水泥篦冷机出口熟料温度自适应辨识模型

为了优化控制系统,建立篦冷机温度熟料出口的识别模型,利用篦冷机内熟料换热机理,找出熟料冷却过程的关键影响因素;利用回声状态网络辨识篦冷机运行数据,基于递归最小二乘法推导网络的在线学习算法,实现权值自适应调整,从而建立了篦冷机出口熟料温度的自适应辨识模型.仿真实验可知,在系统发生变化时构建的模型能够自适应调整网络输出权值矩阵,使模型快速收敛.与其他离线方法相比,提出的熟料出口温度的自适应模型更加持久有效,可以作为辨识模型指导篦冷机的控制.

Adaptive identification model for export clinker temperature in cement grate cooler

In order to optimize the heat recovery efficiency of the cement cooler, the heat exchange mechanism of the clinker in the grate cooler was studied, the key factors affecting the clinker cooling process were found and the operating data of grate cooler is trained by echo state network (ESN). In order to solve the mismatch problem of off-line identification model, t a network online learning algorithm was deduced based on the recursive least squares, an adaptive model of the clinker temperature in the grate cooler was built. Base on that, adaptive identification approach and process for the export of clinker temperature is given. Simulation experiments show that the model can adjust the output weight matrix of the network adaptively when the system changes, so that the model converges quickly. Compared with other off-line methods, the RLS-ESN self-adaptive model of the clinker temperature of the grate cooler can ensure prolonged and effective. The completion of these basic work can guides grate cooler control and lays the foundation for the optimal control and energy conservation and emission reduction of grate coolers.