The performance of soft sensor based on neural network can be affected significantly because of the problems. How to solve those problems is proposed.
这些问题具有一定的普遍性,若不能很好地解决,将直接影响软测量的性能。
On the basis of analysis of several methods for modeling, a soft sensor based on kernel principal component analysis (KPCA) and least square support vector machine (LSSVM) is proposed.
在具体分析了多种建模方法的基础上,提出了核主元分析结合最小二乘支持向量机软测量建模方法。
The reason that the residual test method based on Kalman filter is insensitive to sensor soft fault detection is analyzed.
分析了基于卡尔曼滤波器的残差检验法对传感器缓变故障检测的不敏感性原因。
应用推荐