Deep reinforcement learning framework for dynamic pricing demand response of regenerative electric heating

Applications of electric heating, which can improve carbon emission reduction and renewable energy utilization, have brought new challenges to the safe operation of energy systems around the world. Regenerative electric heating with load aggregators and demand response is an effective means to mitigate the wind curtailment and grid operational risks caused by electric heating. However, there is still a lack of models related to demand response, which results in participants not being able to obtain maximum benefits through dynamic subsidy prices. This study uses the Weber-Fechner law and a clustering algorithm to construct quantitative response characteristics models. The deep Q network was used to build a dynamic subsidy price generation framework for load aggregators. Through simulation analysis based on the evolutionary game model of a project in a rural area in Tianjin, China, the following conclusions were drawn: compared with the benchmark model, regenerative electric heating users can save up to 8.7% of costs, power grid companies can save 56.6% of their investment, and wind power plants can increase wind power consumption by 17.6%. The framework proposed in this study considers user behavior quantification of demand response participants and the differences among users. Therefore, the framework can provide a more reasonable, applicable, and intelligent system for regenerative electric heating.

Automated summary

This study constructs quantitative response characteristics models for demand response, providing a more reasonable and intelligent solution for regenerative electric heating systems. Through simulation analysis, it is proven that regenerative electric heating users can save costs, power grid companies can reduce investment, and wind power plants can increase wind power consumption.