摘要
针对汽车驾驶环境热舒适性指标预测平均热感觉(predicted meanvote,PMV)计算复杂、预测精度不高的问题,提出了改进布谷鸟搜索(cuckoo search,CS)算法优化RBF神经网络的汽车热舒适性预测模型(改进CSRBFNN)。采用自适应步长和高斯扰动因子对CS算法进行改进,并用其对RBF神经网络的中心点c和宽度参数b进行优化。将改进CS-RBFNN与CS-RBFNN和PSO-RBFNN模型的预测结果进行对比,结果表明:改进CSRBFNN模型的均方根误差(root meansquareerror,RMSE)值分别降低了9.2%和35.5%,具有更高的预测精度。当RBFNN隐含层神经元个数增加时,预测精度有所提高,但收敛速度降低,运行时间变长。
To solve the problem that the calculation of the thermal comfort index PMV of automobile driving environment is complex and the prediction accuracy is not high,an improved cuckoo search(CS)algorithm is proposed to optimize the prediction model of automobile thermal comfort of RBF neural network(improved CS-RBFNN).The CS algorithm is improved by using adaptive step size and Gaussian disturbance factor,and used to optimize the center point c and width parameter b of RBF neural network.The prediction results of the improved CS-RBFNN model are compared with those of the CS-RBFNN and PSO-RBFNN models.The results show that the RMSE values of the improved CS RBFNN model are reduced by 9.2%and 35.5%,respectively,with higher prediction accuracy.When the number of RBFNN hidden layer neurons increases,the prediction accuracy improves,but the convergence speed decreases and the running time becomes longer.
作者
徐熊飞
周晓华
杨艺兴
XU Xiongfei;ZHOU Xiaohua;YANG Yixing(School of Automation,Guangxi University of Science and Technology,Liuzhou 545616,China;Guangxi Key Laboratory of Automobile Components and Vehicle Technology(Guangxi University of Science and Tech‐nology),Liuzhou 545616,China;Dongfeng Liuzhou Automobile Co.,Ltd.,Liuzhou 545005,China)
出处
《广西科技大学学报》
CAS
2023年第4期111-116,共6页
Journal of Guangxi University of Science and Technology
基金
广西自然科学基金重点项目(2020GXNSFDA238011)
广东省基础与应用基础研究基金项目(2021B1515420003)资助。
