基于神经网络模型的冷库冷风机除霜控制研究

Research on Defrost Control of Cold Storage Cooling Fan Based on Neural Network Model

冷库冷风机“按需除霜”,可有效降低冷库能耗、提高能源利用率。本文将湿空气物性参数、冷风机运行时间作为神经网络输入变量,建立基于BP算法训练的多层前馈神经网络结霜量与除霜时长预测模型,并利用相关试验数据进行模型训练与测试。结果表明:结霜量预测模型计算值与试验测量值平均误差为10.11%,除霜预测模型计算值与试验测量值的误差均小于5%。本文所建立的基于人工神经网络结霜量预测模型与除霜时长预测模型可较好地预测冷风机结霜量与除霜时长,为实际工程应用中通过确定除霜起始点和除霜时长实现冷风机“按需除霜”提供了参考价值。

The“defrosting on demand”of cold storage chillers can effectively reduce the energy consumption of cold storage and improve energy efficiency.The wet air physical parameters and the cooling fan running time were taken as the neural network input variables,and the frosting amount of the multi-layer feedforward neural network trained based on BP algorithm was used to establish the frosting amount and defrosting duration prediction model.The experimental data was used to train and test the model.The results show that the average error between the calculated value of the frosting prediction model and the experimental measurement value was 10.11%,and the average error between the calculated value of the defrosting prediction model and the experimental measurement value was less than 5%.The artificial neural network frosting prediction model and the defrosting duration prediction model established can well predict the frosting amount and defrosting time of the chiller,and realize the defrosting starting point and the defrosting time in practical engineering applications.The“Defrost on Demand”of the air cooler provides a reference value.