Research On Online Operation Optimization Methods Of Hydrogen-based Energy Systems Under Uncertainties

According to the China Energy Development Report in 2022,83.4% of China’s energy demand in 2021 was met by fossil fuels.With the increasing demand for energy,non-renewable energy sources such as fossil fuels are in danger of exhaustion.Moreover,the environmental impact of high carbon emissions from fossil fuels cannot be ignored.In order to address the above challenges,renewable energy substitution should be valued and energy conservation should be strengthened.Vigorously developing hydrogen-based energy system and optimizing its operation is helpful for the local consumption of distributed renewable energy,which is of great significance and application value for realizing the strategic goal of "double carbon" and reducing the economic cost of energy system.Therefore,online operation optimization methods of hydrogen-based energy systems under uncertainties are studied in this paper.Firstly,the operation cost minimization problem of hydrogen-based energy systems under uncertainties is formulated.It is very difficult to solve this problem because of uncertain parameters,time-coupling constraints related to the storage level in the hydrogen tank and nonlinear constraints.Therefore,an online operation optimization method based on Lyaplov optimization technology for hydrogen-based energy systems is proposed in this paper.The proposed method does not require any prior knowledge of uncertain parameters.Simulation results based on real data show that the proposed online operation optimization method can flexibly reduce carbon emissions according to the given carbon emission penalty parameter.Compared with the rule-based comparison scheme,the proposed online operation optimization method can reduce the total operation cost by0.298%-13.436%.Secondly,by taking the model nonlinearities of the electrolyzer and fuel cell into consideration,the operation cost minimization problem of hydrogen-based energy systems under uncertainties is formulated.Due to the existence of uncertain parameters,time-coupling constraints related to the storage level in the hydrogen tank,nonlinear constraints and nonlinear working efficiency models,it is difficult to solve this problem effectively using traditional optimization methods.Therefore,an intelligent operation optimization method based on Proximal Policy Optimization and human expert knowledge for hydrogen-based energy systems is proposed in this paper.The proposed method can deal with nonlinear models and does not require any prior information of uncertain parameters.The simulation results based on real data show that compared with the rule-based comparison scheme,the proposed intelligent operation optimization method can reduce the total operation cost by0.8-20.5%.Finally,the research work of this paper is summarized and the future research is prospected.