Application and research of improved artificial immune network to power short-term load forecasting
Article
Figures
Metrics
Preview PDF
Reference
Related
Cited by
Materials
Abstract:
According to the deficiencies of load forecasting model at present，a short-term load forecasting model based on optimized clone immune and BP neural network (BPNN) is presented. In the design of artificial immune network (AIN)，the principle of immune network regulation is used in a creative way and the method of immune programming is used to evolve the network structure. The probability of selective antibody concentration，a new fitness function of neurons，a new mutation operator and a new self-adaptive chaos mutation operator are adopted in the AIN. The excitation function controls the BP algorithm which greatly accelerates convergence of BP training，the self adaptable strategy based on clone immune optimizes the controlled BP algorithm，and it improves its global searching ability better than the BP algorithm optimized by chaos and avoids the algorithm to be trapped in local minimum and improves the prediction accuracy.
