【摘要】 由于受到传感器本身的特性以及迟滞、蠕变、温度、湿度等外界因素的影响,传感器的被测参量与传感器输出之间的关系一般是非线性的,需要对其输出特性进行拟合或非线性补偿。介绍了传感器输出特性拟合原理、现代传感器拟合的方法,包括基于切比雪夫算法的传感器输出特性拟合,以及基于径向基、BP、切比雪夫、傅里叶基函数神经网络的传感器特性拟合,讨论了粒子群、遗传算法对传感器输出特性拟合的优化,对存在的问题及未来的研究方向进行了分析和展望。更多还原

【Abstract】 Due to the characteristics of sensor itself,as well as affected by hysteresis,creep,temperature,humidity and other external factors,The relationship between measured parameters and sensor output is nonlinear in general,Its output characteristics need to be fitted or non-linear compensated.The sensor output characteristics fitting principle and modern sensor fitting method are introduced,Including sensor output characteristic fitting based on chebyshev algorithm and sensor characteristic fitting based on RBF,BP,chebyshev,fourier basis functions NN,Optimizing of sensor output characteristics fitting with PSO and genetic algorithm are discussed,And analysis and outlook on the problems and future research directions.更多还原