论文部分内容阅读
应变和温度的改变能够使光纤布拉格光栅(FBG)反射波的中心波长产生漂移,FBG与超磁致伸缩材料的结合可以用于测量电流,但是温度和应变的交叉敏感严重影响测量电流的精度。神经网络具有强大的非线性映射能力,能够自适应地发现传感器的内部规律,从而对温度进行有效补偿。针对神经网络容易陷入局部极小值的问题,采用遗传算法优化神经网络的权值和阈值,以在全局范围内更快速、准确地找到权值和阈值的最优解。针对样本较少的问题,采取K折交叉验证的方法提高网络预测的可靠性。经实验验证,优化的神经网络对电流预测的均方误差为0.0038,提高了FBG电流传感器的测量精度。
Strain and temperature changes can shift the center wavelength of FBG reflected waves. The combination of FBG and Giant Magnetostrictive Materials can be used to measure current, but the temperature and strain cross-sensitivity seriously affect the accuracy of the measurement current. Neural network has a strong ability of non-linear mapping, which can adaptively detect the internal laws of the sensor, so as to effectively compensate the temperature. Aiming at the problem that the neural network is apt to fall into the local minimum, the genetic algorithm is used to optimize the weights and threshold of the neural network to find the optimal solution of the weight and the threshold more quickly and accurately in the global scope. Aiming at the problem of fewer samples, the method of K-fold cross-validation is adopted to improve the reliability of network prediction. The experimental verification shows that the mean square error of the current prediction by the optimized neural network is 0.0038, which improves the measurement accuracy of the FBG current sensor.