Heat supply digital twin modeling method and system based on U3D engine and medium

Abstract

The invention provides a heat supply digital twin modeling method and system based on a U3D engine and a medium. The method comprises the steps that a historical static data set and a historical dynamic data set are obtained for preprocessing, a historical static optimized data set and a historical dynamic optimized data set are obtained, a U3D engine is combined with a preset parameterized component library to conduct heat supply digital twinning model construction, initialization is conducted based on the historical static data set, and a heat supply digital twinning model is obtained. Obtaining an initial heat supply digital twinborn model, generating an LOD model set with different precisions through an LOD rendering method, inputting the historical dynamic optimization data set into the initial heat supply digital twinborn model for processing, obtaining a simulation data set and equipment operation state evaluation data, and evaluating the qualified state of the model through threshold comparison; according to the method, data security fusion is realized through the block chain, the hydraulic and thermal coupling model is constructed by using the U3D engine, and predictive maintenance and multi-objective optimization are realized through an intelligent algorithm, so that intelligent construction of the heat supply word twinborn model is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of digital management and intelligent operation and maintenance of heating systems, in particular to a heating digital twin modeling method, system and medium based on a U3D engine. BACKGROUND

[0002] The application of digital twin technology in the field of heating has many deficiencies, for example, the traditional modeling process based on professional industrial software (such as ANSYS, COMSOL) is complex and requires professional personnel to operate, and it is difficult to realize real-time interaction. Although the modeling method based on BIM+GIS can construct a static geometric model, it lacks the ability to dynamically couple with physical fields and cannot reflect the hydraulic and thermal correlation characteristics of the system during operation. The data of the heating system has the characteristics of "multi-source heterogeneity, spatio-temporal correlation, and noise interference". The existing technology mostly uses centralized data processing architecture, which is prone to data delay and security risks, and the accuracy of the missing data completion is insufficient. Traditional digital twin systems only support two-dimensional visualization or static three-dimensional display, and lack immersive interaction capabilities. Multi-physical field coupling simulation mostly uses offline calculation mode, which cannot meet the real-time needs of dynamic regulation and control. In the existing technology, the application of U3D engine is mostly limited to geometric modeling and simple visualization, and its advantages in real-time rendering, physical engine, and cross-platform deployment are not fully utilized, especially the lack of customized development for multi-physical field coupling of heating systems. Therefore, there is an urgent need for a heating digital twin modeling method that integrates multi-source data fusion architecture, customized physical field coupling algorithms, and immersive interaction design.

[0003] In view of the above problems, an effective technical solution is currently needed. SUMMARY

[0004] The purpose of the present application is to provide a heating digital twin modeling method, system and medium based on a U3D engine, which can realize data security fusion through a blockchain, construct a hydraulic and thermal coupling model using a U3D engine, and achieve predictive maintenance and multi-objective optimization using intelligent algorithms, thereby realizing intelligent construction of a heating digital twin model.

[0005] In a first aspect, the present application provides a heating digital twin modeling method based on a U3D engine, comprising the following steps: Obtaining multi-source heterogeneous data of a heating system in a preset time period, including a historical static data set and a historical dynamic data set, and preprocessing to obtain a historical static optimization data set and a historical dynamic optimization data set; Constructing a heating digital twin model through a U3D engine combined with a preset parameterized component library, and initializing according to the historical static optimization data set to obtain an initial heating digital twin model; inputting the historical dynamic optimization dataset into the initial heating digital twin model for processing to obtain a simulation dataset and equipment operation state evaluation data; comparing the simulation dataset with the corresponding measured dataset and comparing the equipment operation state evaluation data with the corresponding equipment operation state measured data to obtain a model validity evaluation parameter of the initial heating digital twin model; comparing the model validity evaluation parameter with a preset model validity early warning threshold, and determining a model qualified state according to a threshold comparison result.

[0006] Optionally, in the heating digital twin modeling method based on the U3D engine, the multi-source heterogeneous data of the heating system in the preset time period, including the historical static dataset, the historical dynamic dataset, and the historical operation monitoring dataset, are obtained and preprocessed to obtain the historical static optimization dataset and the historical dynamic optimization dataset, including: obtaining the multi-source heterogeneous data of the heating system in the preset time period, including the historical static dataset and the historical dynamic dataset; the historical static dataset includes GIS map data, BIM model parameter data, and equipment operation and maintenance log data; the historical dynamic dataset includes pipe network node water supply operation data, heat exchange station operation data, and heating operation environment data; the historical static dataset is preprocessed by encryption fusion based on a blockchain to obtain the historical static optimization dataset; the historical dynamic dataset is repaired for missing values by using a preset spatiotemporal GAN model to obtain the historical dynamic optimization dataset.

[0007] Optionally, in the heating digital twin modeling method based on the U3D engine, the heating digital twin model is constructed by the U3D engine in combination with the preset parameterized component library, and is initialized according to the historical static optimization dataset to obtain an initial heating digital twin model and an LOD model set, including: creating the preset parameterized component library by ScriptableObject of the U3D engine; constructing the heating digital twin model by the U3D engine in combination with the preset parameterized component library to obtain a heating three-dimensional geometric model; inputting the historical static optimization dataset into the heating three-dimensional geometric model for model initialization to obtain the initial heating digital twin model; processing the initial heating digital twin model by an LOD rendering method to generate the LOD model set, including an LOD0 model, an LOD1 model, an LOD2 model, and an LOD3 model.

[0008] Optionally, in the U3D engine-based heating digital twin modeling method described in the present application, the historical dynamic optimization dataset is input into the initial heating digital twin model for processing to obtain a simulation dataset and equipment operation state evaluation data, including: The historical dynamic optimization dataset is input into the initial heating digital twin model as boundary conditions for coupling simulation processing by a pre-set hydraulic model and a thermal model in the model to obtain a simulation dataset and equipment operation state evaluation data. The simulation dataset includes simulation pressure, simulation flow rate, and simulation temperature. The equipment operation state evaluation data includes pump set efficiency, pipeline corrosion rate, and heat transfer coefficient of a heat exchanger.

[0009] Optionally, in the U3D engine-based heating digital twin modeling method described in the present application, the simulation dataset and the corresponding measured dataset, and the equipment operation state evaluation data and the corresponding equipment operation state measured data are compared to obtain a model validity evaluation parameter of the initial heating digital twin model, including: The simulation dataset and the corresponding measured dataset are compared to obtain simulation pressure deviation rate, simulation flow rate deviation rate, and simulation temperature deviation rate, and weighted sum processing is performed to obtain simulation data deviation rate. The equipment operation state evaluation data and the equipment operation state measured data are compared to obtain pump set efficiency deviation rate, pipeline corrosion rate deviation rate, and heat transfer coefficient of a heat exchanger deviation rate, and weighted sum processing is performed to obtain equipment operation evaluation deviation rate. The simulation data deviation rate and the equipment operation evaluation deviation rate are subjected to weighted sum processing to obtain a model validity evaluation parameter of the initial heating digital twin model.

[0010] Optionally, in the U3D engine-based heating digital twin modeling method described in the present application, the model validity evaluation parameter is compared with a pre-set model validity early warning threshold value, and the model qualified state is determined according to the threshold value comparison result, including: Obtain heating type feature data, query a pre-set heating type and model validity early warning threshold value mapping table according to the heating type feature data to obtain the corresponding model validity early warning threshold value. The model validity evaluation parameter is compared with the model validity early warning threshold value. If the model validity evaluation parameter is less than or equal to the model validity early warning threshold value, it is determined that the initial heating digital twin model is unqualified. If the model validity evaluation parameter is greater than the model validity early warning threshold value, it is determined that the initial heating digital twin model is qualified.

[0011] Optionally, in the U3D engine-based heating digital twin modeling method described in the present application, further comprising: acquiring real-time environmental data of a preset range of the heating system; acquiring real-time simulation data sets and device operation state real-time evaluation data, as well as historical simulation data and corresponding execution feedback data of the heating system in a preset time period; performing feature engineering processing according to the real-time environmental data, real-time simulation data sets, device operation state real-time evaluation data, historical simulation data, and execution feedback data to obtain a heating decision analysis data set, including a time feature data set, an operation feature data set, and an environmental feature data set; processing according to the operation feature data set in combination with preset historical fault data through a preset predictive maintenance model to obtain a fault probability value of the heating system; processing according to the time feature data set and the environmental feature data set in combination with preset historical heat consumption and preset historical environmental feature data through a preset heat load prediction model to obtain heat load prediction data; sending the fault probability value and the heat load prediction data to a user terminal for display.

[0012] In a second aspect, the present application provides a U3D engine-based heating digital twin modeling system, which comprises a memory and a processor, the memory comprising a U3D engine-based heating digital twin modeling method program, and the U3D engine-based heating digital twin modeling method program being executed by the processor to implement the following steps: acquiring multi-source heterogeneous data of the heating system in a preset time period, including historical static data sets and historical dynamic data sets, and performing preprocessing to obtain historical static optimization data sets and historical dynamic optimization data sets; constructing a heating digital twin model through a U3D engine in combination with a preset parameterized component library, and initializing according to the historical static optimization data sets to obtain an initial heating digital twin model; inputting the historical dynamic optimization data sets into the initial heating digital twin model for processing to obtain simulation data sets and device operation state evaluation data; comparing the simulation data sets with corresponding measured data sets, and comparing the device operation state evaluation data with corresponding device operation state measured data to obtain model validity evaluation parameters of the initial heating digital twin model; comparing the model validity evaluation parameters with a preset model validity warning threshold, and determining a model qualified state according to the threshold comparison result.

[0013] Optionally, in the U3D engine-based heating digital twin modeling system described in the application, the multi-source heterogeneous data of the heating system in the preset time period is obtained, including a historical static data set, a historical dynamic data set, and a historical operation monitoring data set, and is preprocessed to obtain a historical static optimization data set and a historical dynamic optimization data set, including: The multi-source heterogeneous data of the heating system in the preset time period is obtained, including a historical static data set and a historical dynamic data set; The historical static data set includes GIS map data, BIM model parameter data, and equipment operation and maintenance log data; The historical dynamic data set includes pipe network node water supply operation data, heat exchange station operation data, and heating operation environment data; The historical static data set is preprocessed by encryption fusion based on a blockchain, and the historical dynamic data set is repaired for missing values by a preset spatiotemporal GAN model to obtain a historical static optimization data set and a historical dynamic optimization data set.

[0014] In a third aspect, the application further provides a computer-readable storage medium, wherein a U3D engine-based heating digital twin modeling method program is stored in the computer-readable storage medium, and the U3D engine-based heating digital twin modeling method program is executed by a processor to implement the steps of the U3D engine-based heating digital twin modeling method according to any one of the above.

[0015] As can be seen from the above, the U3D engine-based heating digital twin modeling method, system, and medium provided by the application realize data security fusion through a blockchain, utilize a U3D engine to construct a hydraulic and thermal coupling model, and implement predictive maintenance and multi-objective optimization by an intelligent algorithm, so as to realize intelligent construction of a heating digital twin model.

[0016] Other features and advantages of the application will be described in the following description, and some will become apparent from the description, or will be understood from the practice of the application. The purpose and other advantages of the application can be achieved and obtained by the structures specifically pointed out in the written description and the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the application, the following will briefly introduce the drawings needed to be used in the embodiments of the application. It should be understood that the following drawings only show some embodiments of the application, and therefore should not be considered as limiting the scope. For those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0018] Figure 1A flowchart of the U3D engine-based heating digital twin modeling method provided in the embodiments of the present application is shown in FIG. 1. Figure 2 A flowchart of obtaining the historical static optimization dataset and the historical dynamic optimization dataset in the U3D engine-based heating digital twin modeling method provided in the embodiments of the present application is shown in FIG. 2. Figure 3 A flowchart of obtaining the initial heating digital twin model and the LOD model set in the U3D engine-based heating digital twin modeling method provided in the embodiments of the present application is shown in FIG. 3. DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. The components of the embodiments of the present application described and shown in the accompanying drawings can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of the present application.

[0020] It should be noted that similar reference numerals and letters represent similar items in the following drawings, and therefore, once an item is defined in one drawing, it need not be further defined and explained in subsequent drawings. Meanwhile, in the description of the present application, the terms "first", "second", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0021] Please refer to Figure 1 , Figure 1 A flowchart of the U3D engine-based heating digital twin modeling method in some embodiments of the present application is shown in FIG. 1. The U3D engine-based heating digital twin modeling method is used in terminal devices such as computers, mobile phones, and the like. The U3D engine-based heating digital twin modeling method includes the following steps: S11, obtaining multi-source heterogeneous data of a heating system in a preset time period, including a historical static dataset and a historical dynamic dataset, and preprocessing to obtain a historical static optimization dataset and a historical dynamic optimization dataset; S12, constructing a heating digital twin model through a U3D engine combined with a preset parameterized component library, and initializing according to the historical static optimization dataset to obtain an initial heating digital twin model; S13, inputting the historical dynamic optimization dataset into the initial heating digital twin model for processing to obtain a simulation dataset and equipment operation state evaluation data; S14, compare the simulation data set with the corresponding measured data set, and compare the equipment running state evaluation data with the corresponding equipment running state measured data, to obtain a model validity evaluation parameter of the initial heating digital twin model; S15, compare the model validity evaluation parameter with a preset model validity early warning threshold, and determine the model qualified state according to the threshold comparison result.

[0022] It should be noted that by using the encryption fusion preprocessing based on the blockchain for the historical static data set, and using the preset spatiotemporal GAN model for the historical dynamic data set to repair missing values, the historical static optimization data set and the historical dynamic optimization data set are obtained, the U3D engine is combined with the preset parameterized component library to construct the heating digital twin model, and the model is initialized and simulated and evaluated, and according to the threshold comparison, it is determined whether the constructed model is qualified. If not, adjust the model parameters according to the deviation, so as to improve the data security, model construction and optimization.

[0023] Please refer to Figure 2 , Figure 2 is the flowchart of obtaining the historical static optimization data set and the historical dynamic optimization data set in the U3D engine-based heating digital twin modeling method in some embodiments of the present application. According to the embodiment of the present application, the multi-source heterogeneous data of the heating system in a preset time period is obtained, including historical static data set, historical dynamic data set and historical running monitoring data set, and is preprocessed to obtain historical static optimization data set and historical dynamic optimization data set, including: S21, obtaining multi-source heterogeneous data of the heating system in a preset time period, including historical static data set and historical dynamic data set; S211, the historical static data set includes GIS map data, BIM model parameter data and equipment operation and maintenance log data; S212, the historical dynamic data set includes pipe network node water supply operation data, heat exchange station operation data and heating operation environment data; S221, the historical static data set is preprocessed by encryption fusion based on the blockchain to obtain the historical static optimization data set; S222, the historical dynamic data set is repaired by using a preset spatiotemporal GAN model to repair missing values to obtain the historical dynamic optimization data set.

[0024] It should be noted that the edge node + regional gateway + cloud platform three-level collection network is adopted to collect multi-source heterogeneous data in the embodiment of the present application, the edge node is deployed at the heat exchange station and the key node of the pipe network, historical dynamic data is collected, the regional gateway converges edge data through the 5G private network, local preprocessing (such as filtering and format conversion) is performed, the cloud platform receives global data, and GIS map (precision 1:500), BIM (component level parameter) model parameter data and equipment operation log data are used to construct static data fusion; a data interaction protocol is designed based on the alliance chain architecture, each edge node is used as a chain node, SHA-256 algorithm is used to hash and encrypt the data, the data right and access control are realized through the smart contract, for example, the heat exchange station operation data is only authorized to be read and written by the administrator node, the historical data cannot be tampered with and can be traced, and the data island and safety hazards are solved; a data repair model is constructed by fusing the space-time attention mechanism and the generative adversarial network (GAN), the space-time coordinates of the missing data (such as the data missing of a pipe network node from 8:00 to 8:10) are input, the model learns the correlation of the historical data of adjacent nodes (spatial dimension) and the 24-hour cycle law (time dimension) of the node, and generates the filling data, wherein, the preset space-time GAN model collects historical temperature, pressure and other data of each node of the pipe network, constructs a sample set containing space-time correlation, learns the normal data law by the generator, generates pseudo data, and distinguishes the true and false data by the discriminator, the two are trained in opposition, the space-time attention mechanism is used to make the model pay attention to the correlation of adjacent nodes and the 24-hour cycle law, and finally the filling data output by the generator is used for missing value repair.

[0025] Please refer to Figure 3 , Figure 3 is a flowchart of obtaining an initial heat supply digital twin model in the U3D engine-based heat supply digital twin modeling method in some embodiments of the present application. According to the embodiment of the present application, the U3D engine is combined with a preset parameterized component library to construct a heat supply digital twin model, and the initial heat supply digital twin model and LOD model set are obtained by initializing according to the historical static optimization data set, including: S31, creating a preset parameterized component library through ScriptableObject of the U3D engine; S32, constructing a heat supply digital twin model through the U3D engine combined with the preset parameterized component library, and obtaining a heat supply three-dimensional geometric model; S33, inputting the historical static optimization data set into the heat supply three-dimensional geometric model for model initialization, and obtaining an initial heat supply digital twin model; S34, processing the initial heat supply digital twin model through the LOD rendering method to generate an LOD model set, including LOD0 model, LOD1 model, LOD2 model and LOD3 model.

[0026] It should be noted that the preset parameterized component library includes pipes, valves, heat exchangers and other standardized components, each component includes geometric parameters, physical properties and operation and maintenance parameters, through a preset conversion plug-in from BIM to U3D model, based on simplified non-critical components and reserved key parameters, the model volume compression is realized, then the GIS map data is integrated, the terrain model is constructed through the Terrain component of U3D, the accurate matching modeling of the pipe network and the geographical environment is realized, and then the heating three-dimensional geometric model is constructed, and the model is initialized to obtain an initial heating digital twin model; in order to balance the visual effect and the consumption of computing resources when a user observes the model through a VR device, simplified versions of the pipe network main branch, heat exchange station and other models are automatically generated according to a four-level LOD standard to obtain a LOD model set, LOD (Level of Detail) is a technology for balancing performance and visual effect and optimizing rendering efficiency by dynamically adjusting the detail level of an object, for example, an LOD1 model retains 90% of the key size and deletes the detail texture, an LOD3 model only retains the center line of the pipe network and the outline of the equipment, and the number of triangular faces is reduced to 10% of the original model, so that the fineness of the model is dynamically adjusted according to the observation distance, and the problem of lag caused by too many details in the traditional digital twin model is solved.

[0027] According to the embodiment of the present application, the historical dynamic optimization data set is input into the initial heating digital twin model for processing to obtain a simulation data set and equipment operation state evaluation data, which comprises: The historical dynamic optimization data set is input into the initial heating digital twin model as a boundary condition for coupled simulation processing by a preset hydraulic model and a thermal model in the model to obtain a simulation data set and equipment operation state evaluation data; The simulation data set comprises simulation pressure, simulation flow and simulation temperature; The equipment operation state evaluation data comprises pump group efficiency, pipe corrosion rate and heat exchanger heat transfer coefficient.

[0028] It should be noted that the historical dynamic optimization data set is input into the initial heating digital twin model as a boundary condition for coupled simulation processing by a preset hydraulic model and a thermal model in the model, wherein the preset hydraulic model is trained by solving the modified Darcy-Weisbach equation by the finite volume method, and the preset thermal model is trained by solving the three-dimensional unsteady heat conduction equation by the explicit difference method, the hydraulic model transmits flow field data to the thermal model every 10 steps, the thermal model feeds back temperature field data to the hydraulic model every 5 steps, and the two-way interaction between the hydraulic model and the thermal model is realized through a "data buffer pool", so that the hydraulic and thermal coupled model is constructed.

[0029] According to the embodiment of the present application, the comparison of the simulation data set and the corresponding measured data set and the comparison of the device running state evaluation data and the corresponding device running state measured data obtain the model validity evaluation parameter of the initial heating digital twin model, which comprises: The comparison of the simulation data set and the corresponding measured data set obtains the simulation pressure deviation rate, the simulation flow deviation rate and the simulation temperature deviation rate, and the weighted sum processing is performed to obtain the simulation data deviation rate; The comparison of the device running state evaluation data and the device running state measured data obtains the pump group efficiency deviation rate, the pipeline corrosion rate deviation rate and the heat exchanger heat transfer coefficient deviation rate, and the weighted sum processing is performed to obtain the device running evaluation deviation rate; The weighted sum processing of the simulation data deviation rate and the device running evaluation deviation rate obtains the model validity evaluation parameter of the initial heating digital twin model.

[0030] It should be noted that the comparison of the simulation data set and the corresponding measured data set obtains the simulation pressure deviation rate, the simulation flow deviation rate and the simulation temperature deviation rate, wherein the simulation pressure deviation rate is the ratio of the absolute value of the difference between the simulation pressure data and the measured pressure data to the measured pressure data, and the simulation flow deviation rate and the simulation temperature deviation rate can be obtained in the same way. The pump group efficiency deviation rate, the pipeline corrosion rate deviation rate and the heat exchanger heat transfer coefficient deviation rate can be obtained by weighted sum processing, and finally the model validity evaluation parameter of the evaluation model is obtained.

[0031] According to the embodiment of the present application, the threshold comparison of the model validity evaluation parameter and the preset model validity early warning threshold value determines the model qualified state according to the threshold comparison result, which comprises: Obtain the heating type feature data, query the preset heating type and model validity early warning threshold value mapping table according to the heating type feature data, and obtain the corresponding model validity early warning threshold value; The threshold comparison of the model validity evaluation parameter and the model validity early warning threshold value; If the model validity evaluation parameter is less than or equal to the model validity early warning threshold value, it is determined that the initial heating digital twin model is unqualified; If the model validity evaluation parameter is greater than the model validity early warning threshold value, it is determined that the initial heating digital twin model is qualified.

[0032] It should be noted that, in order to improve the accuracy of model evaluation, different evaluation thresholds are determined based on different types of heating type feature data, wherein the heating type includes a city-level central heating system and a regional heating system, and the heating type feature data is represented by different numerical values, such as 1 for the city-level central heating system and 0 for the regional heating system. If the model validity evaluation parameter is less than or equal to the model validity early warning threshold, it is determined that the initial heating digital twin model is unqualified, and the model parameter iteration correction process is started, such as correcting the physical property parameters of the component and the simulation algorithm coefficient.

[0033] According to the embodiment of the present application, further comprising: acquiring real-time environmental data of a preset range of the heating system; acquiring real-time simulation data set and equipment running state real-time evaluation data, and historical simulation data and execution feedback data corresponding to the heating system in a preset time period; performing feature engineering processing according to the real-time environmental data, real-time simulation data set and equipment running state real-time evaluation data, and the historical simulation data and execution feedback data, to obtain a heating decision analysis data set, including a time feature data set, a running feature data set and an environmental feature data set; obtaining a fault probability value of the heating system by processing the running feature data set in combination with preset historical fault data through a preset predictive maintenance model; obtaining heat load prediction data by processing the time feature data set and the environmental feature data set in combination with preset historical heat consumption and preset historical environmental feature data through a preset heat load prediction model; sending the fault probability value and the heat load prediction data to the user end for display.

[0034] It should be noted that the real-time simulation data and the historical simulation data are combined to predict and optimize the heating system in the heating operation environment, wherein the time feature data set includes hour segments (0-23), workday / weekend label data (represented by 0 / 1 respectively) and holiday enumeration values, the operation feature data set includes pressure fluctuation amplitude, temperature change rate and equipment cumulative running time in a preset time period, and the environment feature data set includes outdoor temperature and historical same period difference value and wind speed level (1-6 levels are adopted in the embodiment), then the time feature data set, the operation feature data set and the environment feature data set are associated and mapped with the original data, a decision analysis data set containing preset feature dimensions is constructed, and standardized processing is performed to obtain a heating decision analysis data set, then the preset predictive maintenance model and the preset heat load prediction model are used for processing respectively to generate fault probability values and heat load prediction data in a preset time period, wherein the random forest classifier is used to construct the predictive maintenance model based on the preset historical operation feature data set and the preset historical fault data, the LSTM network is used to construct the heat load prediction model based on the preset historical actual heat consumption, the preset historical environment feature data and the preset historical time feature data set in a preset time period (such as 24h), and finally the fault probability values and the heat load prediction data are sent to the user end for display.

[0035] It is worth mentioning that, according to the embodiment of the present application, further comprising: generating a plurality of sets of optimization strategies according to a preset multi-objective optimization model; performing simulation analysis on the initial heating digital twin model according to the optimization strategy set respectively to obtain corresponding optimization strategy evaluation data, including energy consumption data and room temperature compliance rate; comparing the energy consumption data with the historical energy consumption average in a preset time period to obtain the energy consumption reduction rate, and performing weighted sum processing on the room temperature compliance rate to obtain a strategy evaluation value; arranging the strategy evaluation values in descending order, and determining the optimization strategy with the largest strategy evaluation value as the optimal optimization strategy; It should be noted that, with the lowest energy consumption and the highest room temperature standard rate as the target, the constraint condition (such as pipe network pressure ≤ 1.5 MPa) is determined, the decision variable set (such as heat source output 5-20 MW) is adopted, and the non-dominated sorting and congestion calculation are screened by using the NSGA-III algorithm to solve the high-quality solution, so as to build a multi-objective optimization model. Each set of optimization strategies includes the control parameters of heat source output, water pump frequency and valve opening. The simulation is carried out by evaluating the qualified initial heating digital twin model, and the energy consumption data and room temperature standard rate of each scheme are output. The energy consumption data is compared with the average energy consumption in the historical preset time period, and the energy consumption reduction rate is obtained. The energy consumption reduction rate is the ratio of the difference between the average energy consumption and the energy consumption data to the average energy consumption. If it is negative, it is recorded as 0. The energy consumption reduction rate is weighted and summed with the room temperature standard rate to obtain the strategy evaluation value. The optimal optimization strategy with the highest strategy evaluation value is obtained by descending arrangement, and the optimal optimization strategy is determined.

[0036] It is worth mentioning that, according to the embodiment of the present application, further comprising: According to the optimal optimization strategy, the operation of the heating system is controlled, and the real-time operation data of the heating system is obtained, including real-time energy consumption data, real-time room temperature distribution data and real-time equipment operation state data; According to the real-time room temperature distribution data, the real-time standard rate of room temperature is obtained; The real-time energy consumption data, the real-time standard rate of room temperature and the real-time equipment operation state data are compared with the corresponding predicted data to obtain the energy consumption prediction deviation rate, the room temperature standard prediction deviation rate and the equipment operation deviation rate, and the prediction deviation value is obtained by weighted sum processing; If the prediction deviation value is greater than the preset prediction deviation warning threshold, the model parameter correction is triggered.

[0037] It should be noted that the real-time energy consumption data includes heat source gas consumption, water pump power consumption and pipe network heat loss. The real-time room temperature distribution data refers to the temperature of each room, which is aggregated by building to obtain the real-time standard rate of room temperature (i.e. the proportion of users with temperature in the range of 18-22℃). The real-time equipment operation state data includes pump outlet pressure, valve actual opening and heat exchanger inlet and outlet temperature difference. The energy consumption prediction deviation rate is obtained by averaging the heat source gas consumption deviation rate, water pump power consumption deviation rate and pipe network heat loss deviation rate. The heat source gas consumption deviation rate is the ratio of the absolute value of the difference between the heat source gas consumption and the heat source gas prediction value to the heat source gas prediction value. Similarly, the room temperature standard prediction deviation rate and the equipment operation deviation rate can be obtained. Finally, the energy consumption prediction deviation rate, the room temperature standard prediction deviation rate and the equipment operation deviation rate are weighted and summed to obtain the prediction deviation value. If it is greater than the threshold value, the model parameter correction is triggered. Otherwise, normal monitoring can be performed.

[0038] The application also discloses a U3D engine-based heating digital twin modeling system, comprising a memory and a processor, wherein the memory comprises a U3D engine-based heating digital twin modeling method program, and the U3D engine-based heating digital twin modeling method program is executed by the processor to implement the following steps: Multi-source heterogeneous data of the heating system in a preset time period is acquired, including a historical static data set and a historical dynamic data set, and is preprocessed to obtain a historical static optimization data set and a historical dynamic optimization data set; A heating digital twin model is constructed by a U3D engine in combination with a preset parameterized component library, and is initialized according to the historical static optimization data set to obtain an initial heating digital twin model; The historical dynamic optimization data set is input into the initial heating digital twin model for processing to obtain a simulation data set and equipment operation state evaluation data; The simulation data set is compared with a corresponding measured data set, and the equipment operation state evaluation data is compared with corresponding equipment operation state measured data to obtain a model validity evaluation parameter of the initial heating digital twin model; The model validity evaluation parameter is compared with a preset model validity early warning threshold value, and a model qualified state is determined according to a threshold comparison result.

[0039] It should be noted that the historical static data set is preprocessed by encryption fusion based on a blockchain, the historical dynamic data set is repaired for missing values by a preset spatiotemporal GAN model, the historical static optimization data set and the historical dynamic optimization data set are obtained, a heating digital twin model is constructed by a U3D engine in combination with a preset parameterized component library, the model is initialized and simulated, and according to the threshold comparison, it is determined whether the constructed model is qualified, and if not, the model parameter is adjusted according to the deviation, thereby improving data security, model construction and optimization.

[0040] According to the embodiment of the application, the multi-source heterogeneous data of the heating system in a preset time period is acquired, including a historical static data set, a historical dynamic data set and a historical operation monitoring data set, and is preprocessed to obtain a historical static optimization data set and a historical dynamic optimization data set, comprising: Multi-source heterogeneous data of the heating system in a preset time period is acquired, including a historical static data set and a historical dynamic data set; The historical static data set comprises GIS map data, BIM model parameter data and equipment operation and maintenance log data; The historical dynamic data set comprises pipe network node water supply operation data, heat exchange station operation data and heating operation environment data; The historical static data set is preprocessed by adopting a blockchain-based encryption fusion to obtain a historical static optimization data set; The historical dynamic data set is repaired by adopting a preset space-time GAN model to obtain a historical dynamic optimization data set.

[0041] It should be noted that the edge node + regional gateway + cloud platform three-level acquisition network is adopted to acquire multi-source heterogeneous data in the embodiment of the application, the edge node is deployed at the heat exchange station and the key node of the pipe network, historical dynamic data is acquired, the regional gateway converges the edge data through the 5G private network, local preprocessing (such as filtering and format conversion) is performed, the cloud platform receives global data, and GIS map (precision 1:500), BIM (component level parameter) model parameter data and equipment operation log data are fused to construct a static data fusion; a data interaction protocol is designed based on the alliance chain architecture, each edge node is used as a chain node, SHA-256 algorithm is used to hash and encrypt the data, data right and access control are realized through the smart contract, for example, the heat exchange station operation data is only authorized to be read and written by the administrator node, the historical data cannot be tampered with and is traceable, and the data island and safety hazards are solved; a data repair model is constructed by fusing the space-time attention mechanism and the generative adversarial network (GAN), the space-time coordinates of the missing data (such as the data missing of a pipe network node at 8:00-8:10) are input, the model learns the correlation (spatial dimension) of the historical data of adjacent nodes and the 24-hour cycle rule (time dimension) of the node, and generates the filling data, wherein the preset space-time GAN model collects historical temperature, pressure and other data of each node of the pipe network, constructs a sample set containing space-time correlation, learns the normal data rule by the generator, generates pseudo data, and distinguishes the true and false data by the discriminator, the two are trained in opposition, the space-time attention mechanism is used to make the model pay attention to the correlation of adjacent nodes and the 24-hour cycle rule, and finally the filling data output by the generator is used for missing value repair.

[0042] According to the embodiment of the application, the heat supply digital twin model is constructed by the U3D engine in combination with the preset parameterized component library, and is initialized according to the historical static optimization data set to obtain an initial heat supply digital twin model and a LOD model set, including: The preset parameterized component library is created by ScriptableObject of the U3D engine; The heat supply digital twin model is constructed by the U3D engine in combination with the preset parameterized component library to obtain a heat supply three-dimensional geometric model; The historical static optimization data set is input into the heat supply three-dimensional geometric model for model initialization to obtain an initial heat supply digital twin model; The initial heat supply digital twin model is processed by the LOD rendering method to generate a LOD model set, including LOD0 model, LOD1 model, LOD2 model and LOD3 model.

[0043] It should be noted that the preset parameterized component library includes pipes, valves, heat exchangers and other standardized components, each component includes geometric parameters, physical properties and operation and maintenance parameters, through a preset conversion plug-in from BIM to U3D model, based on simplified non-critical components and reserved key parameters, the model volume is compressed, GIS map data is integrated, a terrain model is constructed through the Terrain component of U3D, accurate matching modeling of the pipe network and the geographical environment is realized, a heating three-dimensional geometric model is constructed, and the model is initialized to obtain an initial heating digital twin model; in order to balance the visual effect and the consumption of computing resources when a user observes the model through a VR device, simplified versions of the main and branch lines of the pipe network, heat exchange stations and other models are automatically generated according to a four-level LOD standard to obtain a set of LOD models, LOD (Level of Detail) is a technology for balancing performance and visual effect and optimizing rendering efficiency by dynamically adjusting the level of detail of an object, for example, an LOD1 model retains 90% of the key size and deletes detail textures, an LOD3 model only retains the center line of the pipe network and the outline of the equipment, and the number of triangular faces is reduced to 10% of the original model, the fineness of the model is dynamically adjusted according to the observation distance, and the problem of lag caused by too many details in the traditional digital twin model is solved.

[0044] According to the embodiment of the present application, the historical dynamic optimization data set is input into the initial heating digital twin model for processing to obtain a simulation data set and equipment operation state evaluation data, which comprises: The historical dynamic optimization data set is input into the initial heating digital twin model as a boundary condition for coupled simulation processing by a preset hydraulic model and a thermal model in the model to obtain a simulation data set and equipment operation state evaluation data; The simulation data set comprises simulation pressure, simulation flow and simulation temperature; The equipment operation state evaluation data comprises pump group efficiency, pipe corrosion rate and heat exchanger heat transfer coefficient.

[0045] It should be noted that the historical dynamic optimization data set is input into the initial heating digital twin model as a boundary condition for coupled simulation processing by a preset hydraulic model and a thermal model in the model, wherein the preset hydraulic model is trained by solving the modified Darcy-Weisbach equation using the finite volume method, and the preset thermal model is trained by solving the three-dimensional unsteady heat conduction equation using the explicit difference method, the hydraulic model transmits flow field data to the thermal model every 10 steps, the thermal model feeds back temperature field data to the hydraulic model every 5 steps, and the hydraulic model and the thermal model are bidirectionally interacted through a "data buffer pool", so that a hydraulic and thermal coupled model is constructed.

[0046] According to the embodiment of the present application, the comparison of the simulation data set and the corresponding measured data set and the comparison of the device running state evaluation data and the corresponding device running state measured data obtain the model validity evaluation parameter of the initial heating digital twin model, which comprises: The comparison of the simulation data set and the corresponding measured data set obtains the simulation pressure deviation rate, the simulation flow deviation rate and the simulation temperature deviation rate, and the weighted sum processing is performed to obtain the simulation data deviation rate; The comparison of the device running state evaluation data and the device running state measured data obtains the pump group efficiency deviation rate, the pipeline corrosion rate deviation rate and the heat exchanger heat transfer coefficient deviation rate, and the weighted sum processing is performed to obtain the device running evaluation deviation rate; The weighted sum processing of the simulation data deviation rate and the device running evaluation deviation rate obtains the model validity evaluation parameter of the initial heating digital twin model.

[0047] It should be noted that the comparison of the simulation data set and the corresponding measured data set obtains the simulation pressure deviation rate, the simulation flow deviation rate and the simulation temperature deviation rate, wherein the simulation pressure deviation rate is the ratio of the absolute value of the difference between the simulation pressure data and the measured pressure data to the measured pressure data, and the simulation flow deviation rate and the simulation temperature deviation rate can be obtained in the same way. The pump group efficiency deviation rate, the pipeline corrosion rate deviation rate and the heat exchanger heat transfer coefficient deviation rate can be obtained, and the model validity evaluation parameter of the evaluation model validity is finally obtained through the weighted sum processing.

[0048] According to the embodiment of the present application, the threshold comparison of the model validity evaluation parameter and the preset model validity early warning threshold value determines the model qualified state according to the threshold comparison result, which comprises: Obtain the heating type feature data, query the preset heating type and model validity early warning threshold value mapping table according to the heating type feature data, and obtain the corresponding model validity early warning threshold value; The threshold comparison of the model validity evaluation parameter and the model validity early warning threshold value; If the model validity evaluation parameter is less than or equal to the model validity early warning threshold value, it is determined that the initial heating digital twin model is unqualified; If the model validity evaluation parameter is greater than the model validity early warning threshold value, it is determined that the initial heating digital twin model is qualified.

[0049] It should be noted that, in order to improve the accuracy of model evaluation, different evaluation thresholds are determined based on different types of heating type feature data, wherein the heating type includes a city-level central heating system and a regional heating system, and the heating type feature data is represented by different numerical values, such as 1 for the city-level central heating system and 0 for the regional heating system. If the model validity evaluation parameter is less than or equal to the model validity early warning threshold, it is determined that the initial heating digital twin model is unqualified, and the model parameter iteration correction process is started, such as correcting the physical property parameters of the component and the simulation algorithm coefficient.

[0050] According to the embodiment of the present application, further comprising: acquiring real-time environmental data of a preset range of the heating system; acquiring real-time simulation data set and equipment running state real-time evaluation data, and historical simulation data and execution feedback data corresponding to the heating system in a preset time period; performing feature engineering processing according to the real-time environmental data, real-time simulation data set and equipment running state real-time evaluation data, and the historical simulation data and execution feedback data to obtain a heating decision analysis data set, including a time feature data set, a running feature data set and an environmental feature data set; obtaining a fault probability value of the heating system by processing the running feature data set in combination with the preset historical fault data through a preset predictive maintenance model; obtaining heat load prediction data by processing the time feature data set and the environmental feature data set in combination with the preset historical heat consumption and the preset historical environmental feature data through a preset heat load prediction model; sending the fault probability value and the heat load prediction data to the user end for display.

[0051] It should be noted that the real-time simulation data and the historical simulation data are combined to predict and optimize the heating system in the heating operation environment, wherein the time feature data set includes hour segments (0-23), workday / weekend label data (represented by 0 / 1 respectively) and holiday enumeration values, the operation feature data set includes pressure fluctuation amplitude, temperature change rate and equipment cumulative running time in a preset time period, and the environment feature data set includes outdoor temperature and historical same period difference value and wind speed level (1-6 levels are adopted in the embodiment), then the time feature data set, the operation feature data set and the environment feature data set are associated and mapped with the original data, a decision analysis data set containing preset feature dimensions is constructed, and standardized processing is performed to obtain a heating decision analysis data set, then the preset predictive maintenance model and the preset heat load prediction model are used for processing respectively to generate fault probability values and heat load prediction data in a preset time period, wherein the random forest classifier is used to construct the predictive maintenance model based on the preset historical operation feature data set and the preset historical fault data, the LSTM network is used to construct the heat load prediction model based on the preset historical actual heat consumption, the preset historical environment feature data and the preset historical time feature data set in a preset time period (such as 24h), and finally the fault probability values and the heat load prediction data are sent to the user end for display.

[0052] It is worth mentioning that, according to the embodiment of the present application, further comprising: generating a plurality of sets of optimization strategies according to a preset multi-objective optimization model; performing simulation analysis on the initial heating digital twin model according to the optimization strategy set respectively to obtain corresponding optimization strategy evaluation data, including energy consumption data and room temperature compliance rate; comparing the energy consumption data with the historical average energy consumption in a preset time period to obtain the energy consumption reduction rate, and performing weighted summation processing on the room temperature compliance rate to obtain a strategy evaluation value; arranging the strategy evaluation values in descending order, and determining the optimization strategy with the largest strategy evaluation value as the optimal optimization strategy; It should be noted that, with the lowest energy consumption and the highest room temperature standard rate as the target, the constraint condition (such as pipe network pressure ≤ 1.5 MPa) is determined, the decision variable set (such as heat source output 5-20 MW) is adopted, and the non-dominated sorting and congestion calculation are screened by using the NSGA-III algorithm to solve the high-quality solution, so as to build a multi-objective optimization model. Each set of optimization strategies includes the control parameters of heat source output, water pump frequency and valve opening. The simulation is carried out by evaluating the qualified initial heating digital twin model, and the energy consumption data and room temperature standard rate of each scheme are output. The energy consumption data is compared with the average energy consumption in the historical preset time period, and the energy consumption reduction rate is obtained. The energy consumption reduction rate is the ratio of the difference between the average energy consumption and the energy consumption data to the average energy consumption. If it is negative, it is recorded as 0. The energy consumption reduction rate is weighted and summed with the room temperature standard rate to obtain the strategy evaluation value. The optimal optimization strategy with the highest strategy evaluation value is obtained by descending arrangement, and the optimal optimization strategy is determined.

[0053] It is worth mentioning that, according to the embodiment of the present application, further comprising: According to the optimal optimization strategy, the operation of the heating system is controlled, and the real-time operation data of the heating system is obtained, including real-time energy consumption data, real-time room temperature distribution data and real-time equipment operation state data; According to the real-time room temperature distribution data, the real-time standard rate of room temperature is obtained; The real-time energy consumption data, the real-time standard rate of room temperature and the real-time equipment operation state data are compared with the corresponding predicted data to obtain the energy consumption prediction deviation rate, the room temperature standard prediction deviation rate and the equipment operation deviation rate, and the prediction deviation value is obtained by weighted sum processing; If the prediction deviation value is greater than the preset prediction deviation warning threshold, the model parameter correction is triggered.

[0054] It should be noted that the real-time energy consumption data includes heat source gas consumption, water pump power consumption and pipe network heat loss. The real-time room temperature distribution data refers to the temperature of each room, which is aggregated by building to obtain the real-time standard rate of room temperature (i.e. the proportion of users with temperature in the range of 18-22℃). The real-time equipment operation state data includes pump outlet pressure, valve actual opening and heat exchanger inlet and outlet temperature difference. The energy consumption prediction deviation rate is obtained by averaging the heat source gas consumption deviation rate, water pump power consumption deviation rate and pipe network heat loss deviation rate. The heat source gas consumption deviation rate is the ratio of the absolute value of the difference between the heat source gas consumption and the heat source gas prediction value to the heat source gas prediction value. Similarly, the room temperature standard prediction deviation rate and the equipment operation deviation rate can be obtained. Finally, the energy consumption prediction deviation rate, the room temperature standard prediction deviation rate and the equipment operation deviation rate are weighted and summed to obtain the prediction deviation value. If it is greater than the threshold value, the model parameter correction is triggered. Otherwise, normal monitoring can be performed.

[0055] The third aspect of the present application provides a readable storage medium, wherein a U3D engine-based heat supply digital twin modeling method program is stored in the readable storage medium, and the U3D engine-based heat supply digital twin modeling method program, when executed by a processor, implements the steps of the U3D engine-based heat supply digital twin modeling method according to any one of the preceding aspects.

[0056] The U3D engine-based heat supply digital twin modeling method, system and medium disclosed in the present application realize data security fusion through a blockchain, utilize a U3D engine to construct a hydraulic and thermal coupling model, and utilize an intelligent algorithm to realize predictive maintenance and multi-objective optimization, so as to realize intelligent construction of a heat supply digital twin model.

[0057] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other manners. The above-described device embodiments are merely illustrative. For example, the division of the units is merely a logical function division. There can be another division manner for the actual implementation, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the various components shown or discussed can be through some interfaces, indirect coupling or communication connection between the devices or units, and can be electrical, mechanical or other forms.

[0058] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units; they can be located in one place, or distributed on multiple network units; and part or all of the units can be selected according to actual needs to achieve the purpose of the present embodiment.

[0059] In addition, each functional unit in each embodiment of the present application can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be realized in the form of hardware or in the form of hardware plus software functional unit.

[0060] Those of ordinary skill in the art can understand that all or part of the steps of the above-mentioned method embodiments can be completed by program instruction-related hardware, and the foregoing program can be stored in a readable storage medium, and the program, when executed, executes steps including the above-mentioned method embodiments; and the foregoing storage medium includes mobile storage devices, read-only memories (ROMs), random access memories (RAMs), magnetic discs or optical discs, and various storage media that can store program codes.

[0061] Alternatively, the above-mentioned integrated unit of the present application, if realized in the form of a software function module and sold or used as an independent product, can also be stored in a readable storage medium. Based on such understanding, the technical solutions of the embodiments of the present application can be embodied in the form of a software product, which is stored in a storage medium, includes several instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: mobile storage devices, ROM, RAM, magnetic or optical disks, and various media that can store program codes.

Claims

1. A heating digital twin modeling method based on the U3D engine, characterized in that, Includes the following steps: Acquire multi-source heterogeneous data of the heating system within a preset time period, including historical static datasets and historical dynamic datasets, and preprocess them to obtain historical static optimized datasets and historical dynamic optimized datasets. A heating digital twin model is constructed using the U3D engine in conjunction with a pre-defined parametric component library, and initialized based on the historical static optimization dataset to obtain an initial heating digital twin model. The historical dynamic optimization dataset is input into the initial heating digital twin model for processing to obtain simulation dataset and equipment operation status evaluation data; The simulation dataset is compared with the corresponding measured dataset, and the equipment operation status evaluation data is compared with the corresponding measured equipment operation status data to obtain the model effectiveness evaluation parameters of the initial heating digital twin model. The model validity evaluation parameters are compared with the preset model validity warning threshold, and the model qualification status is determined based on the threshold comparison result.

2. The heating digital twin modeling method based on the U3D engine according to claim 1, characterized in that, The process involves acquiring multi-source heterogeneous data of the heating system within a preset time period, including historical static datasets, historical dynamic datasets, and historical operation monitoring datasets, and preprocessing this data to obtain historical static optimization datasets and historical dynamic optimization datasets, including: Acquire multi-source heterogeneous data of the heating system within a preset time period, including historical static datasets and historical dynamic datasets; The historical static dataset includes GIS map data, BIM model parameter data, and equipment operation and maintenance log data. The historical dynamic dataset includes water supply operation data of pipeline nodes, operation data of heat exchange stations, and heating operation environment data; The historical static dataset is preprocessed using blockchain-based encrypted fusion to obtain an optimized historical static dataset. The historical dynamic dataset is then used with a pre-defined spatiotemporal GAN ​​model to repair missing values, resulting in a historical dynamic optimized dataset.

3. The heating digital twin modeling method based on the U3D engine according to claim 2, characterized in that, The process involves constructing a heating digital twin model using the U3D engine and a pre-defined parametric component library, and initializing it based on the historical static optimization dataset to obtain an initial heating digital twin model and a LOD model set, including: Create a pre-defined parametric component library using the Unity 3D engine's ScriptableObject; A three-dimensional geometric model of the heating system is obtained by constructing a digital twin model of the heating system using the U3D engine and a pre-set parametric component library. The historical static optimization dataset is input into the heating three-dimensional geometric model for model initialization to obtain the initial heating digital twin model; The initial heating digital twin model is processed using the LOD rendering method to generate a set of LOD models, including LOD0, LOD1, LOD2 and LOD3 models.

4. The heating digital twin modeling method based on the U3D engine according to claim 3, characterized in that, The process of inputting the historical dynamic optimization dataset into the initial heating digital twin model for processing to obtain simulation datasets and equipment operation status evaluation data includes: The historical dynamic optimization dataset is used as a boundary condition input into the initial heating digital twin model. Through coupling simulation processing of the preset hydraulic model and thermal model in the model, simulation dataset and equipment operation status evaluation data are obtained. The simulation dataset includes simulated pressure, simulated flow rate, and simulated temperature. The equipment operation status assessment data includes pump efficiency, pipeline corrosion rate, and heat exchanger heat transfer coefficient.

5. The heating digital twin modeling method based on the U3D engine according to claim 4, characterized in that, The step of comparing the simulation dataset with the corresponding measured dataset and comparing the equipment operation status evaluation data with the corresponding measured equipment operation status data to obtain the model effectiveness evaluation parameters of the initial heating digital twin model includes: The simulation dataset is compared with the corresponding measured dataset to obtain the simulation pressure deviation rate, simulation flow deviation rate, and simulation temperature deviation rate, and then a weighted summation is performed to obtain the simulation data deviation rate. The equipment operation status assessment data is compared with the actual equipment operation status measurement data to obtain the pump group efficiency deviation rate, pipeline corrosion rate deviation rate and heat exchanger heat transfer coefficient deviation rate, and then weighted summation is performed to obtain the equipment operation assessment deviation rate. The simulation data deviation rate and the equipment operation evaluation deviation rate are weighted and summed to obtain the model effectiveness evaluation parameters of the initial heating digital twin model.

6. The heating digital twin modeling method based on the U3D engine according to claim 5, characterized in that, The step of comparing the model validity evaluation parameters with a preset model validity early warning threshold, and determining the model's qualification status based on the threshold comparison result, includes: Obtain heating type feature data, and query the preset heating type and model effectiveness warning threshold mapping table based on the heating type feature data to obtain the corresponding model effectiveness warning threshold; The model effectiveness evaluation parameters are compared with the model effectiveness early warning threshold. If the model validity assessment parameter is less than or equal to the model validity warning threshold, the initial heating digital twin model is deemed unqualified. If the model validity assessment parameter is greater than the model validity warning threshold, the initial heating digital twin model is deemed qualified.

7. The heating digital twin modeling method based on the U3D engine according to claim 6, characterized in that, Also includes: Obtain real-time environmental data within a preset range of the heating system; Acquire real-time simulation datasets, real-time equipment operating status evaluation data, historical simulation data for a preset time period, and corresponding execution feedback data of the heating system; Based on the real-time environmental data, real-time simulation dataset, real-time equipment operating status evaluation data, historical simulation data, and execution feedback data, feature engineering is performed to obtain a heating decision analysis dataset, including a time feature dataset, an operating feature dataset, and an environmental feature dataset. Based on the operational feature dataset and the preset historical fault data, the system is processed through a preset predictive maintenance model to obtain the fault probability value of the heating system. Based on the time feature dataset and environmental feature dataset, combined with the preset historical heat consumption and preset historical environmental feature data, heat load prediction data is obtained by processing the data through a preset heat load prediction model. The fault probability value and heat load prediction data are sent to the user terminal for display.

8. A heating digital twin modeling system based on the U3D engine, characterized in that, The system includes a memory and a processor. The memory contains a program for a heating digital twin modeling method based on the U3D engine. When the processor executes the program for the heating digital twin modeling method based on the U3D engine, it performs the following steps: Acquire multi-source heterogeneous data of the heating system within a preset time period, including historical static datasets and historical dynamic datasets, and preprocess them to obtain historical static optimized datasets and historical dynamic optimized datasets. A heating digital twin model is constructed using the U3D engine in conjunction with a pre-defined parametric component library, and initialized based on the historical static optimization dataset to obtain an initial heating digital twin model. The historical dynamic optimization dataset is input into the initial heating digital twin model for processing to obtain simulation dataset and equipment operation status evaluation data; The simulation dataset is compared with the corresponding measured dataset, and the equipment operation status evaluation data is compared with the corresponding measured equipment operation status data to obtain the model effectiveness evaluation parameters of the initial heating digital twin model. The model validity evaluation parameters are compared with the preset model validity warning threshold, and the model qualification status is determined based on the threshold comparison result.

9. The heating digital twin modeling system based on the U3D engine according to claim 8, characterized in that, The process involves acquiring multi-source heterogeneous data of the heating system within a preset time period, including historical static datasets, historical dynamic datasets, and historical operation monitoring datasets, and preprocessing this data to obtain historical static optimization datasets and historical dynamic optimization datasets, including: Acquire multi-source heterogeneous data of the heating system within a preset time period, including historical static datasets and historical dynamic datasets; The historical static dataset includes GIS map data, BIM model parameter data, and equipment operation and maintenance log data. The historical dynamic dataset includes water supply operation data of pipeline nodes, operation data of heat exchange stations, and heating operation environment data; The historical static dataset is preprocessed using blockchain-based encrypted fusion, and the historical dynamic dataset is repaired for missing values ​​using a preset spatiotemporal GAN ​​model to obtain the historical static optimized dataset and the historical dynamic optimized dataset.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a heating digital twin modeling method program based on the U3D engine. When the heating digital twin modeling method program based on the U3D engine is executed by a processor, it implements the steps of the heating digital twin modeling method based on the U3D engine as described in any one of claims 1 to 7.

Images