论文部分内容阅读
光纤传感器受到多种因素影响,测量误差具有非线性,当前方法难以对测量误差进行准确补偿,导致光纤传感器的测量精度低。为了降低光纤传感器的测量误差,提出了粒子群算法和神经网络的光纤传感器测量误差补偿模(PSO-BPNN)。首先通过实验获得光纤传感器测量误差的样本,然后采用神经网络对光纤传感器测量误差变化特性进行拟合,对光纤传感器测量误差进行补偿,并引入粒子群算法对模型进行优化,最后实现光纤传感器测量误差补偿仿真实验。PSO-BPNN构建了良好的光纤传感器测量误差预测模型,根据预测值对测量值进行补偿,使得测量精度超过95%。
Optical fiber sensor is affected by many factors, the measurement error is nonlinear, the current method is difficult to accurately compensate for the measurement error, resulting in low accuracy fiber optic sensor. In order to reduce the measurement error of optical fiber sensor, a PSO-BPNN (Optical Fiber Sensor Measurement Error Compensation Model) based on PSO and neural network is proposed. Firstly, the samples of optical fiber sensor’s measurement error were obtained through experiments. Then, the neural network was used to fit the measurement error variation of optical fiber sensor, and the error of optical fiber sensor was compensated. Particle swarm optimization algorithm was introduced to optimize the model. Finally, Compensation simulation experiment. PSO-BPNN has constructed a good prediction model of optical fiber sensor measurement error, and compensates the measurement value according to the predicted value, so that the measurement accuracy exceeds 95%.