摘要
针对最小二乘等传统方法的不足 ,提出了解决传感器系统非线性校正问题的遗传神经网络算法。计算机仿真表明 ,采用这里所提出的观点和方法 ,其拟合精度远高于传统上应用最为广泛的最小二乘法 ,并且具有较强的鲁棒性。而且 ,从所做的研究显示 。
Genetic neural network model of solving the problems on the non linearity rectification of sensor systems,is put forward,for the shortcoming of least square and other conventional methods.Computer simulations are given to demonstrate that approximation accuracy of the model is far higher than least square methods that are extensively applied conventionally and the model possesses stronger robustness,through adopting the standpoints and methods in this paper.And the research indicates that the model can be also used to realize the non linearity rectification in other similar systems.
出处
《仪器仪表学报》
EI
CAS
CSCD
北大核心
2003年第2期201-204,214,共5页
Chinese Journal of Scientific Instrument
基金
国家 8 63高技术项目 ( 2 0 0 2 AA742 0 48)
关键词
传感器
遗传算法
BP神经网络
非线性校正
精度
Sensor Genetic algorithm BP neural network Non linearity rectification Accuracy