基于改进BP神经网络的风冷热泵除霜智能控制方法研究

Research on intelligent control method ofair-cooled heat pump defrosting based on improved BP neural network

当前的智能除霜过程没有考虑机组本身对除霜的影响,造成以BP神经网络为主的控制方法能耗高,效率低。该文提出改进BP神经网络控制风冷热泵除霜控制方法。结合质量守恒定理,明确风冷热泵中湿空气和风机翅片管热器间的换热关系,通过计算控制单元进口和出口处含湿量差值得到风冷热泵的结霜量;使用北方苍鹰优化算法改进BP神经网络,将空气侧换热器结霜相关参数和机组系统对结霜的影响参数作为网络输入,基于结霜量输出换热器除霜开始时间和加热结束时间,实现风冷热泵除霜智能控制。实验结果表明,所提控制方法可在170 s以内解除冻霜,并将结霜过程中的运行压力波动维持在0.3 MPa,降低能源消耗的同时加快了除霜效率,确保空调设备能高性能平稳运行。

The current intelligent defrosting process does not consider the impact of the unit itself on defrosting,resulting in high energy consumption and low efficiency of the control method mainly based on BP neural network.This article proposes an improved BP neural network control method for defrosting of air-cooled heat pumps.Based on the principle of conservation of mass,clarify the heat transfer relationship between moist air and the finned tube heat exchanger in the air-cooled heat pump.Calculate the frost formation of the air-cooled heat pump by calculating the difference in moisture content between the inlet and outlet of the control unit.Using the Northern Eagle Optimization Algorithm to improve the BP neural network,the frosting related parameters of the air side heat exchanger and the influence parameters of the unit system on frosting are taken as network inputs.Based on the amount of frosting,the defrosting start time and heating end time of the heat exchanger are output,achieving intelligent control of defrosting for air-cooled heat pumps.The experimental results show that the proposed control method can remove frost within 170 seconds and maintain the operating pressure fluctuation during the frosting process at 0.3 MPa,reducing energy consumption while accelerating defrosting efficiency,ensuring high-performance and stable operation of air conditioning equipment.