遗传寻优神经网络在平台温控系统建模中的应用

温度直接影响惯性仪表及惯性平台的使用精度,而高精度温控系统的设计依赖于准确的平台加温模型,针对平台系统中多种惯性仪表加温过程复杂度高,当前采用的阶跃响应辨识方法存在模型适应性差、精度不高等情况,且针对基于梯度下降的BP学习算法存在局部收敛的问题.采用基于遗传算法寻优的神经网络辨识的方法,对惯性仪表加温模型进行建模,试验验证通过遗传寻优后的BP神经网络学习算法,提高了网络的学习精度,进而提高了平台系统中惯性仪表加温过程数学模型的精度,模型适应性较高,为后续惯性仪表的加温控制方法的设计提供了必要的条件.

Application of Genetic Optimization Neural Network in Modeling of Platform Temperature Control System

Temperature directly affects the accuracy of inertial instrument and inertial platform, the design of high precision temperature control system depends on the accurate platform heating model. In this paper, due to the high complexity in heating process of a variety of inertial instruments in the platform system, the current step response identification methods have poor adaptability model and accuracy is not high. At the same time, BP learning algorithm based on gradient descent has the problem of local convergence. A neural network identification method based on genetic algorithm is used to model the heating model of inertial instrument. The experiment prove the BP neural network based on genetic algorithm improves the learning accuracy of the network. The accuracy of the mathematical model of inertial instrument heating process in the platform system is improved. It provides the necessary conditions for the design of the heating control method for the following inertial instruments