features of this technology are as follows.

1. Model predictive control (MPC) technology achieves a flexible operational optimization (Figure 2, (1))
By performing feedback control while predicting future behavior based on a mathematical model, multiple energy resource facilities, such as power generation facilities and battery storage systems, are dynamically allocated, thereby achieving optimal operation that balances supply and demand, and satisfies various constraints. MPC is highly versatile as it can control a variety of targets by changing the mathematical model used.

2. Preference learning technology reflects diverse values (Figure 2, (2))
Community participants can intuitively adjust the control parameters without possessing any specialist knowledge by selecting the preferred operational outcome from two choices. Simply by repeating this selection process, the system will learn the participants’ preferences and decision-making criteria and automatically make adjustments. This in turn makes it possible to operate a VPP by flexibly reflecting the different objectives of different communities, such as enhancing economic value or reducing environmental impact.

3. Robust Control technology supports a stable operation
A robust control technology has been developed that utilizes double-loop feedback control*5 based on a combination of asset allocation control for reflecting the values of communities (Figure 2, (3)A: the inner control loop) and frequency stabilization control*6 for stabilizing the supply of electricity (Figure 2, (3)B: the outer control loop). This allows the deployment of individual energy resources to be adjusted flexibly in accordance with the intentions of the communities, while ensuring the overall balance in supply and demand and stabilizing the frequency of the electricity supplied. The technology even responds to unforeseen changes in circumstances, such as the uncertainties associated with models used in MPC, sudden fluctuations in demand, or changes in operational policy, while compensating for the resulting errors.

*5 Double-loop feedback control: A mechanism to achieve a more stable and flexible operation of a system by combining two control loops. The inner loop is used to make fine adjustments to the system, while the outer loop is used to maintain the overall balance and stability of the system.
*6 Frequency stabilization control: When there is an imbalance in the supply and demand of electricity, a fluctuation of the frequency of the electric current (the speed of the electrical wave) occurs. Frequency stabilization control adjusts the amount of power generation or consumption so as to stabilize this frequency, thereby preventing power outages and problems that affect electrical equipment and infrastructure.

Confirmed results

In simulated studies, when emphasis was placed on the reduction of CO2 emissions, a reduction of up to approximately 20% was achieved, and when emphasis was placed on cost, a reduction of up to approximately 16% was achieved. These results confirm the stable operation that reflects different values of different communities. Based on the findings, optimized energy deployment that responds to the diverse needs of different communities is expected to contribute to a sustainable society.

Future prospects

Going forward, collaboration with customers and partners will continue to make advances in the technology and validate its effectiveness in real-world applications, while repeatedly carrying out verification with community participation. In addition, efforts will be made to apply this technology to transportation infrastructure and other forms of distributed infrastructure systems and promote research of novel control concepts. Moreover, by leveraging AI, further advancements of this technology will be pursued as one of the core technologies supporting Lumada 3.0. Through energy deployment that reflects the diverse values of different regions and communities, Hitachi aims to realize a harmonized society where the environment, wellbeing, and economic growth are in balance.

Some of these results are scheduled for presentation at the 2025 IEEE International Conference on Systems, Man, and Cybernetics (SMC) to be held October 5–8, 2025, in Austria.

For more information, use the inquiry form below to contact the Research & Development Group, Hitachi, Ltd. Please make sure to include the title of the article.

https://www8.hitachi.co.jp/inquiry/hitachi-ltd/hqrd/news/en/form.jsp