摘要
为改善锅炉吹灰的盲目性,基于改进的径向基函数神经网络递归正交最小二乘算法构建了工厂锅炉积灰预测的非线性模型,用来预测运行过程中锅炉各个受热面积灰程度。根据电厂锅炉实际运行情况,确定了多个特征变量来决定锅炉的工作状况。Matlab仿真实验证明通过所建立的非线性模型,能有效预测出锅炉工作时受热面清洁状况下的吸热量,因此能够实时的反映受热面的积灰程度。
To improve the blindness of the boiler soot blowing,the nonlinear model of boiler ash fouling prediction based on improvement of radial basis function neural network（ RBF NN） recursive orthogonal least squares（ ROLS） algorithm was presented to predict the extent of the boiler heating area during operation.According to the actual operation of power plant boiler,multiple characteristic variables were determined to decide the boiler operating conditions. The Matlab simulation experiment proves the presented model can predict the heat absorption of the heat surface cleaning effectively,which can reflect the degree of the ash fouling of the heating surface in real time.
出处
《哈尔滨理工大学学报》
CAS
北大核心
2017年第5期30-34,共5页
Journal of Harbin University of Science and Technology
基金
黑龙江省教育厅科学技术研究项目(12533028)
关键词
神经网络
径向基函数
递归正交最小二乘算法
锅炉积灰
neural network
radial basis function
recursive orthogonal least squares
boiler ash fouling
