基于沙猫群优化算法和BP模型的光纤陀螺温度补偿研究  

A Fiber Optic Gyroscope Temperature Compensation Method Based on SCSO-BP Model

作　　者：张志利 刘瑾 周召发 李洪才 梁哲 ZHANG Zhili;LIU Jin;ZHOU Zhaofa;LI Hongcai;LIANG Zhe(Rocket Force University of Engineering,Xi’an 710025,Shaanxi)

机构地区：[1]火箭军工程大学,陕西西安710025

出　　处：《火箭军工程大学学报》2024年第4期54-60,67,共8页Journal of Rocket Force University of Engineering

基　　金：航空科学基金(201808U8004)。

摘　　要：为有效补偿光纤陀螺输出精度的温度误差,分析了光纤陀螺温度误差产生机理,基于沙猫群优化算法(Sand Cat Swarm Optimization,SCSO)优化后的反向传播(Back Propagation,BP)神经网络,建立了SCSO-BP温度补偿模型,并对某高精度光纤陀螺进行了温度补偿实验。实验结果表明:在-40~70℃环境下,该方法补偿后的光纤陀螺温度漂移相较于补偿前减小了近95%,相较于BP神经网络补偿算法减小了86%左右,相较于蜣螂优化算法优化后的BP温度补偿模型减小了近58%;该模型在对新鲜样本的补偿过程中表现出了较为优越的泛化性能。To effectively compensate temperature-induced errors of the fiber optic gyroscope(FOG)output accuracy,the cause of these errors were analyzed firstly.Accordingly,a temperature compensation model,named the SCSO-BP model,was established based on a back propagation(BP)neural network optimized by Sand Cat swarm optimization(SCSO)algorithm.Moreover,temperature compensation experiments were conducted on a high-precision FOG.Experimental results showed that under a temperature between-40℃and 70℃,the temperature drift of the FOG compensated by the proposed method was reduced by nearly 95%compared to that of the FOG without compensation,while approximately 86%compared to that of the FOG compensated by BP neural network compensation algorithm,and nearly 58%compared to that of the FOG comp ensated the Dung Beetle optimizer(DBO)-BP model.Furthermore,the model demonstrated superior generalization performance when compensating for new samples.

关 键 词：光纤陀螺 温度补偿 BP神经网络 沙猫群优化算法 零偏 

分 类 号：TN253[电子电信—物理电子学] TN29