There are many factors affecting the short-term load forecasting of power systems, therefore, when performing short-term load forecasting, the more types of factors are considered, the higher the accuracy of predicting is. On the basis of considering environmental factors, the improved whale algorithm optimization Elman neural network model considering the factors of energy storage scheduling under time-of-use price is constructed. Under the smart grid, because the energy storage scheduling can change the traditional load curve, firstly, based on the time-sharing price, the energy storage scheduling model is constructed to analyze the charging and discharging behavior of the energy storage users in each period. Then, because the Elman neural network has the disadvantages of slow convergence and easy to fall into local optimum, a Modified Whale Optimization Algorithm (MWOA) is proposed to optimize the weight and threshold of the neural network, which further improves the convergence and global search of the neural network excellent ability. Finally, constructing a short-term load forecasting model considering energy storage scheduling factors, the analysis of the data of a certain power grid from July to August of 2018 is carried out and compared with other prediction models mentioned. The error results analysis shows that the proposed method has higher prediction accuracy and faster convergence. This work is supported by Science and Technology Project of State Grid Corporation of China (No. SGTJDK00D WJS1700034) “Research and Application of Stochastic Power Sources Oriented Multiple Load Active Response and Predictive Control Technology”.