Research And Application Of On-demand Precise Regulation Technology Of Urban Heating System Based On Digital Twin

The shift towards clean,safe,low-carbon,and efficient energy is causing urban heating systems to move towards multiple heat sources and complementary operations.This makes heat production more complex and dynamic.As a result,the heat supply system needs to be operated and controlled in a safe,reliable,energy-efficient,environmentally friendly,and balanced manner,facing several new challenges.It’s mostly shown in the following ways: On the supply side of the heating system,the actively controlled heat source and the passive and variable heat source work together in real time to maintain the heating load.This requires coordination with the specific heating parameters.On the demand side,the heat load fluctuates due to outdoor weather,building structure,and heating methods.Each demand-side heat load varies to a certain degree.On the demand side,the heat load exhibits a degree of volatility influenced by outdoor weather,building structure,and heat supply method,among other factors.The actual demand loads of different endpoints are characterized by differentiation.To achieve on-demand heat supply,it is imperative to integrate the inherent conditions and characteristics of the heat to anticipate the heat loaThe old-fashioned way of operating and controlling heating systems,which combines artificial experience and automated systems,makes it hard to understand the real-time network operation status.The control objectives are focused on individual parameters,which may not work together,creating uneven heating.This traditional route fails to meet the complex and diverse heating needs of China’s urban heating system.In this paper,we use digital tools to connect the virtual digital twin model to the physical world and create a coordinated control strategy for the "source-networkload" heating system.This approach allows for accurate and on-demand regulation of urban heating in different times and locations.The main work and innovations include:(i)Starting from the level of integrating information and physical components,this text explains the essential elements in managing a heating system using digital twin technology.Based on engineering practices,it explores the use of model-based system engineering methods in heating systems and proposes the technical architecture of an intelligent heating system using digital twins,divided into three layers: the heating equipment layer,monitoring and control layer,and intelligent decision-making layer.With new information technology and digital twin technology,Urban heating systems can be monitored using a virtual model.This model integrates "state sensing and monitoring-autonomous analysis and decision-making-demand adjustment and control" to facilitate quantitative and refined management.(ii)Aiming at the problems of load prediction in heat supply system and temperature transmission delay in heat network,this paper proposes a dynamic supply and demand balance analysis model of heat supply system taking into account the hysteresis characteristics.First,using graph theory,model the structure mechanism of the heat supply system.Identify and correct resistance coefficients for the pipe network and pump valve characteristics through system identification methods.Then,construct a digital twin model based on the heat supply system’s structural mechanism model.Using the machine learning method for predicting heat station demand,the digital twin model is established.This model takes into consideration the sliding time window and calculating the average value for predicting temperature delay response time.The result is an accurate prediction of demand load and temperature response time for the station.At the same time,we predict the heat station’s demand heat load using machine learning.We also predict the heat station’s temperature delay response using the sliding time window and mean value calculation method.Then,we establish the predictive regulation process of the heat station based on digital twin.After that,we construct a prediction model of the heating system’s operating conditions based on a digital twin.Finally,we combine the temperature monitoring of room temperature and the distribution of temperature transmission delay in the heat network to propose a dynamic supply-demand balance analysis model for the heat supply system that considers hysteresis characteristics.In simpler terms,we use the deep long and short term network algorithm to make precise predictions and assess the heat load on the demand side.We analyze the dynamic supply-demand balance of the heat supply system by considering its delayed temperature transmission characteristics.Taking into account the changes in supply and demand,we generate a production plan for the heat source.This allows us to monitor the heat supply system in real-time and evaluate future heat supply according to demand.(iii)This paper suggests a real-time optimal scheduling method for a heating system using model predictive control.Firstly,we propose a real-time optimization and scheduling framework for urban heating systems based on digital twin technology.We also summarize the real-time optimization and scheduling model for heating systems.To address the limitations of conventional methods,we suggest a real-time optimization approach that uses deep reinforcement learning.This method is capable of accurately regulating the heat load of the collaborative heating system source network for intelligent distribution across various heating stations.On the basis of optimizing scheduling,we propose a process for real-time optimization scheduling of a heat supply system.This process is based on model predictive control and in combination with the actual system,can accurately regulate and control the urban heat supply system ondemand.(iv)Based on the ideas and techniques explained in this paper,we created an urban heating system control platform that employs digital twins for precise on-demand control.We tested the effectiveness and reliability of this platform in partnership with the intelligent heating demonstration project in northern cities and towns.Practical engineering tests reveal that a city in Hebei province with a multi-heat source networked heating system achieves better energy efficiency when using the new solenoid operation scheme based on the digital twin simulation platform.This scheme saves 14,677.38 tce during the heating season compared to the original solenoid operation scheme.In Beijing’s Mentougou District heating demonstration project,the digital twin simulation platform facilitates dynamic supply-demand balance among the heating system’s various districts.In a heating project in Mentougou District,Beijing,we used a digital twin simulation platform to analyze and efficiently regulate the supply and demand of heat in each zone.This reduced the heat consumption in each area,resulting in a total energy savings of 5.59%.This reduction is equal to approximately1.622 t of carbon.In summary,this paper proposes a general digital twin-based intelligent heating system technical architecture from the level of information-physical fusion,establishes a dynamic supply-demand balance analysis model of the heating system and a real-time optimization scheduling method of the heating system based on the model prediction and control,and then builds an on-demand precision regulation platform of the urban heating system based on digital twins,which breaks through the limitations of the traditional operation and control methods of the heating system at the urban level.The results of the research in this paper can promote the smart heating system in northern cities and towns.The research results of this paper can promote the construction and development of intelligent heat supply system in northern towns,which is conducive to promoting the low-carbon transformation of China’s heat supply system.