Research On Energy Saving And Operation Strategy Of Industrial Air Conditioning Based On Neural Network

Energy conservation and emission reduction is an important part of enterprise management,and energy costs directly affect the economic benefits of enterprises.The issue of energy management and control of industrial air conditioners has long been the focus of enterprises.This paper collects and cleans the historical data of enterprise air-conditioning operation(Climate,capacity),and establishes a scientific air-conditioning energy consumption model and air-conditioning startup time prediction model.The theory and technology proposed in this paper are applied in practice,and the intelligent factory air-conditioning energy management and control system is designed and implemented to realize the startup time prediction of the air-conditioning in the production workshop.The results show that the system can help companies achieve the goal of energy smart control.The research work is mainly as follows:1)Establishment of a comprehensive industrial air conditioning energy consumption model for enterprises to produce the whole environment.Through the existing 98 temperature and humidity collection points in the workshop,comprehensive and no-dead data collection in the production workshop,and use OPC technology to parse and persist data to the database.Based on the historical data collected from the operation of the air-conditioning equipment,an energy consumption model is established for the main parts of the air-conditioning system(chiller,cooling water pump,chilled water pump,cooling tower,fan coil,air-conditioning unit).Taking the overall energy saving of enterprise air-conditioning system as the research goal,the energy efficiency ratio of all components is taken as the objective function of energy-saving optimization,and the universe energy consumption model of the workshop air-conditioning system is finally established.2)Establishment of a prediction model for air conditioning start-up time based on neural network.Aiming at the instability of traditional prediction methods and the problem of gradient disappearance in RNN neural network,combined with the characteristics of air-conditioning running time series data,the Long Short-Term Memory neural network is used to establish the air conditioning start-time prediction model based on production requirements and environmental comfort.The air conditioning start-time prediction model established in this paper is compared with the prediction model established by Random Forest and Support Vector Regression.The experimental results show that the model established in this paper shows the stability superior to the other two,and the optimized air conditioning energy consumption reduced by approximately 27.9%.3)Establishment of a prediction model for air conditioning start-up time based on IASFA-LSTM.In order to improve the prediction accuracy of Long Short-Term Memory network models,an artificial fish swarm algorithm with improved visual field and step size is proposed based on the existing artificial fish swarms algorithm.The improved algorithm is used to optimize the parameters of LSTM network.A Long Short-Term Memory network prediction model based on improved artificial fish swarm algorithm is constructed.At the same time,the datasets of four different dimensions are selected to verify the model,and the model is applied to the prediction of air conditioning startup time.The results show that the improved model prediction accuracy has been improved,and have a good predictive effect on air conditioning start-up time.4)Design and implementation of intelligent factory air conditioning energy management and control system.Based on the above research results,build a corporate air conditioning energy management and control platform,these include automatic start-up and energy consumption prediction modules for air conditioners.Starting from the actual needs of the enterprise,the functions of air-conditioning information inquiry,prediction model training,and prediction result query are implemented on the platform and applied in production.At the same time provide air conditioning energy consumption panoramic view board,so that enterprises can transform from traditional experience to precision,refinement and scientificization in air-conditioning energy management and control,reducing the energy costs required for production.