基于GA-BP神经网络的光纤陀螺温度补偿

摘要

The temperature characteristic of the scale factor and the zero bias is the main factor that affects the performance of FOG,a temperature compensation model based on BP neural network was designed to improve the precision of the FOG about the scale fact and the zero offset.Aiming at the deficiency of BP neural network,the temperature compensation model of GA-BP neural network was established by the genetic algorithm optimizing the structure of BP network.The optimized network model was validated by using the test data of the scaling factor and the zero bias at different temperature environment.The experimental re-sults show that the compensation effect of the optimized network model is greatly improved,the accuracy of the compensation er-ror is improved by one level.%标度因数和零偏的温度特性是影响光纤陀螺工作性能的主要因素,为提高光纤陀螺仪的输出精度,分别建立了基于BP神经网络的标度因数和零偏的温度补偿模型.在此基础上提出了利用遗传算法优化网络参数来弥补BP神经网络算法所存在的不足,最终建立了GA-BP神经网络温度补偿模型.使用在不同温度下的标度因数和零偏测试数据对改进后的神经网络补偿模型进行验证并与原网络模型进行对比分析,实验结果表明,优化过的模型其补偿效果具有很大的提高,其补偿的误差精度提高了一个级别.