论文部分内容阅读
温度漂移是影响光纤陀螺精度的重要因素之一。在对光纤陀螺温度漂移特性进行实验分析的基础上,对零偏温度漂移进行了多项式拟合补偿。为了解决传统曲面拟合方法无法精确描述标度因数温度漂移与温度、转速之间的关系导致其补偿精度低的问题,提出了一种基于自适应网络模糊推理的光纤陀螺温度漂移补偿新方法。该方法基于模糊逻辑,结合最小二乘和误差反向传播混合算法,设计了自适应网络模糊推理系统,从而有效提高了光纤陀螺温度漂移补偿精度。实验结果表明,在-30~60℃温度范围和-165~165(°)/s载体角速率范围,应用新方法对光纤陀螺温度漂移进行补偿,得到的训练误差均方根不超过0.003(°)/s,预测误差均方根不超过0.005(°)/s。
Temperature drift is one of the important factors affecting the accuracy of FOG. Based on the experimental analysis of the temperature drift characteristics of fiber optic gyroscope, the polynomial fitting compensation is applied to the temperature drift of zero bias. In order to solve the problem that the traditional surface fitting method can not accurately describe the relationship between the temperature drift of scale factor and temperature and rotational speed, the compensation accuracy is low. A new method of temperature drift compensation based on adaptive network fuzzy inference is proposed. This method based on fuzzy logic, combined with least square and error backpropagation hybrid algorithm, designed an adaptive network fuzzy inference system to effectively improve the temperature drift of FOG compensation accuracy. Experimental results show that a new method is used to compensate the temperature drift of fiber optic gyroscope at -30 ~ 60 ℃ and -165 ~ 165 (°) / s, and the root mean square error of training error is less than 0.003 ) / s, the root mean square of prediction error does not exceed 0.005 (°) / s.