遗传神经网络优化预测方法研究及其应用

Optimization of Forecasting Method Based on Genetic Neural Network and its Application

根据短期电力负荷预测的特点，将遗传算法和人工神经网络相结合，提出一种负荷预测新算法─—遗传神经网络优化预测方法。该方法明显地提高了模型的优化能力，有效地克服了人工神经网络学习速度慢、存在局部极小点的固有缺陷。经实例验证，该方法能有效地提高预测精度和速度。

According to the characteristics of the short-time power load forecasting, the authors propose a newload forecasting model based on the combination of genetic algorithm and neural network. The intrinsic defectsof artificial neural network, such as its slow learning speed and local minimum points, are overcome. It can beseen horn a case study that the proposed method can improve the accuracy and the speed of load forecasting effectively.