基于数据驱动的多模型软传感器在一阶水箱中的应用

The Study and Application of Multi-model Soft Sensor Based on Data-driven on one-order Tank

针对一类非线性系统建立精确机理模型困难、仅用单一模型构建软传感器不甚可靠等问题,提出一种基于数据驱动的多模型传感器构建方法,该方法利用数据驱动技术建立一类非线性系统RBF神经网络、最小二乘支持向量机和核部分最小二乘3种预测模型,用系统多个预测模型的融合值代替传感器的输出构建软传感器;最后将所提出的方法应用于一阶水箱液位控制系统,实验结果表明多模型软传感器预测输出和实际系统响应基本重合,说明多模型软传感器能为非线性系统建立准确的模型,对复杂工业过程的建模有一定的普适性.

There are some problems in domain of fault detection of sensor and fault-tolerant control, such as difficult establishment of a precise mechanism model for a type of nonlinear system and the unreliability of the fault detection and fault tolerance with a single model. In order to solve those problems, this paper pro- poses an approach of multi-model soft sensor based on data-driven, and this paper establishes three type pre- diction models of RBF neural network, least square support vector machine and kernel partial least squares for the system using data-driven technology, and the fusion values of the multi-predictive-models for the system instead of the output of the sensor. And finally, the experiment applies the proposed method into the one-or- der tank liquid level control system. The experimental results show that the output of this multi-model soft sensor coincides with the actual output, which means multi-model soft sensor can effectively establish a pre- cise model for nonlinear system, and this model is universal for the complex industry process.