Research On Energy Efficiency Improvement Strategy Of Integrated Energy System Based On Deep Learning

With the continuous development of the national economy,the acceleration of economic restructuring and the gradual increase in the demand for energy from the rapid development of society,the number and scale of integrated energy systems are constantly increasing and its energy efficiency control has become more and more significantly important.An integrated energy system is an energy system that uses coal,heat,and solar energy to meet different load requirements such as cold,heat,and electricity.The the large absorption of distributed renewable energy and the uncertainty of the demand side load have brought great challenges to the energy efficiency management and control of the integrated energy system.Through the development of energy big data,coupled with the development of cloud computing and other technological developments,the improvement of computer computing capabilities has made it possible for deep learning algorithms that have inherent advantages in processing large sample data to be applied to the energy efficiency,which can eliminate the cumbersome steps of data feature construction during using conventional predictive model.A comprehensive energy system energy efficiency improvement strategy based on deep learning is proposed in this circumstance.Deep learning algorithms can address the main problems existing in the current comprehensive energy system management.The uncertainty of demand-side cooling,heating,electric load and renewable energy supply in the energy system are predicted to provide a reference for making a scientific decision plan.This can improve the utilization of energy and eliminate waste in the integrated energy system.On the other hand,it can also help integrated energy companies to fully understand the operation of the integrated energy system and take countermeasures to save energy as much as possible.Eventually these companies can provide users with personalized and differentiated service,so as to further expand future comprehensive services.Therefore,this article firstly determines two indicators to evaluate the overall energy efficiency index of the integrated energy system,and makes models for different equipment in the integrated energy system are made.Secondly,a comprehensive energy system energy efficiency improvement strategy based on deep learning is proposed,combined with the photovoltaic output and multi-load data predicted by the deep learning framework.Finally,the particle swarm is reused to evaluate the proposed model and comparing with the EMA model and the SVM model in terms of energy efficiency improvement.The energy efficiency improvement model of the integrated energy system based on deep learning proposed in this paper compensated the weak point of conventional static models.The model uses the accurate prediction characteristics of deep learning,and integrates the multi-load curve on the demand side and the photovoltaic output on the energy supply side.These curves are used to correct parameters in the existing energy efficiency optimization model.Meanwhile,the corresponding weather conditions descriptions are transformed into digital vectors as input variables of the deep learning model.The experimental results show that model has fabulous advantages in training time and prediction,and has a significant role in promoting the energy efficiency of the integrated energy system.The final results show that the ability of automatic feature extraction of big data samples based on big data deep learning models makes up for the shortcomings of traditional predictive models.Experiments have proved that the deep learning model has certain application prospects in comprehensive energy efficiency management.Through accurate prediction of uncertain parameters in the comprehensive energy system,it can effectively promote the overall energy efficiency,which means introducing the deep learning model to predict the parameters with lower stability in the optimized operation model will promote the improvement of the whole energy efficiency of the integrated energy system to a certain extent.