基于神经网络的FCD絮凝控制系统的研究

Research on FCD Coagulation Control System Based on Neural Network

神经网络具有很强的自学习和自适应能力,对于大滞后、非线性的复杂系统有较好的控制效果。针对水处理过程中系统环境不断变化,特别是流量有较大、频繁变化的情况下,在FCD絮凝控制系统的基础上,提出了神经网络控制方法,解决了混凝控制中的目标等效直径动态确定问题。实际运行表明,该方法取得了很好的控制效果,具有广阔的应用前景。

Neural network is of good ability of self-learning and self-adapting, therefore, it has better control effectness for nonlinear large time-delay complex system. By analyzing the different situation in water supply, especially the sudden change of flow, raises the method of neural network based on flocculation image manipulation, resolve the problem that the target-value of equivalent diameter is always uncertain. The application shows that the method can obtain satisfactory control, and has broad application foreground.