Central Heating And Heat Load Forecasting And Management System Research And Implementation

The central heating system is an important infrastructure for people’s livelihood in northern China,and its operating conditions are directly related to the heating quality of users and the economic benefits of heating enterprises.The informatization and intelligentization of centralized heating management work have a profound impact on improving the efficiency and quality of management in the heating industry,and on environmental protection,energy conservation and emission reduction.This paper focuses on how to achieve more accurate load forecasting for central heating,intelligent and information management of heating management,and combined with the actual needs of heating company management,we developed Management System ".The main research content and results are as follows:(1)Sorted out and optimized the business processes of the heating company,and constructed business demand analysis models such as organizational structure and core business processes.The requirements of the central heating management system are analyzed,and the system use-case model is constructed.The overall system design is carried out on the basis of requirements analysis,and the system architecture design,database design and system function module design results are given.(2)Aiming at the problems of central heating,the existing heat load prediction method relies on artificial experience,low prediction accuracy and easy to fall into local minimum,etc.,a heat load prediction model based on GA-BP neural network is designed,and here Based on this,a prediction method combining the genetic algorithm with global search capability and the traditional BP neural network is proposed.After the analysis of the factors influencing the heat load,the five factors influencing the average daily outdoor temperature,working day type,average daily outdoor wind speed,sunshine time and the first three days of daily heat load were determined as input variables,and the heat load of the day Output variables.After optimizing the initial weights and thresholds of the BP neural network using genetic algorithm,a heat load prediction model is trained.Compare and analyze the prediction effect of GA-BP neural network and BP neural network.The experimental results show that the error of the prediction value obtained by the prediction method proposed in this paper is between-15%and 10%,which is more effective than the current prediction method for central heating.(3)Adopt object-oriented,unified modeling language and Java Web development methods and technologies,and developed the "central heating and heat exchange station temperature adjustment and management system".The system has functions such as customer management,charge management,heat exchange station monitoring,inspection management,measurement management,and equipment information statistics.The "central heating heat load forecasting and management system" developed in this paper has been tested and operated.The operation results show that the system meets the actual needs of the central heating system management,and can accurately predict the central heating load to make the heating operation.The state of satisfaction is of great significance to the safety of heating,stable operation,energy saving and emission reduction,and environmental protection.