Steam load forecasting based on chaos theory and LSSVM.

The heating steam is an important secondary energy, so it is of great significance to predict the required steam load in the future hours, which is important for the thermal power plant to provide users with high quality heat load securely and economically. Steam load time series proves to be with chaos characteristics. According to Takens theorem, delay time and embedding dimension are calculated respectively using C-C method and Cao method, and the steam load time series is reconstructed in phase space, and then the steam load forecasting model is established using least squares support vector machine (LSSVM). A SA_WPSO algorithm (improved particle swarm optimization (PSO) with simulated annealing algorithm (SA)) is proposed to implement the optimization of LSSVM parameters. The simulation results show that the method can achieve good prediction results. [ABSTRACT FROM AUTHOR]