基于最小二乘支持向量机的复合肥装置养分含量的软测量建模

针对复合肥装置养分含量无法用常规的传感器在线测量的问题,提出了基于最小二乘支持向量机(LS-SVM)的软测量方法来在线估计养分含量.LS-SVM用等式约束代替传统的标准支持向量机中的不等式约束,求解过程从解二次规划问题变成解线性方程组,求解速度相对加快.工业实例表明LS-SVM所建模型的预测精度较高,能满足实际工业应用的需求.

Least Squares Support Vector Machines Based Soft Sensor Modeling of Nutrient Concentration for Multiplex Fertilizer Device

According to the problem that the nutrient(N,P_2O_5 and K_2O)concentrations can not be measured accurately on-line,soft sensors based on least squares support vector machines(LS-SVM)are proposed to estimate the nutrient concentrations.Due to equality type constraints in the formulation of the LS-SVM,the solution follows from solving a set of linear equations,instead of quadratic programming for classical SVM's,as a result,the calculation of the LS-SVM is faster than the SVM's.The simulation results by use of real industrial data show that the soft sensors based on LS-SVM is accurate in prediction,so they are suitable to practical application.