摘要
针对电厂过热汽温控制中存在强非线性和大时滞的特点,利用基于径向基函数的最小二乘支持向量机方法进行建模。由最小二乘支持向量机建立被控对象的模型,并在保证模型逼近性能的前提下,使用修剪算法得到具有稀疏性的SVM模型,该算法的优点是训练过程遵循结构风险最小化原则,不易发生过拟合现象。应用某电厂超临界600MW直流锅炉高温过热系统进行仿真,结果表明该模型可以较好地适应非线性和较大时滞特性的变化。
Aiming at the strong nonlinearity and large time-varying characteristics in controlling of super-heater temperature in plant, the method of LS-SVMs based on radial basis function are used to model. Under the condition of modeling approximating to perfor-mance, the sparse modeling is gotten by the pruning algorithm. The merits of the algorithm are conforming to the least structural risk in training process and hardly leading to over-fitting. The simulation of a superheating system, in one supercritical concurrent 600MW boiler in one power plant, is taken. The result shows that the controlling system can be adapt to the variation of the object characteristic well with strong nonlinearity and large time-varying characteristics rapidly.
出处
《微计算机信息》
北大核心
2007年第10期270-272,289,共4页
Control & Automation
基金
国家自然科学基金项目(60604023)
