Research On Optimal Scheduling Of Air Compressor Number

With the advent of industrial 4.0,manufacturing industry is becoming more and more automated and intelligent,compressed air as a very common energy medium in industry,the scope of applications will become more and more extensive.As a mechanical equipment for producing compressed air,the energy consumption of air compressor occupies more than 95%of the whole compressed air system,and there are still many unreasonable situations in the field use,resulting in a lot of energy waste.In order to reduce unnecessary energy consumption,how to reasonably schedule the operation of air compressor becomes more and more important.The more meaningful it is.This paper is based on the Dongfeng Nissan plant energy-saving transformation project as the background,the number of air compressor scheduling research,so as to reduce energy consumption waste,achieve energy-saving effect.There are two main parts in the design of air compressor number dispatching.The first part is to forecast the load at each time of air consumption,and the other part is to select the best combination of air compressors and the optimal output for the load forecasted at each time.Load forecasting is the basis for scheduling.Only accurate load forecasting can provide the unit with optimized combination and optimal gas production allocation.The designed number scheduling is more reasonable and safe.In this paper,the least squares support vector machine model is used to predict the load of the gas at each moment of the next day.In order to improve the accuracy and reliability of the prediction,the shortcomings of the manual selection parameters in the least squares support vector machine model are presented.The particle swarm optimization algorithm is introduced to automatically optimize the parameters in the model.Although the accuracy is improved,the particle swarm optimization algorithm has premature convergence,and the particle swarm optimization algorithm is improved.An improved particle swarm optimization algorithm is proposed to select the model parameters.The simulation results show that the improved particle swarm least squares support vector machine model improves the prediction accuracy and enhances the reliability,and has strong theoretical basis and practical value.By obtaining the results of load forecasting,under the safe and stable operation of the system,the gas load at each time is reasonably arranged for the unit combination and the optimal gas yield distribution of the unit.The purpose is to meet the field production of gas at the same time,make the energy consumption of the air compressor unit minimum,and achieve the effect of energy saving.After studying the working principle of air compressor and understanding the field constraints,the energy consumption model of air compressor unit is established.Then the Lagrange relaxation algorithm,the dynamic programming algorithm and the differential evolution algorithm are applied to solve the energy consumption model of air compressor unit.The simulation results of ten air compressor units are given.The number of air compressors can be scheduled one day ahead of schedule,which shows that the method adopted in this paper can effectively solve the number of air compressors scheduling problem.