论文部分内容阅读
针对压力传感器在应用中存在温度漂移这一缺点,提出了一种基于蚁群聚类算法的RBF(Radial Basis Function)神经网络温度补偿方法。利用蚁群算法的并行寻优特征和一种自适应调整挥发系数的方法作为聚类算法来确定RBF神经网络的基函数的位置,并通过裁减的方法约简隐层的神经元达到简化网络结构的目的。通过仿真可以看出,该算法具有误差小,精度高等优点,对压力传感器的温度漂移有较好的补偿效果。
In view of the shortcoming of temperature drift of pressure sensor in application, this paper proposes a Radial Basis Function (RBF) neural network temperature compensation method based on ant colony clustering algorithm. Using the parallel optimization of ant colony algorithm and a method of adaptively adjusting the volatility coefficient as the clustering algorithm to determine the position of the basis function of the RBF neural network and reducing the neurons of the hidden layer through the reduction method to achieve a simplified network structure the goal of. It can be seen through the simulation that the algorithm has the advantages of small error and high accuracy, and has better compensation effect on the temperature drift of the pressure sensor.