Load energy storage double-optimization prediction method and device and medium

Abstract

The embodiment of the invention discloses a load energy storage double-optimization prediction method and device and a computer readable storage medium. The method comprises the steps of initializingnetwork parameters of a neural network; wherein the network parameters comprise a weight parameter, a bias parameter and a newly introduced energy storage parameter; constructing a neural network model by utilizing the initialized network parameters; establishing an objective function containing load cost and energy storage cost and constraint conditions; and based on an objective function and constraint conditions, training the neural network model by using a gradient descent method to obtain a training result containing a load cost matrix and an energy storage cost matrix. By combining the improved neural network model with the gradient descent method, double-optimization prediction of the optimal load and energy storage of a power selling company and a large user in the microgrid is realized, the time and cost spent on energy storage and load prediction are effectively saved, and the power selling company can submit a better application scheme in the power market conveniently.

Description

technical field [0001] The invention relates to the technical field of power grids, in particular to a load-energy-storage double-optimized prediction method, device and computer-readable storage medium. Background technique [0002] The electricity market is an important link to maintain the stable operation and management of the power system. With the deregulation of the electricity market intensifying around the world, load forecasting has become a very important part of the electricity market to maintain the balance between supply and demand of the power system. As a transferable load in the electricity market, energy storage can effectively deal with the stability and reliability of the power system. [0003] Energy storage is widely used in microgrids, which can effectively reduce the fluctuations caused by the intermittency of new energy power generation in the grid, and also play an important role in the "peak shaving and valley filling" of demand-side response. The...

AI Technical Summary

Problems solved by technology

With the wide application of new energy and the time delay in power system scheduling, the power market is often changing rapidly. At this time, the time spent on the two optimizations will miss the best trading opportunity, reducing the market sensitivity of power sales companies and large users. and decision accuracy

Images