The optimal bidding strategy for time-shiftable loads in the day-ahead market and real-time market is studied respectively in an advanced power market environment. In the bidding optimization

it must be guaranteed that the time-shiftable loads consume a certain amount of electricity within deadline. Considering the deviation of short-term load forecasting results of the traditional average value method

an autoregressive integrated moving average(ARIMA)model is proposed to analyze the optimal bidding strategy for time-shiftable loads. The model parameters are estimated by the characteristics of historical electricity price

so as to predict the price curve of the day-ahead price and the real-time price on the second day

based on which the bidding strategy is optimized. Then

a greedy algorithm is proposed to adjust the energy bidding considering the power usage limit in each period. Finally

the proposed method is verified by a case study

which shows that the method can help save more energy procurement cost than the methods simply using the expected value of historical data as parameters

and adjust the energy bidding effectively according to the power usage limit as well.