Building energy consumption management system based on BIM building

Abstract

The invention discloses a building energy consumption management system based on a BIM building. The system comprises a BIM model construction module, a data acquisition module, an AI feedback and decision module, a strategy output module and a self-learning optimization module. And the BIM model construction module is used for acquiring building design data and constructing a BIM model comprising energy consumption monitoring nodes. The invention relates to the technical field of building energy consumption management. According to the building energy consumption management system based on the BIM building, building design data, structure parameters and energy consumption monitoring node space are mapped through the BIM model building module, accurate association of building components and energy consumption data is achieved, and a unified energy consumption data space model is formed; the AI feedback and decision module can predict a future energy consumption trend based on an environment-energy consumption correlation model and a neural network prediction model; and in combination with a reinforcement learning algorithm, a multi-target energy efficiency optimization strategy is automatically generated, and cooperative energy-saving control of air conditioning, illumination, ventilation and equipment operation is realized.

Description

Technical Field

[0001] This invention relates to the field of building energy management, and more specifically, to a building energy management system based on BIM (Building Information Modeling). Background Technology

[0002] With the acceleration of urbanization and the expansion of building scale, building energy consumption has become a significant component of social energy consumption. Statistics show that building energy consumption accounts for more than 30% of total energy consumption. Traditional building energy management mainly relies on manual statistics and experience-based judgment, which suffers from problems such as scattered energy consumption data, poor real-time performance, delayed decision-making, and a lack of scientific basis for energy-saving strategies, making it difficult to achieve refined and intelligent management of building energy efficiency.

[0003] In recent years, the rise of BIM technology has provided a unified data carrier for the entire life cycle management of buildings, enabling the integration of information from the building design, construction, and operation and maintenance phases in three-dimensional space. However, existing BIM applications mostly remain at the design and visualization stage, lacking deep integration with energy consumption data during the building operation phase, and failing to form a closed-loop management system of "design-monitoring-feedback-optimization".

[0004] Meanwhile, while artificial intelligence algorithms have demonstrated significant advantages in energy consumption prediction and optimization control, most AI models still rely on offline data training and are not integrated with the real-time operating environment of buildings. This makes it difficult to automatically generate energy-saving strategies based on dynamic changes in the building's internal environment and external climate. Furthermore, existing systems generally lack self-learning mechanisms, making it impossible to continuously correct and optimize models based on historical performance, leading to performance degradation and insufficient adaptability after long-term operation.

[0005] Therefore, there is an urgent need for a building energy management system that can integrate BIM models with multi-source energy consumption data, AI intelligent decision-making, and self-learning optimization technology to achieve real-time perception, intelligent analysis, and dynamic optimization control of building energy consumption. Summary of the Invention

[0006] The purpose of this invention is to provide a building energy consumption management system based on BIM architecture, which solves the problem that existing systems generally lack a self-learning mechanism and cannot continuously correct and optimize the model based on historical execution results, resulting in performance degradation and insufficient adaptability after long-term operation, thus failing to meet usage requirements.

[0007] This invention achieves the above objectives through the following technical solution: a building energy consumption management system based on BIM (Building Information Modeling), the system comprising:

[0008] BIM model building module, data acquisition module, AI feedback and decision-making module, strategy output module, and self-learning optimization module;

[0009] The BIM model building module is used to acquire building design data and build a BIM model that includes energy consumption monitoring nodes.

[0010] The data acquisition module is used to collect data on the building's internal and external environment, real-time energy consumption data, and building usage status data.

[0011] The AI ​​feedback and decision-making module is used to automatically analyze the collected data and generate energy efficiency optimization strategies;

[0012] The strategy output module is used to provide dynamic energy-saving suggestions to building managers;

[0013] The self-learning optimization module is used to iteratively optimize the decision model based on historical data and strategy execution results.

[0014] Furthermore, the BIM model building module acquires architectural design drawings, structural parameters, equipment parameters, and energy consumption monitoring point layout data to build a basic BIM model;

[0015] The BIM basic model includes building spatial coordinates, thermal parameters of the building envelope, rated power of equipment, and energy consumption monitoring node numbers.

[0016] Furthermore, the BIM model building module associates the energy consumption monitoring nodes with the physical spatial location of the BIM model, establishing a mapping relationship between energy consumption data and building components.

[0017] The building component types include at least one of the following: walls, floors, doors and windows, air conditioning equipment, lighting systems, and water supply and drainage equipment.

[0018] Furthermore, the BIM model building module expands the BIM model to include an energy consumption dimension, adding attribute fields required for energy consumption calculation:

[0019] The attribute fields include at least one of the component heat transfer coefficient, equipment operating efficiency, and lighting system power density, and the BIM model supports real-time linkage with the data acquisition module, which can automatically update the corresponding energy consumption attribute fields and monitoring node mapping relationships after changes to building components or equipment.

[0020] Furthermore, the data acquisition module integrates a sensor network and a third-party data interface, supporting real-time synchronization of multi-source data;

[0021] The building's external environmental data includes at least one of outdoor temperature, solar radiation intensity, wind speed, and precipitation.

[0022] The building interior environmental data includes at least one of indoor temperature, relative humidity, and indoor light intensity.

[0023] Furthermore, the real-time energy consumption data collected by the data acquisition module includes:

[0024] Electricity consumption, heating energy consumption, cooling energy consumption;

[0025] Building usage status data includes door and window opening / closing status, occupancy density, and equipment operating status;

[0026] All collected data are aligned with timestamps to form a time-series dataset, stored in a time-series database, and support the retention of at least 3 years of historical data. Data transmission is secured using an encryption protocol.

[0027] Furthermore, the AI ​​feedback and decision-making module integrates an artificial intelligence-based feedback mechanism and intelligent decision-making system, possessing the capabilities of environment-energy consumption correlation analysis, energy consumption prediction, and multi-objective optimization decision-making. It calculates the impact coefficient of environmental changes on energy consumption by constructing an environment-energy consumption correlation model, and obtains future energy consumption prediction values ​​by constructing an energy consumption prediction model based on neural networks.

[0028] Furthermore, the AI ​​feedback and decision-making module combines the energy consumption prediction results with the preset energy efficiency target to generate an optimization objective function;

[0029] Energy efficiency optimization strategies are generated using reinforcement learning algorithms;

[0030] The energy efficiency optimization strategy includes at least one of the following: air conditioning temperature regulation strategy, lighting control strategy, equipment operation strategy, and ventilation strategy.

[0031] Furthermore, the strategy output module supports visual display, multi-terminal push, and strategy execution status feedback;

[0032] Translate energy efficiency optimization strategies into energy-saving recommendations in natural language, including adjustment measures, expected effects, and implementation priorities;

[0033] Based on the BIM model, energy consumption distribution and key optimization areas are identified through color coding.

[0034] Furthermore, the self-learning optimization module constructs an evaluation index for the effectiveness of the strategy;

[0035] The evaluation indicators include: energy saving rate, cost reduction rate, and comfort deviation;

[0036] Machine learning algorithms are used to update the parameters of the decision model, the iteration cycle is set, and the abnormal data filtering function is supported to continuously improve the energy efficiency management effect.

[0037] The beneficial effects of this invention are as follows:

[0038] 1. By using the BIM model building module to spatially map building design data, structural parameters and energy consumption monitoring nodes, the precise association between building components and energy consumption data is achieved, forming a unified energy consumption data spatial model.

[0039] 2. The data acquisition module integrates sensor networks and third-party interfaces to collect data on the building's internal and external environment and energy consumption status in real time, forming traceable time-series data to provide high-precision data support for dynamic energy efficiency assessment.

[0040] 3. The AI ​​feedback and decision-making module is based on the environment-energy consumption correlation model and the neural network prediction model, which can predict future energy consumption trends. Combined with reinforcement learning algorithms, it automatically generates multi-objective energy efficiency optimization strategies to achieve coordinated energy-saving control of air conditioning, lighting, ventilation and equipment operation.

[0041] 4. The strategy output module presents the optimization results in the form of visual graphics and natural language energy-saving suggestions, supports multi-terminal display and strategy feedback, and facilitates quick understanding and implementation by building managers.

[0042] 5. The self-learning optimization module continuously evaluates indicators such as energy saving rate, cost reduction rate, and comfort deviation, and uses machine learning algorithms to automatically update the decision model parameters, constructing a closed-loop optimization system so that the longer the system runs, the better the energy efficiency management effect. Attached Figure Description

[0043] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0044] Figure 1 This is a closed-loop diagram of the entire process of the system of this invention;

[0045] Figure 2 This is a flowchart of the BIM model construction process of the present invention;

[0046] Figure 3 This is a flowchart of the AI ​​decision optimization process of the present invention. Detailed Implementation

[0047] The present application will now be described in further detail with reference to the accompanying drawings. It should be noted that the following specific embodiments are only used to further illustrate the present application and should not be construed as limiting the scope of protection of the present application. Those skilled in the art can make some non-essential improvements and adjustments to the present application based on the above application content.

[0048] Example 1:

[0049] Please see Figure 1-3This invention provides a technical solution: a building energy consumption management system based on BIM (Building Information Modeling), the system comprising:

[0050] The module includes BIM model building, data acquisition, AI feedback and decision-making, strategy output, and self-learning optimization.

[0051] The BIM model building module is used to acquire building design data and build a BIM model that includes energy consumption monitoring nodes.

[0052] Architectural design data encompasses various data generated during the design phase of a building, including its geometry, dimensions, structural type, material information, and spatial layout. This data forms the foundation for building the BIM model. The BIM model, or Building Information Modeling, is built upon various relevant information and data of a building project. It simulates the real-world information of a building through digital information and features visualization, coordination, simulation, optimization, and the ability to generate drawings. Energy monitoring nodes are specific locations set within the BIM model to monitor the energy consumption of various parts of the building in real time. These nodes correspond to energy monitoring equipment in the actual building, accurately acquiring energy consumption data and feeding it back into the model.

[0053] The data acquisition module is used to collect data on the building's internal and external environment, real-time energy consumption data, and building usage status data. The data acquisition module integrates sensor networks and third-party data interfaces, and supports real-time synchronization of multi-source data.

[0054] Among the building's internal and external environmental data, internal environmental data refers to environmental parameters inside the building, such as temperature, humidity, air quality, and light intensity. These data reflect the building's internal environmental conditions and directly impact energy consumption. External environmental data includes outdoor meteorological data, such as temperature, wind speed, wind direction, and solar radiation intensity. External environmental factors affect the building's heating and cooling loads, thus influencing energy consumption. Real-time energy consumption data is the actual energy consumed by the building during operation, such as the consumption of electricity, gas, and water, collected in real time by energy metering devices installed inside the building. Building usage status data reflects the building's usage, such as the number of people inside, equipment operating status, and usage time. This data will affect... The system monitors the building's energy consumption patterns. Sensor networks are network systems composed of numerous sensor nodes distributed across different locations within the building. These nodes collect various environmental and energy consumption data in real time and transmit the data to the data acquisition module for processing. Third-party data interfaces are used for data interaction with other systems or devices. Through these interfaces, data from other relevant systems, such as smart meters and smart water meters, can be acquired, enabling the integration of multi-source data. Real-time synchronization of multi-source data ensures that data from different data sources, such as sensor networks and third-party data interfaces, remains consistent over time and is updated to the system in real time, ensuring data accuracy and timeliness and providing a reliable basis for subsequent analysis and decision-making.

[0055] The AI ​​feedback and decision-making module integrates an AI-based feedback mechanism and intelligent decision-making system, automatically analyzes and collects data and generates energy efficiency optimization strategies. The module has the ability to perform environmental-energy consumption correlation analysis, energy consumption prediction and multi-objective optimization decision-making.

[0056] Among these, the AI ​​feedback mechanism utilizes AI technology to evaluate and provide feedback on the system's operational status based on collected data, promptly identifying energy consumption anomalies or low energy efficiency, and feeding these issues back to the decision-making system; the intelligent decision-making system, based on AI algorithms and models, analyzes and processes collected data to automatically generate energy efficiency optimization strategies to achieve building energy conservation goals; environment-energy consumption correlation analysis analyzes the relationship between internal and external environmental factors and energy consumption, identifying key factors affecting energy consumption and providing a basis for developing targeted energy conservation strategies; energy consumption prediction uses predictive models to forecast future building energy consumption based on historical data and current environmental and usage information, helping building managers to plan and schedule energy in advance; and multi-objective optimization decision-making capability comprehensively considers multiple objectives, such as energy conservation, comfort, and cost, when formulating energy efficiency optimization strategies, finding the optimal balance between these objectives through optimization algorithms to maximize overall benefits;

[0057] The strategy output module is used to provide dynamic energy-saving suggestions to building managers, and supports visualization, multi-terminal push and strategy execution status feedback.

[0058] Among them, dynamic energy-saving suggestions provide building managers with specific energy-saving measures and recommendations based on energy efficiency optimization strategies generated by the AI ​​feedback and decision-making module. These suggestions are dynamically adjusted according to the building's real-time operating status and environmental changes. Visualization displays energy consumption data, energy-saving suggestions, and other information are presented to building managers in intuitive graphical and chart formats, allowing them to quickly understand the building's energy consumption and energy-saving effects. For example, bar charts can be used to compare energy consumption over different time periods, and pie charts can be used to show the consumption ratio of various energy sources. Multi-terminal push notifications deliver energy-saving suggestions and related data to building managers through various terminal devices (such as computers, mobile phones, and tablets), enabling them to access information anytime, anywhere and make timely decisions. Strategy execution status feedback provides building managers with feedback on the implementation status of energy-saving strategies, including whether the strategies are effectively implemented and any problems encountered during implementation, so that managers can adjust strategies in a timely manner to ensure the achievement of energy-saving goals.

[0059] The self-learning optimization module is used to iteratively optimize the decision-making model based on historical data and strategy execution results, continuously improving energy efficiency management. The module dynamically adjusts model parameters through strategy evaluation indicators.

[0060] Historical data comprises various types of data collected and accumulated by the system during its past operation, including energy consumption data, environmental data, usage status data, and strategy execution effect data. This data forms the basis for self-learning and optimization. Strategy execution effect refers to the actual changes in building energy consumption after implementing the generated energy efficiency optimization strategy. The effectiveness of the strategy is evaluated by comparing energy consumption data before and after the strategy implementation. The iterative optimization decision model continuously improves and refines the decision model based on historical data and strategy execution effect. Through multiple iterations, the accuracy and decision-making ability of the model are gradually improved, thereby generating more effective energy efficiency optimization strategies. Strategy evaluation indicators are used to measure the effectiveness of energy efficiency optimization strategy implementation, such as energy consumption reduction rate, energy saving cost-effectiveness ratio, and comfort improvement degree. These indicators are used to quantitatively evaluate the strategy and provide a basis for dynamically adjusting model parameters. Dynamically adjusting model parameters automatically adjusts the parameters of the decision model based on the results of the strategy evaluation indicators, enabling the model to better adapt to the actual operating conditions of the building and constantly changing environmental conditions, and continuously improve the effectiveness of energy efficiency management.

[0061] It should be noted that during use, the BIM model building module acquires design data and sets energy consumption monitoring nodes, laying the foundation for subsequent precise management. The data acquisition module integrates multiple methods, synchronizing multi-source data in real time, enabling comprehensive and timely understanding of the building's internal and external environment, energy consumption, and usage status information, providing rich data for decision-making. The AI ​​feedback and decision-making module, leveraging artificial intelligence, possesses various analytical capabilities and can automatically generate scientific energy efficiency optimization strategies. The strategy output module provides dynamic suggestions to managers in various forms, facilitating timely decision-making and providing feedback on execution status. The self-learning optimization module iteratively optimizes the model based on historical data and execution results, dynamically adjusting parameters through evaluation indicators to continuously improve energy efficiency management. All modules of the entire system collaborate to achieve closed-loop management from data acquisition to decision optimization and feedback improvement, effectively improving building energy utilization efficiency and reducing operating costs.

[0062] In one embodiment, the specific implementation of the BIM model building module includes:

[0063] Obtain CAD drawings of the building design, structural parameters, equipment parameters, and energy consumption monitoring point layout data to construct a basic BIM model;

[0064] The parameters of the BIM base model include building space coordinates. , , thermal parameters of building envelope (Unit: W / (m²・℃)), Rated power of equipment (Unit: kW), Energy consumption monitoring node number ,in It covers all functional areas of the building and key energy-consuming equipment;

[0065] By associating energy consumption monitoring nodes with their physical spatial locations in the BIM model, a mapping relationship between energy consumption data and building components is established. The mapping expression is as follows:

[0066]

[0067] in , , For the first Spatial coordinates of each monitoring node (unit: m). For the corresponding building component type, the values ​​include walls, floors, doors and windows, air conditioning equipment, lighting systems, and water supply and drainage equipment;

[0068] The BIM model is expanded to include an energy consumption dimension, adding attribute fields required for energy consumption calculation, including component heat transfer coefficients. (Unit: W / (m²・℃), value range 0.2) 3.5), Equipment operating efficiency (Value range 0.6) 0.98), power density of lighting systems (Unit: W / m², range 3) 15);

[0069] The BIM model supports real-time linkage with the data acquisition module, automatically updating the corresponding energy consumption attribute fields and monitoring node mapping relationships when building components or equipment change.

[0070] This design, by acquiring various data to construct a BIM basic model including energy consumption monitoring nodes, clarifies parameter dimensions, associates monitoring nodes with physical locations, expands energy consumption dimension attributes, and supports real-time linkage updates. The comprehensive and accurate BIM model provides a solid foundation for energy consumption management. Detailed parameter dimensions accurately reflect the characteristics of various parts of the building, associating energy consumption monitoring nodes with physical locations clearly locates energy consumption, and the expanded energy consumption dimensions enrich the model information, facilitating accurate energy consumption calculation. The real-time linkage update function ensures that the model is always consistent with the actual state of the building, enabling energy consumption management to be based on the latest data, improving the accuracy and timeliness of decision-making, effectively improving building energy utilization efficiency, and reducing management costs.

[0071] In one embodiment, the data acquisition process of the data acquisition module includes:

[0072] Collect building external environment data:

[0073]

[0074] in, Outdoor temperature (unit: °C, measurement range -40) 60℃, accuracy ±0.1℃). Solar radiation intensity (unit: W / m², measurement range 0) 2000W / m², accuracy ±1%) Wind speed (unit: m / s, measurement range 0) 60m / s, accuracy ±0.1m / s). Rainfall (unit: mm, measurement range 0) 500mm, accuracy ±0.1mm);

[0075] External environmental data is acquired through both weather stations deployed on building rooftops and third-party weather data interfaces. When data conflicts occur, a weighted fusion algorithm is used to correct them.

[0076] Collect building interior environmental data:

[0077]

[0078] in, Number the indoor areas. Number of functional areas in a building For the first Indoor temperature of the area (unit: °C, measurement range 10) 40℃, accuracy ±0.1℃). Relative humidity (unit: %, measurement range 20%) 90%, accuracy ±1%) Indoor light intensity (unit: lux, measurement range 0) 10000 lux, accuracy ±1%).

[0079] At least two environmental sensors are deployed in each indoor area, and the average data is taken as the final data for that area.

[0080] Collect real-time energy consumption data:

[0081]

[0082] in, Electricity consumption (unit: kWh, measurement accuracy ±0.5%). Energy consumption for heating (unit: MJ, 1kWh=3.6MJ). Cooling energy consumption (unit: MJ);

[0083] Energy consumption data is collected in real time through smart meters and heat meters in building power distribution systems and heating and cooling systems.

[0084] Collect building usage status data:

[0085]

[0086] in, 0 indicates that the doors and windows are closed, and 1 indicates that the doors and windows are open. Data is collected by magnetic sensors on the doors and windows, and the collection frequency is consistent with the data sampling period. Personnel density (unit: people / m², measurement range 0) (1.5 people / m²), obtained through infrared sensors or video analysis algorithms; The device operating status vector (elements take values ​​of 0 or 1, where 0 indicates stop operation and 1 indicates normal operation) is read in real time through the device controller interface;

[0087] All collected data is timestamped. Alignment to form a time series dataset:

[0088]

[0089] The sampling period is set to Data storage uses a time-series database, supporting the retention of at least 3 years of historical data, and data transmission uses an encryption protocol to ensure security.

[0090] This design allows the data acquisition module to collect data on the building's internal and external environment, real-time energy consumption, and usage status from multiple dimensions. It clearly defines the acquisition methods, accuracy, data processing, storage, and transmission requirements. The multi-dimensional data comprehensively reflects the building's operational status, providing rich evidence for energy consumption analysis. High-precision acquisition ensures data accuracy and reduces errors. Dual acquisition of external environmental data and a weighted fusion algorithm improve data reliability. Reasonable data processing methods guarantee data quality. Time-series database storage supports long-term data retention, facilitating trend analysis. Encrypted transmission ensures data security and prevents information leakage, providing reliable and secure data support for subsequent AI analysis and decision-making.

[0091] In one embodiment, the energy efficiency optimization strategy generation process of the AI ​​feedback and decision-making module includes: constructing an environment-energy consumption correlation model and calculating the impact coefficient of environmental changes on energy consumption. The expression is:

[0092]

[0093] in, , , Let be the weighting coefficient, satisfying The algorithm is optimized using gradient descent, with the objective of minimizing the prediction error of the correlation model. The number of iterations is no less than 1000, and the convergence threshold is set to... ;

[0094] An energy consumption prediction model is built based on an LSTM neural network. The network structure includes an input layer, and the number of neurons equals the data dimension × 100%. , For historical time steps, Hidden layers, 2 layers in total, 256 neurons per layer; Output layer, number of neurons = energy consumption data dimension × , To predict the time step, Input as history Dataset at each time step Output for the future Energy consumption prediction at each time step The model loss function is:

[0095]

[0096] The model was trained using the Adam optimizer with a learning rate of 0.001, a batch size of 64, and at least 500 training iterations. The validation set accuracy was at least 90%.

[0097] Combining forecast results with energy efficiency targets , The value is set based on building type, national standards, and user needs, and is taken as 60% of the building design energy consumption quota. 90%, generate the optimization objective function:

[0098]

[0099] in, Strategy execution cost (unit: yuan, including equipment adjustment energy consumption cost and labor cost). Cost weight (value range 0.1) 0.5 (dynamically adjusted based on user cost sensitivity).

[0100] Energy efficiency optimization strategies are generated using reinforcement learning algorithms.

[0101]

[0102] Specifically, it includes:

[0103] Air conditioning temperature control strategy :

[0104]

[0105] The reference temperature is 26°C in summer and 20°C in winter. This is an adjustment coefficient (with a value range of 0.1). 0.5℃ / (kWh) This is the optimal energy consumption prediction value. The adjustment range is 24 hours in summer. 28℃, 18℃ in winter 22℃;

[0106] Lighting control strategy :

[0107]

[0108] For the first Area (unit: m²) For standard lighting requirements (500 lux for offices, 300 lux for meeting rooms, and 100 lux for corridors), when When necessary, the artificial lighting in the area will be automatically turned off or adjusted to the lowest power.

[0109] Equipment operation strategy :

[0110] Based on equipment efficiency Sort and run first High-efficiency equipment is used, and redundant operating equipment is shut down. In the same functional area, when the operating load of an equipment is less than 30% and there are other similar equipment, it is considered a redundant equipment. Service continuity is ensured during equipment switching.

[0111] Ventilation strategy :

[0112]

[0113] This is the ventilation volume adjustment coefficient (value range 0.01). 0.1 m³ / (s·℃) The average indoor temperature, when and At that time, the ventilation volume is automatically adjusted to reduce air conditioning energy consumption by utilizing outdoor natural conditions.

[0114] This design allows the AI ​​feedback and decision-making module to construct correlation and prediction models, generate optimization objective functions by combining energy efficiency targets, and use reinforcement learning algorithms to generate energy efficiency optimization strategies. It accurately analyzes the relationship between the environment and energy consumption, accurately predicts energy consumption trends, and provides a scientific basis for strategy formulation. By combining energy efficiency targets and cost factors to generate optimization objective functions, it achieves a balance between energy saving and cost control. The strategies generated by the reinforcement learning algorithm comprehensively cover aspects such as air conditioning, lighting, equipment operation, and ventilation, and are highly targeted. By comprehensively considering multiple factors, the generated strategies can effectively reduce building energy consumption, improve energy utilization efficiency, and at the same time take into account cost and comfort, thereby improving the overall operational efficiency of the building.

[0115] In one embodiment, the optimization process of the self-learning optimization module includes:

[0116] Construct strategy effectiveness evaluation metrics:

[0117]

[0118] in:

[0119] For energy saving rate, Energy consumption before strategy execution Energy consumption after strategy execution;

[0120] To reduce the cost rate, Costs incurred before strategy execution Costs incurred after strategy execution;

[0121] For comfort deviation, The optimal comfortable temperature is 25°C in summer and 21°C in winter. For optimal comfort humidity (40%) 60%)

[0122] , , Weighting coefficients ( , , (Supports user-defined adjustments), determined using the analytic hierarchy process (AHP), to assess the consistency ratio of the matrix. ;

[0123] Based on historical strategy data Corresponding evaluation indicators The random forest algorithm is used to update the parameters of the decision model. A random forest contains 100 decision trees, each with a maximum depth of 10 and a minimum number of splits of 2. The update formula is:

[0124] in, The learning rate has a value ranging from 0.001. 0.01, according to The rate of change is dynamically adjusted. To evaluate the gradient of the metric with respect to the model parameters;

[0125] Set iteration period The model parameters are re-optimized after each iteration cycle. When three consecutive iteration cycles... When the value improvement rate is less than 1%, stop parameter updates and maintain the current optimal model;

[0126] The module also supports abnormal data filtering, using 3 The criteria remove outliers from the collected data to ensure the reliability of the model training data.

[0127] This design allows the self-learning optimization module to construct evaluation indicators, update model parameters using the random forest algorithm based on historical data, set iteration cycles, and support outlier filtering. It scientifically evaluates the effectiveness of strategies, comprehensively considering factors such as energy saving rate, cost reduction rate, and comfort deviation. The random forest algorithm improves model accuracy and adaptability by updating parameters, making decisions more precise. Setting iteration cycles allows for regular model optimization to maintain its advanced nature. The outlier filtering function ensures the quality of training data, avoids interference from outliers, and improves model reliability. Through continuous optimization, it continuously improves energy efficiency management and achieves long-term effective control of building energy consumption.

[0128] In one embodiment, the implementation of the policy output module includes:

[0129] Energy efficiency optimization strategy Energy-saving recommendations translated into natural language include adjustment measures, expected effects, and implementation priorities;

[0130] Specifically, the adjustment measures include: specific operational steps and target parameters; expected effects include: energy saving rate, cost reduction, and changes in comfort; and implementation priorities include: 1. Level 5, with Level 1 being the highest, based on Value and urgency determination;

[0131] Based on the BIM model, the energy consumption distribution and key optimization areas are visualized. High energy consumption areas are marked by color coding, and users can click on the area to view detailed energy consumption data and optimization suggestions.

[0132] In this area, red indicates an over-consumption zone, where energy consumption exceeds 120%. ;

[0133] Yellow indicates the normal range, where energy consumption is between 80% and 120%. ;

[0134] Green indicates energy-saving areas with energy consumption <80%. ;

[0135] Generate dynamic reports It includes real-time energy consumption data, environmental parameters, a list of optimization strategies, expected energy-saving effects, and historical comparison data (daily, weekly, monthly, and yearly dimensions). The report supports export in PDF and Excel formats and supports push notifications to web (computer browser) and mobile (mobile APP, WeChat mini-program). The push frequency can be set to real-time, daily, weekly, or monthly.

[0136] The module also has a strategy execution status feedback function, which monitors the execution of optimization strategies in real time. When the strategy execution fails or the effect does not meet expectations, and the actual energy saving rate is less than 50% of the expected rate, the AI ​​feedback and decision-making module is automatically triggered to regenerate the optimization strategy.

[0137] This design allows the strategy output module to translate optimization strategies into natural language suggestions, visualize energy consumption distribution, generate dynamic reports, and provide execution status feedback. This facilitates building managers' understanding of the strategies, clarifies operational steps, expected results, and priorities, and improves execution efficiency. The visualized energy consumption distribution enables managers to intuitively understand the building's energy consumption situation and quickly locate problem areas. The dynamic reports provide multi-dimensional data, support export in different formats and multi-terminal push, and are convenient for viewing and analysis at any time. The execution status feedback function can promptly identify problems and regenerate strategies, ensuring continuous and effective energy consumption management, forming a closed-loop management system, and improving the overall level and effectiveness of building energy consumption management.

[0138] Those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing related hardware. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0139] The above embodiments provide a detailed description of the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A building energy consumption management system based on BIM (Building Information Modeling), characterized in that, The system includes: BIM model building module, data acquisition module, AI feedback and decision-making module, strategy output module, and self-learning optimization module; The BIM model building module is used to acquire building design data and build a BIM model that includes energy consumption monitoring nodes. The data acquisition module is used to collect data on the building's internal and external environment, real-time energy consumption data, and building usage status data. The AI ​​feedback and decision-making module is used to automatically analyze the collected data and generate energy efficiency optimization strategies; The strategy output module is used to provide dynamic energy-saving suggestions to building managers; The self-learning optimization module is used to iteratively optimize the decision model based on historical data and strategy execution results.

2. The building energy management system based on BIM architecture according to claim 1, characterized in that: The BIM model building module acquires architectural design drawings, structural parameters, equipment parameters, and energy consumption monitoring point layout data to build a basic BIM model. The BIM basic model includes building spatial coordinates, thermal parameters of the building envelope, rated power of equipment, and energy consumption monitoring node numbers.

3. The building energy management system based on BIM architecture according to claim 2, characterized in that: The BIM model building module associates energy consumption monitoring nodes with the physical spatial location of the BIM model, establishing a mapping relationship between energy consumption data and building components. The building component types include at least one of the following: walls, floors, doors and windows, air conditioning equipment, lighting systems, and water supply and drainage equipment.

4. The building energy management system based on BIM architecture according to claim 3, characterized in that: The BIM model building module expands the BIM model to include an energy consumption dimension, adding attribute fields required for energy consumption calculation: The attribute fields include at least one of the component heat transfer coefficient, equipment operating efficiency, and lighting system power density. The BIM model supports real-time linkage with the data acquisition module and can automatically update the corresponding energy consumption attribute fields and monitoring node mapping relationships after changes to building components or equipment.

5. The building energy management system based on BIM architecture according to claim 1, characterized in that: The data acquisition module integrates a sensor network and a third-party data interface, supporting real-time synchronization of multi-source data. The building's external environmental data includes at least one of outdoor temperature, solar radiation intensity, wind speed, and precipitation. The building interior environmental data includes at least one of indoor temperature, relative humidity, and indoor light intensity.

6. The building energy management system based on BIM architecture according to claim 5, characterized in that, The real-time energy consumption data collected by the data acquisition module includes: Electricity consumption, heating energy consumption, cooling energy consumption; Building usage status data includes door and window opening / closing status, occupancy density, and equipment operating status; All collected data are aligned with timestamps to form a time-series dataset, stored in a time-series database, and support the retention of at least 3 years of historical data. Data transmission is secured using an encryption protocol.

7. The building energy management system based on BIM architecture according to claim 1, characterized in that: The AI ​​feedback and decision-making module integrates an artificial intelligence-based feedback mechanism and intelligent decision-making system, and has the ability to perform environment-energy consumption correlation analysis, energy consumption prediction, and multi-objective optimization decision-making. It calculates the impact coefficient of environmental changes on energy consumption by constructing an environment-energy consumption correlation model, and obtains future energy consumption prediction values ​​by constructing an energy consumption prediction model based on neural networks.

8. The building energy management system based on BIM architecture according to claim 7, characterized in that: The AI ​​feedback and decision-making module combines energy consumption prediction results with preset energy efficiency targets to generate an optimization objective function; Energy efficiency optimization strategies are generated using reinforcement learning algorithms; The energy efficiency optimization strategy includes at least one of the following: air conditioning temperature regulation strategy, lighting control strategy, equipment operation strategy, and ventilation strategy.

9. The building energy management system based on BIM architecture according to claim 1, characterized in that: The strategy output module supports visual display, multi-terminal push, and strategy execution status feedback. Translate energy efficiency optimization strategies into energy-saving recommendations in natural language, including adjustment measures, expected effects, and implementation priorities; Based on the BIM model, energy consumption distribution and key optimization areas are identified through color coding.

10. The building energy management system based on BIM architecture according to claim 1, characterized in that: The self-learning optimization module constructs strategy effectiveness evaluation indicators; The evaluation indicators include: energy saving rate, cost reduction rate, and comfort deviation; Machine learning algorithms are used to update the parameters of the decision model, the iteration cycle is set, and the abnormal data filtering function is supported to continuously improve the energy efficiency management effect.

Images