A Power Prediction Method Combining the Principles of Symbiotic Evolutionary Bioimmunity

Minjing Yang
Power Dispatch and Control Center, Guangdong Power Grid Co., Ltd, China Southern Power Grid, Guangzhou, Guangdong, 510000, China
Binghong Su
Power Dispatch and Control Center, Guangdong Power Grid Co., Ltd, China Southern Power Grid, Guangzhou, Guangdong, 510000, China
Qinwei Duan
Power Dispatch and Control Center, Guangdong Power Grid Co., Ltd, China Southern Power Grid, Guangzhou, Guangdong, 510000, China
Yashan Zhong
Power Dispatch and Control Center, Guangdong Power Grid Co., Ltd, China Southern Power Grid, Guangzhou, Guangdong, 510000, China
Jiaxin Zhuo
Beijing TsIntergy Technology Co., Ltd, Beijing, 100084, China
Xuanli Lan
Beijing TsIntergy Technology Co., Ltd, Beijing, 100084, China

Abstract:

In this paper, an intelligent power prediction model is constructed based on the principle of symbiotic evolutionary bio-immunity. Firstly, the power grid control system is established by combining the biological immune response mechanism, and the large number of antibodies produced by using the biological B-cell recognition technology as well as the activation and clonal amplification technology rapidly binds to the invading antigen and inactivates it for elimination. And using the memory nature of immune cells, immunize against the reoccurrence of power failures or disturbances of the same nature. Then, a complete neural network is formed through the immune regulation mechanism in symbiotic evolutionary organisms, and the balance of the network is maintained based on the diversity of the organisms, and the performance in the network where the neurons are located is used as the basis for the evaluation of the power adaptation. Finally, the construction of the power prediction model is completed by predicting the positional weights between power load points in the symbiotic neural network and calculating their similarity values. The results show that the error value of the intelligent power prediction model constructed in this paper for load value detection is always within 3%, and the accuracy of power prediction under different conditions is above 70%. It can be seen that the combination of symbiotic evolutionary bio immunity improves the power prediction accuracy and saves the prediction time.