Bidding strategy for photovoltaic storage station in the electricity spot market based on photovoltaic power prediction

In recent years, with the continuous expansion of the scale of China ‘s new energy units, the proportion of new energy participating in the electricity spot market (ESM) has been increasing. High-precision short-term (ST) and ultra-short-term (UST) prediction results of new energy are needed to meet the needs of the ESM, and the operation mechanism of new energy prediction participating in the ESM needs to be improved. Photovoltaic power prediction (PPP) is the decision-making basis for photovoltaic storage station (PSS) to participate in the ESM, and its results directly affect the revenue of the PSS in the ESM. However, the coupling of PPP in the ESM is not clear, and the bidding strategy (BS) of PSS considering the deviation between real-time (RT) and day-ahead (DA) PPP is not clear. To this end, this paper first proposes a PPP method based on TCN-LSTM-Attention, and combines the scheduling requirements of the ESM to form a multi-time scale collaborative PPP and ESM clearing system. Then, this paper proposes a two-stage power-time division BS for PSS to participate in the ESM, considering how PPP results at different time scales influence the day-ahead market (DAM) pre-clearing and real-time market (RTM) formal clearing processes. Based on the coupling between photovoltaic and energy storage, this paper constructs a two-stage two-layer model for PSS to engage in volume bidding and maximize their profits. Finally, taking the measured data of PSS in Mengxi area of China as an example, the simulation results showed that the prediction model proposed in this paper achieved high accuracy in both ST and UST prediction, and the proposed two-stage BS considering the difference between DAM and RTM time scales can make the PSS have more abundant scheduling space to stabilize the deviation penalty cost caused by the fluctuation of photovoltaic output and maximize the profit of the stations under the condition of satisfying the actual market operation.