Energy Consumption Monitoring System For Heat Source Plant Based On Cloud Platform

As people’s demand for heating increases,the number of heat source plants across the country is increasing.During the operation of heat source plants,a large amount of harmful gases are emitted,causing serious air pollution and causing a great impact on people’s health.The problem of how to achieve effective reduction of polluting gas emissions in heat source plants while safeguarding people’s demand for heat supply needs to be addressed urgently.In this thesis,we design a cloud-based energy consumption monitoring system to achieve real-time monitoring of energy consumption data of heat source plants and sho rt-term prediction of power load of heat source plants in combination with power load prediction models,so as to improve the energy utilisation of heat source plants for the purpose of energy saving and emission reduction.This paper focuses on the design of the system architecture,the research of the power load forecasting method,the design of the data collector and the development of the system software.For the architecture design of the system,the system is divided into field equipment layer,network communication layer,server layer and application layer by analysing the actual functional and performance requirements of the heat source plant.For the data transmission between each layer,a combination of wired and wireless data transmission is used;for the server layer,Aliyun is selected as the cloud server of the system;for the data storage,a data management system is used to save the collected information;for the software architecture of the application layer,a browser/server structure is used.Power load prediction is an extremely important function in the energy consumption monitoring system of a heat source plant,and a research on power load prediction methods is conducted for this function.The key factors influencing the power load of the heat source plant are determined by combining the actual situation of the heat source plant,and a two-way long and short-term memory neural network algorithm,which can handle two-way information flow,is selected as the basis,and the attention mechanism is used to improve it,so as to obtain a two-way long and shortterm memory neural network based on the attention mechanism.Finally,the historical load data of the heat source plant is used to verify the load prediction performance of the method.Designing a dedicated data collector for the energy consumption monitoring system of a heat source plant to centralise the collection of numerous devices in the heat source plant.The functional and performance requirements of the data collector were analysed in relation to the real situation of the heat source plant and it was determined that the data collector consisted of a main circuit module,a power module,a storage module and a communication module,of which the communication module included an RS-485 module,a Lo Ra module,a 4G module and an Ethernet module.Suitable devices are selected and schematic design is carried out for each module’s requirements.For the software module of the energy consumption monitoring system of the heat source plant,the acquisition function of the data collector and the remote communication function are designed and the call to each function is realised through the main program.By building the data management system separately in the cloud platform,the classification and storage of the uploaded data is realised.Real-time monitoring of the energy consumption data of the heat source plant is realised by developing the front-end interface and the back-end service program in the cloud platform using the relevant software,com bined with the database.The function of electricity load forecasting is also realised by calling load forecasting models in the development software.