An Artificial Chromosomes Embedded Genetic Algorithms for Smart Grid Power Demand Forecast

This research develops a hybrid model by integrating K-mean, Neural Network, and Artificial Chromosomes embedded Genetic Algorithms to forecast the electrical load. This hybrid model encompasses two novel concepts: 1. Under the expectations of load balancing, we clustering 272 electrical substations into different clusters, and develop a forecast model of each cluster. 2. Instead of the generation mechanisms of Genetic Algorithms, the artificial chromosome is injected to seek a higher accurate. Numerical data of various affecting factors and actual electrical load of 48 months of 272 electrical substations are collected and fed into the hybrid model for future monthly amount. Experimental results show a higher accurate of our model compare with other two traditional forecasting model.