CAJ | 학술논문

为了提高瓦斯浓度预测的精度和稳定性,提出了将遗传算法( GA)与BP神经网络结合的预测方法。利用BP神经网络能以任意精度逼近非线性函数的优点,结合遗传算法的全局搜索能力,优化神经网络权值和阈值,建立GA—BP混合算法模型预测瓦斯浓度。实验结果表明,GA—BP算法与BP神经网络相比,具有较高的预测精度和较强的稳定性。
위료제고와사농도예측적정도화은정성,제출료장유전산법( GA)여BP신경망락결합적예측방법。이용BP신경망락능이임의정도핍근비선성함수적우점,결합유전산법적전국수색능력,우화신경망락권치화역치,건립GA—BP혼합산법모형예측와사농도。실험결과표명,GA—BP산법여BP신경망락상비,구유교고적예측정도화교강적은정성。
In order to improve the accuracy and stability of the gas concentration prediction, a prediction method of combining genetic algorithm ( GA) and BP neural network was proposed. By using the advantage of BP neural network which can approach the nonlinear function with any accuracy, combining with the overall search ability of the genetic algorithm and optimizing the neural network weights and thresholds, a GA-BP hybrid algorithm model for gas concentration prediction was established. The experimental results show that GA-BP algorithm has higher prediction accuracy and stronger stability as compared with the BP neural network.