Research And Application Of Steam Intelligent Supply Strategy Optimization In Large Manufacturing Enterprises

In the energy management and control of large manufacturing enterprises,the steam supply strategy management module plays an important role.Both the production process and industrial air conditioning require steam energy,and the related energy costs directly affect the economic benefits of the enterprise.At present,the steam supply management and control of large manufacturing enterprises often adopts a fixed preheating start time and a charging boiler start strategy,which lacks scientificity and energy saving.Therefore,there is an urgent need to increase production factors to achieve the "precision supply and demand" of steam in the production process of manufacturing enterprises and reduce the waste of energy in the production of enterprises.In this paper,through the collection and processing analysis of industrial production data,combined with existing production experience,the process production and industrial air conditioning steam flow are predicted and modeled respectively,and the two prediction curves are superimposed over time to obtain a complete steam supply prediction curve,The scheduling optimization problem of the steam supply system is reduced to the mixed integer linear programming problem to be solved,and the best strategy of steam supply is obtained.The specific research contents are as follows:1)Industrial data collection and processing.That is to complete the collection of process and energy management and control data,production plan and energy consumption related data,and steam and industrial air conditioning environment related data.Through the docking with the existing data collection platform,direct database connection,Web interface and other collection methods to complete the collection of enterprise massive data,and establish a stream computing processing architecture and batch processing overall architecture.For real-time calculation and analysis of data and monitoring results,by using a distributed Redis cluster to store data key/value to meet its high real-time requirements.For the data with low access frequency and offline analysis,distributed file database Hbase is used to store it to ensure the reliability and expandability of data storage.2)Optimal energy prediction that matches the production plan and process flow to provide steam with accurate energy supply on demand.According to the process production situation and the timeliness of the original data,a point-by-point regression-curve completion prediction method is proposed to realize the steam flow prediction of the process production.Case studies show that compared with the traditional single-prediction model of undivided time set,the point-by-point regression-curve completion combined prediction method based on time set segmentation not only has high prediction accuracy and is more stable,but also provides steam production and real-time scheduling Basis for decision.3)Air conditioning steam flow prediction based on chaotic time series data.In view of the fact that industrial air conditioning time series data has many influencing factors and has strong chaotic characteristics,an air conditioning steam flow prediction method based on chaotic time series data is used.Analyze the chaotic characteristics of the air-conditioning steam flow time series data,determine the chaotic characteristic parameters and select the support vector regression machine for prediction.Case studies show that this method improves the prediction accuracy and provides a scientific basis for subsequent steam strategy optimization.4)Research on steam supply optimization strategy based on flow forecast.The steam energy supplied by the enterprise's boilers is mainly used in process production and air conditioning.Therefore,based on the process flow and industrial air conditioner's predicted flow superposition curve,the predicted curve of the total steam supply end flow in the production process can be obtained and is consistent with the company's existing experience supply strategy Combined,a mixed integer linear programming mathematical model is constructed to solve it,and the optimal control parameter,that is,the optimal control strategy under the premise of satisfying the steam flow required for production is obtained.Practical application shows that this scheme can not only realize the precise supply and demand of steam energy,but also effectively reduce the production cost of enterprises.5)Build a comprehensive management and control platform,and apply the steam supply strategy optimization model based on flow forecast to actual enterprise production.Starting from the actual needs of the enterprise,the operation management module,early warning strategy module,clustering and correlation analysis module,etc.are implemented on the platform and applied in production,so that the steam supply of manufacturing enterprises from traditional experience to precision,Scientific,reduce the cost of energy required for production.