Study on Application of LMBP Neural Network PID Controller in HVAC System

王彦,刘宏立,杨珂
2010-01-01
Abstract:To address the defects of slow convergence and easy immergence in the partial minimum in the learning process of Back Propagation(BP) Neural Network(NN),a Levenberg-Marquardt(LM) algorithm was presented.In order to solve the two problems which seriously affect the training time and the converging accuracy in the choice of learning rate and the inverse matrix solution in LM algorithm,LU decomposition was proposed to improve the LM algorithm simulated with MATLAB.The LMBP Neural Network PID controller was applied in the cooling water cycle of heating,ventilating and air conditioning(HVAC) system.The results of LMBP neural network PID controller,PID controller,and BP neural network PID controller were simulated and compared.The performance has indicated that the BP Neural Network PID controller optimized with the improved LM algorithm is better than other algorithms in reducing the overshoot,speeding up the convergence rate,and reducing the steady-state error.
What problem does this paper attempt to address?