基于神经网络的蓄电池容量预测研究

Prediction of battery capacity based on neural network

针对当前变电站直流系统中蓄电池容量难以预测的问题,本文引入神经网络对蓄电池容量进行预测研究。为实现对变电站直流系统蓄电池容量的准确预测,首先对当前国内外蓄电池容量预测研究现状进行分析,总结国内外研究成果进行下一步研究工作;然后,在吸取各研究经验的基础上,对蓄电池容量预测模型进行构建。在构建该预测模型时,本文主要采用模糊神经网络与改进学习算法构建起整个蓄电池容量预测模型;最后,通过具体的仿真实验对上述模型的可行性和科学性进行验证,进而验证本文构建的模型在预测变电站直流系统蓄电池容量上的正确性。结果表明,本文构建的神经网络模型在预测方面与实际的均方根误差最小,进而验证了本文构建算法的精度与实际最为接近。由此说明本文构建模型的科学性和正确性。通过以上的研究,为变电站直流系统蓄电池容量预测工作提供参考。

Aiming at the problem of unpredictable capacity of battery in DC system of substation at present,the battery capacity prediction based on neural network is introduced. In order to achieve accurate prediction of battery capacity of DC system in substation,firstly,the current research status of battery capacity prediction at home and abroad was analyzed,and the next research work was summarized. In the construction of the prediction model,on fuzzy neural network and improved learning algorithm are adopted;finally,the feasibility and scientificity of the above model are verified by specific simulation experiments,The validity of the model in predicting the storage capacity of DC system in substation is verified. The results show that the neural network model constructed in this paper has the smallest root mean square error in predicting and the actual,thus verifying the accuracy of the algorithm constructed in this paper is the closest to the actual. It shows that this model is scientific and correct. Through the above research,it provides reference for battery capacity prediction of DC system in substation.