A dynamic fusion system and method for digital twin energy management BIM modeling

Abstract

This invention provides a dynamic fusion system and method for digital twin energy management BIM modeling, aimed at improving the intelligence level of energy management and optimizing energy dispatch. The system includes: data sensing nodes for collecting heterogeneous operational data from energy facilities; a local twin processor for real-time digital twin mapping of the data and generating dynamic feature vectors; regional twin nodes for local optimization based on edge computing and decentralized collaboration through blockchain or knowledge graphs; a cloud-based BIM fusion engine for constructing digital twin BIM models and generating self-optimizing energy dispatch schemes and long-term trend predictions; and an operation control terminal for presenting real-time views and providing control commands. This method combines multi-source data fusion, streaming aggregation optimization, and adaptive twin mapping technologies to improve the accuracy and dynamic response capabilities of energy management.

Description

Technical Field

[0001] This invention relates to digital twin technology, Building Information Modeling (BIM) technology, and energy management, specifically to a dynamic fusion system and method for digital twin energy management BIM modeling. The aim is to achieve efficient management, optimized scheduling, and intelligent decision-making of energy facilities through multi-source data fusion, edge computing, decentralized collaboration mechanisms, and intelligent optimization algorithms, thereby improving energy utilization efficiency, reducing operating costs, and enhancing real-time perception and control capabilities of the energy system. A dynamic fusion system and method for digital twin energy management BIM modeling. Background Technology

[0002] With the development of intelligent energy management and Building Information Modeling (BIM) technology, the application of digital twin technology in the field of energy management has gradually attracted attention. Traditional energy management systems mainly rely on static BIM models and preset optimization rules, making it difficult to perceive the real-time operating status of energy facilities and dynamically adjust optimization strategies, resulting in low energy utilization efficiency and high operation and maintenance costs. In addition, existing energy management methods often rely on centralized computing architectures, which suffer from latency in data transmission and processing, making it difficult to meet the real-time optimization needs of complex energy systems.

[0003] In recent years, the rapid development of multi-source data acquisition technology based on the Internet of Things (IoT), edge computing, blockchain, and artificial intelligence optimization algorithms has provided new technical means for energy management. However, existing technologies still have the following problems: (1) the data acquisition and fusion methods are singular, making it difficult to comprehensively and accurately perceive the dynamic operating status of energy facilities; (2) the lack of efficient local computing and collaborative optimization mechanisms leads to a lag in energy dispatch response; (3) the centralized architecture suffers from problems such as high computational pressure and low data security, making it difficult to adapt to the distributed management needs of large-scale energy systems.

[0004] To address the aforementioned issues, this invention proposes a dynamic fusion system for energy management BIM modeling based on digital twins. Through data perception, edge computing, decentralized collaborative optimization, and cloud-based intelligent analysis, it achieves real-time perception, dynamic optimization, and intelligent decision-making in energy management, thereby improving energy utilization, reducing energy costs, and enhancing the system's robustness and adaptability. Summary of the Invention

[0005] In summary, this invention provides a dynamic fusion system and method for digital twin energy management BIM modeling, aiming to solve problems such as incomplete data collection, delayed optimization response, and single computing architecture in existing energy management systems, and to achieve efficient and intelligent management of energy facilities.

[0006] According to the data sensing node provided by the inventor, deployed around energy facilities, a multi-source sensor array is configured to collect and output heterogeneous operational data of the energy facilities in real time. A local twin processor, coupled with the data sensing node, receives the heterogeneous operational data and performs real-time digital twin mapping to generate dynamic feature vectors. A regional twin node is interconnected with multiple local twin processors, performs local energy optimization based on edge computing, and collaborates with other regional twin nodes through a decentralized collaboration mechanism, wherein the decentralized collaboration mechanism uses blockchain or knowledge graph to achieve distributed decision-making. A cloud-based BIM fusion engine, interconnected with the regional twin nodes, receives data transmitted by the regional twin nodes, constructs a digital twin BIM model, generates a self-optimizing energy scheduling scheme and long-term trend prediction, and sends the results back to the regional twin nodes. An operation control terminal, interconnected with the cloud-based BIM fusion engine, presents a real-time view of the digital twin BIM model, energy anomaly alerts, and scheduling suggestions, and supports users issuing control commands.

[0007] Optionally, the data sensing node performs primary feature extraction on the collected heterogeneous operating data to generate an energy state vector, wherein the energy state vector is calculated using the following formula:

[0008]

[0009] Where (Q1, Q2, …, Q m (k1, k2, …, k) represents the heterogeneous operational data components acquired by the multi-source sensor array. m ) represents the corresponding dynamic calibration coefficient, which is adaptively updated based on the operating mode of the energy facility and satisfies the normalization condition.

[0010] Optionally, the local twin processor incorporates an adaptive twin mapping algorithm to generate the dynamic feature vector and employs a self-organizing hierarchical strategy to determine data hierarchy based on the dynamic feature vector, wherein key data is transmitted through a priority channel and regular data is transmitted through a secondary channel after compression.

[0011] Optionally, the system further includes a streaming aggregation relay connected to multiple local twin processors via a distributed network. This relay receives the compressed regular data transmitted by the local twin processors through a secondary channel, performs streaming clustering on the compressed regular data, and forwards it to the regional twin node. Meanwhile, the critical data is directly transmitted by the local twin processors to the regional twin node through a priority channel.

[0012] Optionally, the streaming aggregation relay performs semantic layering and transmission priority evaluation on the received compressed regular data, forwards highly relevant data to the regional twin node in real time, and buffers low-relevance data locally to reduce network load.

[0013] Optionally, the streaming aggregation relay constructs an adaptive feedback network with multiple local twin processors based on the local optimization results generated by the regional twin nodes, supporting the hierarchical push of the local optimization results to the corresponding local twin processors according to the real-time demand of energy facilities.

[0014] Optionally, the cloud-based BIM fusion engine calculates the energy dynamic index D based on the fused data and generates the long-term trend prediction accordingly, wherein the energy dynamic index D is calculated using the following formula:

[0015]

[0016] Wherein, R is the real-time response value of the heterogeneous operation data, R0 is the reference response value, T is the current model extrapolation value, T0 is the baseline extrapolation value, and α is the sensitivity factor, which is adjusted in the range of [0.8, 2.5] according to the type of energy facility.

[0017] Optionally, the system further includes a real-time optimization unit coupled to the local twin processor, which dynamically adjusts the operating parameters of the energy facility based on the local optimization results of the regional twin node to achieve an optimal balance in energy utilization.

[0018] Optionally, the cloud-based BIM fusion engine sends parameter calibration instructions or model update instructions to the data sensing node through the regional twin node to support online performance improvement of the energy facility and reduce operational losses.

[0019] According to the present invention, a dynamic fusion method for digital twin energy management BIM modeling includes the following steps:

[0020] (1) By deploying data sensing nodes around the energy facility, a multi-source sensor array is used to collect and output the heterogeneous operation data of the energy facility in real time;

[0021] (2) Receive the heterogeneous operating data using the local twin processor, perform real-time digital twin mapping using an adaptive twin mapping algorithm, and generate dynamic feature vectors;

[0022] (3) Utilize regional twin nodes to receive data transmitted by the local twin processor, perform local energy optimization based on edge computing, and collaborate with other regional twin nodes through a decentralized collaboration mechanism, wherein the decentralized collaboration mechanism uses blockchain or knowledge graph to achieve distributed decision-making;

[0023] (4) Utilize the cloud-based BIM fusion engine to receive data transmitted from the twin nodes in the region, construct a digital twin BIM model using a multi-scale dynamic fusion model, perform global energy status simulation, and generate a self-optimizing energy dispatching scheme and long-term trend prediction.

[0024] (5) The self-optimizing energy dispatch scheme and long-term trend prediction generated by the cloud BIM fusion engine are transmitted back to the regional twin node;

[0025] (6) Receive the self-optimizing energy scheduling scheme and long-term trend prediction returned by the cloud BIM fusion engine through the operation control terminal, present the real-time view of the digital twin BIM model, energy anomaly prompts and scheduling suggestions, and support users to issue control commands to the cloud BIM fusion engine.

[0026] This invention improves the intelligence of energy management through digital twin technology, edge computing, and decentralized collaborative optimization, realizing real-time data fusion, dynamic optimization scheduling, and adaptive feedback control. It is suitable for the refined management of large-scale energy systems and has high application value. Attached Figure Description

[0027] Figure 1 This invention provides a schematic diagram of a dynamic fusion system architecture for BIM modeling of a digital twin energy management system.

[0028] Figure 2 A flowchart of a dynamic fusion method for BIM modeling of a digital twin energy management system provided by the present invention.

[0029] Figure 3 This is a schematic diagram of a flow aggregation relay provided by the present invention. Detailed Implementation

[0030] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0031] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature; in the description of this application, unless otherwise stated, "multiple" means two or more.

[0032] To achieve the above objectives, please refer to Figure 1 and Figure 2 As shown, this invention provides a dynamic fusion system for digital twin energy management BIM modeling, applicable to various scenarios. For example, in intelligent building energy management, data sensing nodes are deployed in the building's heating, ventilation, air conditioning (HVAC), lighting, and power distribution systems, configured with multi-source sensor arrays, including temperature, current, and light sensors, collecting data once per second to monitor the building's energy consumption and environmental status in real time. Local twin processors are installed in the computer room on each floor, coupled with the data sensing nodes, receiving heterogeneous operational data and generating dynamic feature vectors through real-time digital twin mapping. For example, based on current temperature and power consumption, a feature vector reflecting the energy efficiency of that floor is generated. Regional twin nodes are deployed in the building's central control room, interconnected with the local twin processors on each floor. They optimize the HVAC operating parameters of each floor based on edge computing and record energy allocation decisions through blockchain to ensure data transparency and immutability. A cloud-based BIM fusion engine runs on a cloud server, receiving data from the regional twin nodes, constructing a digital twin BIM model of the building, predicting energy consumption trends during peak summer periods based on historical data, and generating a scheduling scheme for automatically adjusting air conditioning power. The operation control terminal presents a real-time view via a tablet computer, displaying energy consumption anomalies on each floor (such as lights not being turned off on a certain floor) and optimization suggestions (such as reducing air conditioning power during non-working hours), supporting building administrators in issuing control commands.

[0033] Furthermore, the system is extended to an industrial park encompassing solar power plants, wind turbines, and energy storage systems. Data sensing nodes are equipped with sensor arrays on solar panels, wind turbines, and energy storage batteries, collecting heterogeneous operational data such as light intensity, wind speed, and battery voltage, updating every minute. Local twin processors are deployed alongside each type of energy equipment, receiving data and generating dynamic feature vectors, such as generating an operational efficiency vector for wind turbines based on wind speed and power generation. Multiple regional twin nodes are set up within the park, each responsible for data processing in the solar, wind, and energy storage zones respectively. They perform local optimization through edge computing and utilize knowledge graphs to enable collaboration between nodes, dynamically adjusting power distribution (e.g., storing excess solar energy in batteries). A cloud-based BIM fusion engine receives data from each zone, constructs a digital twin BIM model of the entire park, predicts the energy supply and demand balance for the next 24 hours, and generates scheduling plans (e.g., prioritizing energy storage power at night). The operation control terminal provides the park's energy management team with a real-time view through desktop software, detecting anomalies (e.g., decreased efficiency of a wind turbine) and supporting adjustments to equipment operating parameters based on recommendations.

[0034] Furthermore, the system is applied to the optimization of a smart grid in an urban area. Data sensing nodes deploy sensors at substations, distribution lines, and residential electricity terminals to collect data such as voltage, current, and load, updating every 10 seconds. Local twin processors are installed in each substation to process heterogeneous data within their jurisdiction, generating dynamic feature vectors reflecting the distribution of grid load. Regional twin nodes cover different areas of the city (such as residential and industrial areas), optimizing local grid loads based on edge computing and enabling distributed decision-making for inter-regional electricity transactions through blockchain (e.g., industrial areas selling surplus electricity to residential areas). A cloud-based BIM fusion engine integrates city-wide data to construct a digital twin BIM model of the power grid, predicting peak-hour electricity demand and generating self-optimizing dispatch schemes (e.g., adjusting substation output power). The operation control terminal displays the real-time grid status on a large screen in the operations center, alerts to anomalies (e.g., overload of a line), and supports engineers in issuing control commands to balance the load.

[0035] It is understandable that this invention, through digital twin technology, edge computing, and decentralized collaborative optimization, improves the intelligence level of energy management, achieving real-time data fusion, dynamic optimization scheduling, and adaptive feedback control. It is suitable for the refined management of large-scale energy systems and has high application value. Specifically, the system achieves real-time data acquisition and mapping through a multi-source sensor array and a local twin processor, significantly improving the accuracy and timeliness of energy status monitoring. Edge computing and decentralized collaborative mechanisms (such as blockchain or knowledge graphs) of regional twin nodes optimize local resource allocation, enhancing the system's adaptability and collaborative efficiency in complex scenarios. The cloud-based BIM fusion engine provides a global perspective and long-term trend prediction, providing a scientific basis for energy scheduling. Furthermore, the intuitive interactive design of the operation control terminal further enhances the flexibility of user control, enabling the system to not only automatically optimize operation but also meet the needs of manual intervention. This multi-layered and multi-dimensional technological integration effectively reduces energy loss and improves resource utilization efficiency, making it particularly suitable for diverse scenarios such as buildings, industrial parks, and urban power grids, driving the development of energy management towards intelligence and refinement.

[0036] It is understood that the dynamic fusion system of digital twin energy management BIM modeling can be applied to a variety of scenarios. The following uses an intelligent building energy management system as an example, but it does not limit the present invention.

[0037] In some embodiments, data sensing nodes are deployed in the building's HVAC system, lighting system, and power distribution system, configured with multi-source sensor arrays, including temperature sensors, current sensors, and light sensors, to collect heterogeneous operational data once per second, such as indoor temperature, lighting current, and power distribution on each floor. The data sensing nodes perform primary feature extraction on the collected heterogeneous operational data to generate an energy state vector. For example, for the HVAC system on a certain floor, the sensor collects a temperature Q1 = 25°C. ∘ C. Current Q2 = 10A, combined with dynamic calibration coefficients k1 = 0.6 (based on the current cooling mode) and k2 = 0.4 (based on power consumption weight), using the formula:

[0038] = =15.52;

[0039] The energy state vector V = 15.52 for this floor is calculated, reflecting its energy consumption status. The dynamic calibration coefficients are adaptively updated according to the building's operating mode (e.g., cooling or heating). For example, in heating mode, k1 may be adjusted to 0.7, and all coefficients satisfy the normalization condition (i.e., ∑k... i =1).

[0040] Furthermore, in the lighting system, the data sensing node collects illuminance Q1=300 lux and current Q2=5A, with dynamic calibration coefficients k1=0.3 (based on the influence of natural light) and k2=0.7 (based on electricity priority), respectively, and calculates the energy state vector: = =190.14; generates V=90.14, reflecting the energy status of the lighting system. The dynamic calibration coefficient is adaptively adjusted according to the daytime or nighttime mode and optimized under normalization constraints.

[0041] Furthermore, in the power distribution system, the data sensing node collects voltage Q1=220V and load current Q2=20A, with dynamic calibration coefficients k1=0.5 and k2=0.5 (based on balancing weights), calculates the energy state vector, and generates V=110.45V, reflecting the operating status of the power distribution system. These coefficients are adaptively updated according to peak or off-peak load periods to ensure data accuracy.

[0042] Understandably, the above methods further enhance the precision and intelligence of energy management. Specifically, data sensing nodes collect heterogeneous operational data in real time through multi-source sensor arrays and utilize formulas... The generation of energy state vectors enables precise quantification of the operational status of energy facilities. An adaptive update mechanism for dynamic calibration coefficients can flexibly adjust weights according to different operating modes (such as cooling, heating, or peak periods), ensuring that the energy state reflected in the vector is highly consistent with actual demand. This method not only improves the real-time performance and accuracy of data processing but also provides reliable foundational data for subsequent digital twin mapping and optimized scheduling, reduces energy waste caused by data bias, and enhances the system's applicability in complex scenarios, thus providing efficient support for the refined management of large-scale energy systems.

[0043] In some embodiments, a local twin processor is installed in a server room on each floor, coupled to the building's data sensing nodes, to receive heterogeneous operational data from heating, ventilation, and air conditioning (HVAC) systems, lighting systems, and power distribution systems. The data sensing nodes collect data once per second, including temperature (e.g., 25°C), current (e.g., 10A), illuminance (e.g., 300 lux), and voltage (e.g., 220V). The local twin processor incorporates an adaptive twin mapping algorithm to process this data and generate dynamic feature vectors reflecting the energy operating status of each floor. For example, for the HVAC system on the 5th floor, the input data is temperature Q1 = 25°C and current Q2 = 10A. The algorithm assigns weights based on the current cooling mode (e.g., w1 = 0.6, w2 = 0.4) and calculates the dynamic feature vector components: V5 = [w1 ⋅ Q1, w2 ⋅ Q2 ... 2⋅Q2]=[0.6⋅25,0.4⋅10]=[15,4], generating a dynamic feature vector V5=[15,4]. Similarly, the lighting system of the 10th layer is given by input light intensity Q1=300lux and current Q2=5A, with weights adjusted to w1=0.3 and w2=0.7 (due to the influence of natural daylight), generating V 10 =[90,3.5].

[0044] Specifically, the local twin processor employs a self-organizing hierarchical strategy, processing data hierarchically based on the characteristics of the dynamic feature vector. For example, during peak working hours (9:00-11:00 AM), the dynamic feature vector V5=[15,4] of the 5th layer HVAC has a larger magnitude ( =15.52), and the temperature exceeds the comfortable range (set 22°C—24°C), which is considered critical data; while the V10 of the 10th floor lighting system is [90, 3.5] ( =90.06) is classified as routine data due to normal energy consumption. Critical data is transmitted directly to the regional twin node in the building's central control room via a priority channel, with transmission latency controlled within 50 milliseconds to ensure real-time response; routine data is compressed by the local twin processor (for example, compressed to 60% of its original size using the Lempel-Ziv algorithm) and transmitted via a secondary channel, with a latency of approximately 200 milliseconds.

[0045] Furthermore, the adaptive twin mapping algorithm dynamically adjusts based on the building's real-time status. For example, when the outside temperature rises to 35°C, the weight w1 of the HVAC system increases from 0.6 to 0.8, reflecting the greater impact of temperature on energy consumption, and V5=[20,4] is regenerated. Simultaneously, the self-organizing hierarchical strategy adjusts the threshold based on the building's peak electricity consumption (e.g., a total load of 500kW at 10:00 AM), and dynamically assigns feature vectors with moduli exceeding 20 (such as the new moduli of V5). = 20.4) is prioritized as key data to ensure efficient processing during peak periods. In addition, the local twin processor also records hourly data hierarchical statistics. For example, the 5th layer generates 1,000 sets of dynamic feature vectors between 9:00 and 10:00 am, of which 20% are key data and 80% are regular data, providing a basis for subsequent optimization.

[0046] It is understood that the local twin processor described in this embodiment, along with its built-in adaptive twin mapping algorithm and self-organizing hierarchical strategy, significantly improves the intelligence and efficiency of energy management. Specifically, the local twin processor generates dynamic feature vectors in real time through the adaptive twin mapping algorithm. It can dynamically adjust the mapping weights based on the operating status of energy facilities (such as temperature changes or peak electricity consumption), ensuring that the feature vectors accurately reflect current energy consumption characteristics and providing precise data support for subsequent optimization. The self-organizing hierarchical strategy intelligently classifies the dynamic feature vectors, prioritizing the rapid transmission of key data (such as abnormal energy consumption or peak load) through a priority channel, ensuring real-time performance. Regular data, after compression, is transmitted through a secondary channel, effectively reducing network bandwidth consumption and improving data processing efficiency. This hierarchical transmission mechanism not only optimizes resource allocation but also reduces system latency, enabling building energy management to quickly respond to anomalies and implement control measures in complex and ever-changing environments.

[0047] In some embodiments, such as Figure 3 As shown, the streaming aggregation relay connects to local twin processors on all floors via the building's distributed network, responsible for receiving compressed routine data transmitted through secondary channels. For example, during peak hours from 9:00 AM to 10:00 AM, the streaming aggregation relay receives approximately 800 sets of routine data per minute (such as compressed data from the 10th-floor lighting system), each set being approximately 0.6KB in size. It employs a streaming clustering algorithm (based on online K-means) to cluster this data in real time into categories such as "normal lighting energy consumption" or "low-load power distribution." Taking the lighting data from the 10th and 12th floors as an example, the streaming aggregation relay clusters the two similar sets of data into one class, generates cluster centers, and forwards them in batches to the regional twin node every 5 seconds. Simultaneously, critical data from the 5th-floor HVAC system is transmitted directly to the regional twin node via a priority channel, bypassing the streaming aggregation relay, ensuring real-time performance.

[0048] Specifically, when the external environment changes (such as increased light intensity at 10:00 AM), the streaming aggregation relay dynamically adjusts its clustering parameters, re-integrates the updated regular data, and forwards it. If the network is interrupted for 10 seconds, the streaming aggregation relay can cache up to 500 sets of data and send them in order after recovery; if an anomaly is detected in regular data (such as a sudden increase in current at a certain layer), it is converted into critical data and transmitted directly through a priority channel. This mechanism ensures efficient processing and reliable transmission of the data stream.

[0049] Furthermore, the streaming aggregation relay receives compressed routine data transmitted via secondary channels from local twin processors on each floor through the building's distributed network. During peak hours of 9:00-10:00 AM, approximately 800 sets of routine data are received per minute. This data is semantically layered based on energy type and operating status, categorized into layers such as "Lighting Energy Consumption," "HVAC Low Load," and "Power Distribution Stability." For example, routine data from the 10th-floor lighting system (compressed to reflect illumination and current status) is categorized into the "Lighting Energy Consumption" layer, while HVAC data from the 3rd floor, operating at low power, is categorized into the "HVAC Low Load" layer. Next, the streaming aggregation relay performs transmission priority assessment: if the 10th-floor data indicates normal illumination but slightly high current (e.g., exceeding the average by 10%), it is prioritized as "High Relevance" and immediately forwarded to the regional twin node to optimize lighting adjustments; the 3rd-floor HVAC data, due to minimal changes, is prioritized as "Low Relevance" and sent in batches after a 5-second local buffer to reduce network load.

[0050] Specifically, when the building's operational status changes (e.g., increased external light at 10:00 AM), the streaming aggregation relay dynamically updates the semantic hierarchy. Data from the 10th floor, despite significant changes in lighting, remains highly relevant and is transmitted immediately, while low-load data from other floors continues to be buffered. If regular data from a certain floor becomes abnormal (e.g., a sudden 50% increase in lighting current on the 8th floor), the streaming aggregation relay prioritizes it and forwards it immediately. This hierarchy and evaluation mechanism optimizes data transmission efficiency.

[0051] Understandably, this embodiment further enhances the efficiency and intelligence of the energy management system. Streaming aggregation relay clearly distinguishes different energy types and operating states by semantically layering regular data, providing more targeted data support for regional twin nodes; transmission priority assessment ensures that highly relevant data (such as anomalies or critical changes) is transmitted in a timely manner, while low-relevance data is buffered locally, effectively optimizing network resource utilization and reducing bandwidth pressure, especially ensuring data fluency in dynamic environments or peak periods.

[0052] In some embodiments, the regional twin node generates local optimization results based on high energy consumption data of the HVAC system on the 5th floor (e.g., a temperature of 28°C exceeding the comfort range), suggesting an increase in air conditioning power by 20%; simultaneously, the lighting system on the 10th floor is advised to reduce brightness by 10% due to sufficient lighting. These optimization results are fed back from the regional twin node to the streaming aggregation relay. The streaming aggregation relay constructs an adaptive feedback network with the local twin processors on each floor based on the results, and by analyzing the building's real-time demand (e.g., current floor occupancy and an outside temperature of 35°C), it pushes the optimization results into two levels: "urgent" and "normal." The air conditioning adjustment suggestion on the 5th floor is marked as "urgent" because it involves comfort and is pushed to the local twin processor on the 5th floor via a high-speed feedback channel (latency less than 50 milliseconds); the lighting adjustment suggestion on the 10th floor is marked as "normal" because its impact is relatively small and is pushed via a secondary channel (latency approximately 200 milliseconds).

[0053] Specifically, when the building status changes (e.g., the outside temperature drops to 30°C at 10:30), the regional twin node updates the optimization results, reducing the air conditioning power increase on the 5th floor by 10%. The streaming aggregation relay dynamically adjusts the feedback network, reassessing the push priority to ensure that the 5th floor still receives update instructions first. This hierarchical push mechanism ensures the targeted and timely nature of the feedback.

[0054] In some embodiments, the cloud-based BIM fusion engine runs on a cloud server and receives fusion data transmitted from regional twin nodes, including the operational status of HVAC, lighting, and power distribution systems on each floor. At 10:00 AM, the engine's real-time response value R=28 is based on the HVAC system on the 5th floor. ∘ C (current temperature), reference response value R0 = 24 ∘ The energy dynamic index D is calculated using C (comfort temperature), the current model-estimated value T = 30kW (current energy consumption), and the baseline estimated value T0 = 25kW (historical average energy consumption). The building's HVAC system is a high-sensitivity device with a sensitivity factor α = 2.0 (adjusted within the range of [0.8, 2.5]), determined using the formula:

[0055] = =9.79;

[0056] The calculated D=9.79 indicates that the HVAC energy consumption on the 5th floor is relatively high. Similarly, the lighting system on the 10th floor has R=5A, R0=4A, T=2kW, T0=1.8kW, and α=1.0 (low lighting sensitivity), resulting in a calculated D=1.25⋅1.246=1.56, indicating normal operation.

[0057] The cloud-based BIM fusion engine uses these indices, combined with historical data (energy consumption records for the past three months) and external variables (such as weather forecasts for the next week), to predict long-term building trends. For example, it predicts that HVAC energy consumption on the 5th floor will increase by 15% during the summer peak season (July-August), while the lighting system will only increase by 5% due to increased natural light. These predictions are fed back to the regional twin nodes to adjust long-term scheduling strategies.

[0058] Understandably, the cloud-based BIM fusion engine's ability to calculate energy dynamic indices and generate long-term trend forecasts further enhances the scientific rigor and foresight of energy management. The cloud-based BIM fusion engine utilizes formula D... Quantifying energy dynamic indices accurately reflects the real-time operating status of facilities, and adjusting sensitivity factors according to equipment type ensures the applicability of the indices. This quantitative method provides a reliable basis for long-term trend forecasting, enabling the system to identify energy consumption trends in advance based on historical data and external conditions, such as predicting peak demand growth. The feedback of forecast results supports regional node optimization scheduling strategies, reducing energy waste and equipment overload risks, enhancing the system's planning capabilities in complex environments, and providing crucial support for the refined management and sustainable development of large-scale energy systems.

[0059] In some embodiments, the real-time optimization unit is coupled to a local twin processor in each floor's server room, receiving local optimization results generated by the regional twin node. At 10:00 AM, the HVAC system on the 5th floor triggers an optimization suggestion from the regional twin node due to a temperature rise to 28°C (exceeding the comfort range of 24°C): increase the air conditioning power by 20% (from 50kW to 60kW). Based on this suggestion, and combined with the building's real-time status (e.g., current floor occupancy 80%, outside temperature 35°C), the real-time optimization unit dynamically adjusts the HVAC operating parameters. It lowers the cooling temperature setting from 25°C to 23°C and increases the fan speed by 10% (from 1000rpm to 1100rpm) to quickly restore comfort while avoiding excessive energy consumption. After the adjustment, the temperature on the 5th floor drops to 24°C within 15 minutes, and the energy consumption stabilizes at 58kW, achieving an optimal balance in utilization.

[0060] Specifically, when the outside temperature drops to 30°C at 10:30, the regional twin node updates its optimization results, suggesting a reduction in power increase to 10% (55kW). The real-time tuning unit responds quickly, adjusting the fan speed back to 1050rpm and fine-tuning the cooling temperature to 24°C to ensure a dynamic balance between energy efficiency and comfort. Throughout the process, the real-time tuning unit records the adjustment effects and feeds them back to the local twin processor, providing data support for subsequent optimizations.

[0061] In some embodiments, the cloud-based BIM fusion engine, based on long-term operational data analysis, discovered that the temperature sensor at the data sensing node of the 5th-layer HVAC system consistently exceeded the reading by 2°C in a high-temperature environment (outdoor temperature 35°C), affecting the accuracy of the dynamic feature vector. At 10:00 AM, the engine generated a parameter calibration command, which was sent to the 5th-layer data sensing node via the regional twin node. The command adjusted the calibration coefficient of the temperature sensor from 1.0 to 0.95, making the collected values ​​closer to the actual temperature (e.g., calibrating from 28°C to 26.6°C). After calibration, HVAC operating efficiency improved, reducing overcooling caused by misjudgment, and power consumption decreased from 60kW to 55kW.

[0062] Meanwhile, the cloud-based BIM fusion engine optimized the digital twin model based on energy consumption trends over the past week (e.g., decreased efficiency of the lighting system under low nighttime load), updating the prediction algorithm for the 10th-layer lighting system (upgrading from linear regression to an exponential model with time decay). At 10:30 AM, the engine sent a model update command to the 10th-layer data sensing node through the regional twin node, requesting the sensors to increase the sampling frequency of light changes (from once per second to twice per second) to capture subtle needs for nighttime lighting adjustments. After the update, the lighting system reduced ineffective energy consumption by 10% under low load, significantly reducing operational losses.

[0063] In some embodiments, such as Figure 2 As shown, a dynamic fusion method for BIM modeling of digital twin energy management is also provided, the specific steps of which are as follows:

[0064] (1) The data sensing node collects and outputs heterogeneous operating data in real time. At 10:00 AM, the data sensing node of the 5th floor HVAC system collected data of 28°C and 12A through temperature and current sensors and output it once per second.

[0065] (2) The local twin processor receives data and generates dynamic feature vectors. The fifth-layer local twin processor uses an adaptive twin mapping algorithm to map temperature and current data into dynamic feature vectors V5=[16.8,4.8] (based on weight adjustment), reflecting the current energy consumption status.

[0066] (3) Regional twin nodes perform local energy optimization and collaboration. The regional twin nodes receive data from the 5th layer, analyze the excessive temperature based on edge computing, generate optimization suggestions (increase air conditioning power by 15%), and collaborate with other floor nodes through blockchain to confirm the feasibility of power allocation.

[0067] (4) The cloud-based BIM fusion engine builds a model and extrapolates the global status. The cloud receives regional data and uses a multi-scale dynamic fusion model to build a digital twin BIM model of the building, extrapolates the global energy consumption, finds that the 5th floor needs to be adjusted first, and generates a self-optimizing scheduling scheme (power increased to 57.5kW) and a long-term prediction (energy consumption increased by 10% in summer).

[0068] (5) Optimize the plan and forecasts and send them back to the regional twin node. The cloud sends back the scheduling plan and forecast results at 10:05, and the regional twin node is ready to execute and distribute them to the local processor.

[0069] (6) The operation control terminal presents a view and supports adjustment. The administrator views the real-time BIM model through a tablet computer, sees the temperature anomaly prompt and scheduling suggestion on the 5th floor (power increased to 57.5kW), manually confirms the command, and the cloud immediately coordinates the execution. At 10:15, the temperature dropped to 24°C.

[0070] Understandably, this method achieves closed-loop management of the entire process from data acquisition to optimized execution through real-time data acquisition from data-sensing nodes, dynamic mapping of local twin processors, edge optimization and collaboration of regional twin nodes, global simulation by the cloud-based BIM fusion engine, and interactive control via the operation control terminal. The close integration of each step ensures the effective implementation of real-time data fusion, dynamic optimization scheduling, and adaptive feedback. For example, it can quickly respond to temperature anomalies and optimize air conditioning power, ensuring comfort while avoiding waste. Furthermore, this method supports user participation in decision-making, enhancing flexibility and making it suitable for the refined management of large-scale energy systems. It possesses practical value in terms of high energy efficiency and wide applicability.

[0071] The above description is merely an exemplary embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural transformations made using the contents of the present invention's specification and drawings under the technical concept of the present invention, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. A dynamic fusion system for digital twin energy management BIM modeling, characterized in that, include: Data sensing nodes are deployed around energy facilities and equipped with multi-source sensor arrays to collect and output heterogeneous operating data of the energy facilities in real time. The data sensing node performs primary feature extraction on the collected heterogeneous operating data to generate an energy state vector, wherein the energy state vector is calculated using the following formula: Where (Q1,Q2,…,Q) m (k1,k2,…,k) represents the heterogeneous operational data components acquired by the multi-source sensor array. m ) represents the corresponding dynamic calibration coefficient, which is adaptively updated based on the operating mode of the energy facility and satisfies the normalization condition; A local twin processor, coupled to the data sensing node, is used to receive the heterogeneous operating data and perform real-time digital twin mapping to generate dynamic feature vectors. A regional twin node is interconnected with multiple local twin processors, performs local energy optimization based on edge computing, and collaborates with other regional twin nodes through a decentralized collaboration mechanism, wherein the decentralized collaboration mechanism uses blockchain or knowledge graph to achieve distributed decision-making. A cloud-based BIM fusion engine, interconnected with the regional twin nodes, is used to receive data transmitted by the regional twin nodes, construct a digital twin BIM model, calculate the energy dynamic index D based on the fused data, and generate a long-term trend forecast accordingly. The energy dynamic index D is calculated using the following formula: Wherein, R is the real-time response value of the heterogeneous operation data, R0 is the reference response value, T is the current model extrapolation value, T0 is the baseline extrapolation value, and α is the sensitivity factor. The sensitivity factor α is adjusted in the range of [0.8, 2.5] according to the type of energy facility to generate a self-optimizing energy dispatch scheme and long-term trend prediction, and the results are sent back to the regional twin node. The operation control terminal is interconnected with the cloud-based BIM fusion engine to present a real-time view of the digital twin BIM model, energy anomaly alerts, and scheduling suggestions, and supports users in issuing control commands.

2. The dynamic fusion system for digital twin energy management BIM modeling according to claim 1, characterized in that, The local twin processor has a built-in adaptive twin mapping algorithm to generate the dynamic feature vector, and adopts a self-organizing hierarchical strategy to determine the data hierarchy based on the dynamic feature vector. Key data is transmitted through the priority channel, while regular data is compressed and transmitted through the secondary channel.

3. The dynamic fusion system for digital twin energy management BIM modeling according to claim 2, characterized in that, The system also includes a streaming aggregation relay, which is connected to multiple local twin processors via a distributed network. It is used to receive the compressed regular data transmitted by the local twin processors through a secondary channel, and to perform streaming clustering on the compressed regular data before forwarding it to the regional twin node. Meanwhile, the critical data is directly transmitted by the local twin processors to the regional twin node through a priority channel.

4. The dynamic fusion system for digital twin energy management BIM modeling according to claim 3, characterized in that, The streaming aggregation relay performs semantic layering and transmission priority evaluation on the received compressed regular data, forwards highly relevant data to the regional twin node in real time, and buffers low-relevance data locally to reduce network load.

5. The dynamic fusion system for digital twin energy management BIM modeling according to claim 4, characterized in that, The streaming aggregation relay constructs an adaptive feedback network with multiple local twin processors based on the local optimization results generated by the regional twin nodes, supporting the hierarchical push of the local optimization results to the corresponding local twin processors according to the real-time demand of energy facilities.

6. The dynamic fusion system for digital twin energy management BIM modeling according to claim 1, characterized in that, The system also includes a real-time optimization unit coupled to the local twin processor, which dynamically adjusts the operating parameters of the energy facility based on the local optimization results of the regional twin node to achieve an optimal balance in energy utilization.

7. The dynamic fusion system for digital twin energy management BIM modeling according to claim 1, characterized in that, The cloud-based BIM fusion engine sends parameter calibration instructions or model update instructions to the data sensing nodes through the regional twin nodes to support the online performance improvement of the energy facilities and reduce operational losses.

8. A dynamic fusion method for digital twin energy management BIM modeling based on the system of claim 1, characterized in that, Includes the following steps: (1) By deploying data sensing nodes around the energy facility, a multi-source sensor array is used to collect and output the heterogeneous operation data of the energy facility in real time; (2) Receive the heterogeneous running data using the local twin processor, perform real-time digital twin mapping using an adaptive twin mapping algorithm, and generate dynamic feature vectors; (3) Utilize regional twin nodes to receive data transmitted by the local twin processor, perform local energy optimization based on edge computing, and collaborate with other regional twin nodes through a decentralized collaboration mechanism, wherein the decentralized collaboration mechanism uses blockchain or knowledge graph to achieve distributed decision-making; (4) Utilize the cloud-based BIM fusion engine to receive data transmitted from the twin nodes in the region, construct a digital twin BIM model using a multi-scale dynamic fusion model, and perform global energy status simulation to generate a self-optimizing energy dispatching scheme and long-term trend prediction. (5) The self-optimizing energy dispatch scheme and long-term trend prediction generated by the cloud BIM fusion engine are transmitted back to the regional twin node; (6) Receive the self-optimizing energy scheduling scheme and long-term trend prediction returned by the cloud BIM fusion engine through the operation control terminal, present the real-time view of the digital twin BIM model, energy anomaly prompts and scheduling suggestions, and support users to issue control commands to the cloud BIM fusion engine.

Images