“This paper introduces the research progress of energy storage power stations in the field of power system frequency regulation. Experts propose a dual objective optimization bidding strategy in the current market. The GRU neural network prediction model based on attention mechanism is used to predict the frequency regulation clearing price and electricity price. The probability distance fast reduction method is used to generate a typical frequency regulation signal set to simulate the uncertainty of frequency regulation signals. The cost of frequency regulation capacity and mileage of energy storage power stations are calculated, and the settlement method of frequency regulation revenue is determined. A dual objective bidding strategy is proposed with the goal of improving the winning bid rate and revenue rate. The SOC utilization rate is considered, and the initial SOC is set Scope, improve power station revenue, introduce deviation assessment in real-time stage, optimize frequency regulation capacity through rolling, verify that this strategy can reduce the cost of secondary frequency regulation of energy storage power stations by 0.88%, increase the cost-benefit ratio by 0.61%, and provide a solution to motivate more energy storage power stations to participate in the frequency regulation auxiliary service market.”