Research On The Control System Of Absorption Refrigeration Unit

In China,the energy status is increasingly severe,and resource utilization efficiency is low,and the waste heat resources are not effectively utilized.The absorption refrigeration unit can operate with waste heat resources,but when the absorption refrigeration unit is operated under partial load,the COP of the unit becomes lower,and there is a large waste of resources.Moreover,the absorption refrigeration unit has strong coupling,large time delay and difficulty in establishing accurate mathematical model.In response to the above questions,this paper establishes a BP neural network steady state model,and uses improved particle swarm optimization algorithm to find the optimal set point under partial load,in order to improve the COP of the unit.The obtained set point can be used to control the absorption refrigeration unit by using the improved model-free adaptive control algorithm.The main research contents and results are as follows:Firstly,the experimental platform of single-effect lithium bromide absorption refrigeration unit was built.In the case of different loads,the experimental platform is used to obtain steady state data of the unit.Based on the experimental data,the BP neural network steady state model of 6-9-2 was established.The established steady-state model is used for sensitivity analysis to obtain the set points that need to be optimized.Secondly,aiming at the problem that the COP of refrigeration unit is not high under partial load,an optimization algorithm combining inverse neural network algorithm and particle swarm optimization algorithm is proposed.According to this algorithm,find the optimal set point of the unit operation under partial load.Experimental data was used to verify the validity of the proposed method.By comparing the optimization set points before and after,it can be seen that the COP of the unit is significantly improved under partial load.Finally,aiming at the large time delay and difficult modeling of the refrigeration unit,a model-free adaptive control method is proposed based on the second-order universal model with lag time input variation constraint.Considering the input change of the previous time affects the output change,the universal model is upgraded from first order to second order.And further derive the second-order universal model of double-input and double-output.And then the lag time constant is introduced into the control input criterion function and the estimation criterion function to derive the two-loop second-order universal model.The control scheme is verified by simulation,and the simulation results show that the improved model-free adaptive control algorithm in the refrigeration unit has good control performance.