Data-Driven State Prediction Method for Economic Dispatch Problem

Strategic Insights -

Research at Lingnan University has developed a data-driven state prediction method for the economic dispatch problem, which is essential for power system operations. The new method uses a symmetrical convolutional neural network (CNN) structure to learn the states from the concatenation of input load and operating limits. This architecture effectively extracts multi-scale features through multiple convolutional layers, pooling, and upsampling operations. The proposed method can handle a varying number of energy storage systems in real-world deregulation of electricity markets.

Key Takeaways:

  • The economic dispatch problem is critical for power system operations, and traditional methods have limitations in handling large-scale energy storage integration.
  • The proposed data-driven state prediction method uses a CNN structure to learn the states from input load and operating limits, effectively extracting multi-scale features.
  • The method can handle a varying number of energy storage systems in real-world deregulation of electricity markets.
  • The effectiveness of the proposed method is demonstrated in the IEEE 118-bus test system and a real-world 661-bus utility system.
  • The study highlights the importance of developing efficient methods for economic dispatch problems, particularly in the context of increasing energy storage integration.
  • Xiang Pan, Wei Lin, Linze Yang, and Yanfang Mo are the authors of the study, which was published in the International Journal of Electrical Power & Energy Systems.
  • The study uses a data-driven approach to predict states in the economic dispatch problem, leveraging convolutional neural networks to extract multi-scale features.

Statistics:

  • The proposed method is tested on the IEEE 118-bus test system and a real-world 661-bus utility system.
  • The CNN structure used in the proposed method consists of multiple convolutional layers, pooling, and upsampling operations.
  • The method can handle a varying number of energy storage systems, which is a critical challenge in real-world deregulation of electricity markets.
  • The study demonstrates the effectiveness of the proposed method through simulation results.
  • The International Journal of Electrical Power & Energy Systems is the publisher of the study, and a free version of the article is available online.

Sources:

  • CNN-based state prediction for a varying number of storage in economic dispatch. International Journal of Electrical Power & Energy Systems, 2025,168():110590.
  • NewsRx. Data from Lingnan University Update Knowledge in Electrical Power and Energy Systems (CNN-based state prediction for a varying number of storage in economic dispatch). Energy Weekly News. July 11, 2025; p 66.