Abstract
Energy storage has emerged as a crucial flexible regulation resource in the new power system, playing a pivotal role across the power generation, transmission, and distribution networks. China has entered a period of large-scale development for energy storage, making it imperative to optimize its planning and configuration for economic and efficient utilization. This paper delves into the current research status of energy storage planning methods and examines the impact of evolving trends, such as the diversification of energy storage investors and operators, the complexity of application scenarios and operation modes, and the proliferation of various energy storage technologies. Furthermore, we propose novel planning approaches tailored for renewable energy clusters, grid-side integration, and user-side utilization, considering shared mechanisms, market competition, and multi-energy storage cooperations. We also emphasize the importance of addressing the overall coordination across power system stages, the game mechanisms arising from diverse stakeholders, and the adaptability of planning schemes through transfer learning in multi-scenario and multi-objective contexts.
1. Introduction
China’s commitment to the “dual carbon” goals—peaking carbon dioxide emissions by 2030 and achieving carbon neutrality by 2060—necessitates a fundamental transformation of its energy and power supply system. This transformation is driven by an increasing reliance on green and low-carbon energy sources, particularly renewable energy (RE). However, the intermittency and volatility of RE pose significant challenges to the stability and reliability of the power grid. Energy storage systems offer a viable solution to address these challenges by providing flexibility and balancing capabilities.
1.1 Background and Significance
Energy storage technologies have advanced significantly in recent years, with lithium-ion batteries emerging as a prominent player. China’s National Development and Reform Commission (NDRC) and National Energy Administration (NEA) have set ambitious targets for energy storage development, aiming to achieve a cumulative installed capacity of 30 gigawatts (GW) by 2025. As of 2022, China’s total installed energy storage capacity surpassed 59.8 GW, accounting for 25% of the global total, with new energy storage technologies contributing over 13.1 GW.
1.2 Objectives and Contributions
This paper aims to provide a comprehensive overview of the research progress and future prospects of energy storage planning methods for the new power system. We analyze the current state of energy storage planning, highlight emerging challenges, and propose novel planning approaches tailored to various power system segments. Our key contributions include:
- A detailed analysis of the diversification of energy storage investors and operators, application scenarios, and technology types.
- Novel energy storage planning methods for renewable energy clusters, grid-side integration, and user-side utilization.
- Insights into the game mechanisms between diverse stakeholders and the transfer learning capabilities of planning schemes.
2. Current State of Energy Storage Planning Research
2.1 Overview
Energy storage planning research spans various aspects of the power system, including power generation, transmission, and distribution networks, as well as end-user applications. This section summarizes the current state of energy storage planning across these domains.
2.1.1 Renewable Energy Clusters
In renewable energy clusters, energy storage systems are primarily employed to improve the power output characteristics of RE sources, such as wind and solar, by mitigating output fluctuations, tracking generation plans, and reducing curtailment. Early research focused on single-point configurations within individual RE plants, while recent studies have explored shared energy storage systems across multiple RE sites.
2.1.2 Grid-Side Integration
Grid-side energy storage can defer transmission and distribution upgrades, lower network losses, enhance supply reliability, and provide ancillary services. Planning methods must consider multiple application scenarios, such as congestion relief, peak shaving, and frequency regulation, and address the complex interplay between energy storage systems and the grid infrastructure.
2.1.3 User-Side Utilization
User-side energy storage primarily targets peak shaving and arbitrage opportunities between peak and off-peak electricity prices. Different user types (residential, commercial, and industrial) have distinct load profiles and market participation strategies, necessitating tailored planning approaches. Furthermore, user-side energy storage can participate in demand response programs to maximize economic benefits.
2.2 Emerging Challenges
While energy storage planning research has made significant progress, several emerging challenges necessitate further investigation:
- Diversification of Stakeholders: The increasing diversity of energy storage investors and operators has transformed planning problems from single-entity optimizations to multi-agent games, requiring novel planning frameworks that account for competitive and cooperative interactions.
- Complexity of Application Scenarios: The proliferation of RE sources, electric vehicle charging stations, and distributed generation units has led to more complex and dynamic application scenarios, posing challenges for accurate scenario generation and feature description.
- Variety of Energy Storage Technologies: The availability of multiple energy storage technologies (e.g., lithium-ion batteries, pumped hydro storage, flywheels) necessitates planning methods that can accommodate diverse technical and economic characteristics.
3. Novel Energy Storage Planning Methods
To address the aforementioned challenges, we propose novel energy storage planning methods tailored to renewable energy clusters, grid-side integration, and user-side utilization.
3.1 Renewable Energy Clusters
3.1.1 Energy Storage Planning Considering Time-Space Complementarity
To maximize the efficiency of shared energy storage systems in renewable energy clusters, it is crucial to exploit the temporal and spatial complementarity of RE output. This involves analyzing power output patterns across different spatial scales (e.g., plant, cluster, region) and temporal scales (e.g., seconds, hours, days, seasons).
Table 1: Example of Time-Space Complementarity Analysis
| Spatial Scale | Temporal Scale | Analysis Approach |
|---|---|---|
| Plant Level | Seconds-Hours | Power fluctuation smoothing |
| Cluster Level | Hours-Days | Load shifting and peak shaving |
| Regional Level | Days-Months | Seasonal storage optimization |
3.1.2 Shared Energy Storage Architecture and Revenue Mechanisms
Shared energy storage systems require robust architectures and fair revenue distribution mechanisms to incentivize participation. Blockchain technology can facilitate secure transactions and smart contract execution, enabling efficient and transparent sharing of energy storage resources.
3.2 Grid-Side Integration
3.2.1 Multi-Objective and Multi-Agent Planning
Grid-side energy storage planning must consider multiple objectives (e.g., reliability, economic efficiency, environmental impact) and account for the interactions between diverse stakeholders (e.g., utilities, independent power producers, regulators). A multi-layer planning framework that integrates game theory can effectively address these complexities.
3.2.2 Data-Driven Transfer Learning for Cross-Domain Planning
Transfer learning can enhance the adaptability of grid-side energy storage planning by leveraging existing planning knowledge and experience. This approach allows for rapid deployment of effective planning strategies in new scenarios, reducing computational costs and improving planning efficiency.
3.3 User-Side Utilization
3.3.1 Multi-Energy Load Forecasting and Scenario Generation
User-side energy storage planning necessitates accurate forecasting of electrical, thermal, and potentially hydrogen loads. Advanced machine learning techniques, such as long short-term memory (LSTM) networks, can improve load prediction accuracy, informing more effective planning decisions.
Table 2: Comparison of Load Forecasting Methods
| Method | Accuracy | Computational Complexity | Data Requirements |
|---|---|---|---|
| ARIMA | Moderate | Low | Time-series data |
| Random Forest | High | Moderate | Feature engineering |
| LSTM | Very High | High | Large datasets |
3.3.2 Game-Theoretic Planning under Cooperative and Non-Cooperative Settings
User-side energy storage planning must consider both cooperative and non-cooperative interactions among diverse stakeholders, such as residential users, commercial enterprises, and utilities. Game-theoretic approaches can model these interactions and inform optimal planning decisions.
4. Future Prospects
4.1 Overall Coordination Across Power System Stages
Future energy storage planning must consider the overall coordination across different stages of power system development. This involves quantifying the overall demand for energy storage, developing flexible planning frameworks that can adapt to evolving system conditions, and ensuring that energy storage deployments do not unnecessarily increase electricity costs for consumers.
4.2 Game Mechanisms and Stakeholder Interactions
The increasing diversity of energy storage investors and operators necessitates a deeper understanding of the game mechanisms that govern their interactions. This includes exploring cooperative and non-cooperative game-theoretic models, designing effective incentive mechanisms, and developing robust optimization algorithms that can account for the strategic behaviors of diverse stakeholders.
4.3 Adaptability and Transfer Learning in Multi-Scenario Planning
To enhance the adaptability of energy storage planning methods, future research should focus on transfer learning approaches that can leverage existing planning knowledge and experience to inform planning decisions in new scenarios. This includes developing data-driven models that can identify common patterns and features across different planning tasks and applying these insights to inform optimal planning decisions in dynamic and uncertain environments.
5. Conclusion
Energy storage has emerged as a critical flexible regulation resource for the new power system. This paper has provided a comprehensive overview of the current state of energy storage planning research, highlighted emerging challenges, and proposed novel planning approaches tailored to renewable energy clusters, grid-side integration, and user-side utilization. We have also discussed the importance of overall coordination across power system stages, the game mechanisms that govern stakeholder interactions, and the adaptability of planning methods through transfer learning. Future research should continue to explore these areas to ensure that energy storage deployments are optimized for economic efficiency, environmental sustainability, and grid reliability.