An intelligent building energy-saving management method and system based on digital twin technology

Abstract

The present invention discloses an intelligent building energy-saving management method and system based on digital twin technology, which relates to the field of digital twin technology, including: establishing a three-dimensional building model based on a building information model and three-dimensional laser scanning point cloud data, generating a twin database through multi-source equipment data fusion; performing energy consumption simulation modeling according to the lamp dimming parameters, air conditioning performance curves and elevator operation logs in the twin database, and generating an energy consumption benchmark report; collecting personnel density and environmental parameters, and generating a building environment state matrix mapped to the twin database after time and space calibration; establishing a hierarchical threshold control strategy library based on the building energy consumption benchmark report and the building environment state matrix, and generating an optimized control strategy set through the equipment linkage relationship matrix. The present invention embeds a Pareto optimal solution screening mechanism and equipment mutual exclusion rule verification, so that the optimized control parameter package can achieve a reduction in daily average energy consumption while ensuring the comfort level compliance rate.

Description

Technical Field

[0001] The present invention relates to the technical field of digital twin, and in particular to an intelligent building energy-saving management method and system based on digital twin technology. Background Art

[0002] In recent years, the application of digital twin technology in the field of building energy management has been gradually deepened. By integrating building information model (BIM), Internet of Things sensors and dynamic simulation tools, static analysis of building energy consumption and equipment monitoring have been realized. In the prior art, the energy consumption modeling method based on BIM can generate a preliminary energy consumption benchmark by extracting structural parameters and matching equipment performance curves; while the multi-sensor fusion technology realizes local perception of building states by deploying temperature and humidity, occupancy density and environmental monitoring nodes. In addition, the construction of the equipment control strategy library often adopts the combination of rule engine and historical data mining. For example, lighting dimming or air conditioner start / stop is triggered by preset thresholds. At the optimization level, parameter tuning methods based on energy consumption simulation (such as genetic algorithm) have been used to balance energy efficiency and comfort requirements, and some studies further introduce computational fluid dynamics (CFD) to verify the thermal environment distribution.

[0003] However, there are still significant deficiencies in the prior art in terms of multi-source heterogeneous data fusion and dynamic collaborative control. First, the integration of BIM and point cloud data often results in a decrease in model accuracy due to construction noise interference, and the dynamic coupling of equipment status data and simulation models lacks a real-time calibration mechanism, causing the energy consumption benchmark to deviate from the actual working conditions. Second, the spatio-temporal mapping accuracy between environmental perception data and equipment control strategies is insufficient, and the existing threshold control strategies are difficult to adapt to the dynamic changes of building functional zoning, resulting in an exacerbation of the conflict between energy efficiency optimization and comfort requirements. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides an intelligent building energy-saving management method based on digital twin technology to solve the problem that it is difficult to balance energy efficiency and comfort in existing intelligent building systems due to low multi-source data fusion and lag in dynamic environment response.

[0006] To solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides an intelligent building energy-saving management method based on digital twin technology, which includes establishing a three-dimensional building model based on a building information model and three-dimensional laser scanning point cloud data, and generating a twin database through multi-source device data fusion;

[0008] Performing energy consumption simulation modeling according to the lamp dimming parameters, air conditioner performance curves and elevator operation logs in the twin database, and generating an energy consumption benchmark report;

[0009] Collect personnel density and environmental parameters, and generate a building environment state matrix that maps to the twin database after spatio-temporal calibration;

[0010] Establish a hierarchical threshold control strategy library based on the building energy consumption benchmark report and the building environment state matrix, and generate an optimized control strategy set through the equipment linkage relationship matrix;

[0011] Input the optimized control strategy set into the digital twin simulation environment for multi-objective optimization calculation, generate Pareto optimal solutions through the non-dominated sorting genetic algorithm, and finally output an executable optimized control parameter package.

[0012] As a preferred solution of the intelligent building energy-saving management method based on digital twin technology described in the present invention, wherein: the generation of the twin database includes the following steps,

[0013] Scan the point cloud data and perform RANSAC algorithm denoising, multi-site point cloud registration, and Poisson surface reconstruction to generate a three-dimensional building model;

[0014] Integrate the lighting circuit switch status, air conditioning unit operation parameters, and elevator group control data stream of the building automation equipment through the OPC protocol, establish a lighting component model, a heating, ventilation, and air conditioning component model, and a vertical transportation component model associated with the coordinates of the three-dimensional building model, and perform least squares optimization in the Unity3D engine to generate a twin database.

[0015] As a preferred solution of the intelligent building energy-saving management method based on digital twin technology described in the present invention, wherein: the generation of the building energy consumption benchmark report includes the following steps,

[0016] Extract the LED lamp specification parameters, generate a power curve through cubic polynomial fitting, and combine the historical dimming records to construct a three-dimensional tensor of the actual power value and the three-dimensional operation duration of the lamp to generate a daily average lighting energy consumption benchmark value;

[0017] Retrieve the heating, ventilation, and air conditioning component model parameters of the TRNSYS simulation platform, and generate the instantaneous cooling load peak in combination with the historical temperature and humidity data of the same period;

[0018] Construct a Markov chain state transition probability matrix based on the elevator energy consumption pulse signal, and generate the statistical distribution of the daily average power consumption of the elevator through Monte Carlo simulation;

[0019] Integrate the daily average lighting energy consumption benchmark value, the hourly cooling load peak of the air conditioning system, and the statistical distribution of the daily average power consumption of the elevator to generate a building energy consumption benchmark report.

[0020] As a preferred solution of the intelligent building energy-saving management method based on digital twin technology described in the present invention, wherein: the generation of a building environment state matrix mapped with a building energy consumption benchmark report comprises the following steps:

[0021] The improved YOLOv5 algorithm is used to process the webcam video stream and output the population density data of each area;

[0022] Anomaly marking and noise filtering are performed on the heterogeneous data of temperature and humidity sensors, CO2 concentration transmitters and meteorological parameter acquisition devices. The global environmental parameter distribution map is generated through inverse distance weighted interpolation. The environmental state matrix is formed by combining the personnel density data, and a spatiotemporal mapping relationship is established with the building energy consumption benchmark report.

[0023] As a preferred solution of the intelligent building energy-saving management method based on digital twin technology described in the present invention, the establishment of a hierarchical threshold control strategy library includes the following steps:

[0024] Define three-level thresholds for population density, CO2 concentration classification standards, and allowable ranges for indoor and outdoor temperature differences according to functional areas;

[0025] Formulate coordination rules for air conditioning temperature adjustment, lighting dimming and fresh air valves based on the equipment linkage relationship matrix, and generate a multi-condition combined control strategy;

[0026] The policy matching engine identifies the out-of-limit areas, combines the equipment mutual exclusion rules to perform conflict detection and priority sorting, and outputs the optimized preliminary control policy set.

[0027] As a preferred solution of the intelligent building energy-saving management method based on digital twin technology described in the present invention, wherein: the outputting executable optimization control parameter package includes the following steps:

[0028] Convert the optimized preliminary control strategy set into EnergyPlus input file, perform dynamic energy consumption simulation and computational fluid dynamics analysis, and evaluate thermal comfort;

[0029] A multi-objective function of energy consumption reduction rate and comfort level compliance rate was constructed, and a non-dominated sorting genetic algorithm was used to generate the Pareto optimal solution set.

[0030] Based on the Pareto optimal solution set, the strategy combination with the highest comprehensive score is selected and encoded into an optimization control parameter package in the device executable instruction format.

[0031] As a preferred solution of the intelligent building energy-saving management method based on digital twin technology described in the present invention, the device executable instruction format includes a device address code, a parameter type and a target value triple.

[0032] In a second aspect, the present invention provides an intelligent building energy-saving management system based on digital twin technology, including a modeling and fusion module that establishes a three-dimensional building model based on building information models and three-dimensional laser scanning point cloud data, and generates a twin database through multi-source device data fusion;

[0033] An energy consumption benchmark module that performs energy consumption simulation modeling based on the lighting dimming parameters, air-conditioning performance curves, and elevator operation logs in the twin database to generate an energy consumption benchmark report;

[0034] An environmental perception module that collects personnel density and environmental parameters, and generates a building environmental state matrix mapped to the twin database after spatio-temporal calibration;

[0035] A strategy generation module that establishes a hierarchical threshold control strategy library based on the building energy consumption benchmark report and the building environmental state matrix, and generates an optimized control strategy set through an equipment linkage relationship matrix;

[0036] A simulation and optimization module that inputs the optimized control strategy set into a digital twin simulation environment for multi-objective optimization calculation, generates Pareto optimal solutions through a non-dominated sorting genetic algorithm, and finally outputs an executable optimized control parameter package.

[0037] In a third aspect, the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and: when the computer program is executed by the processor, any step of the intelligent building energy-saving management method based on digital twin technology as described in the first aspect of the present invention is implemented.

[0038] In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, and: when the computer program is executed by the processor, any step of the intelligent building energy-saving management method based on digital twin technology as described in the first aspect of the present invention is implemented.

[0039] The beneficial effects of the present invention are as follows: a global environmental parameter distribution map is generated through an inverse distance weighted interpolation algorithm, and the control strategy priority is dynamically adjusted in combination with the functional area division standard; the fresh air volume and lighting brightness are synchronously increased through the equipment linkage relationship matrix to avoid local thermal environment imbalance caused by traditional single-device adjustment; when performing multi-objective optimization using the non-dominated sorting genetic algorithm, the optimized control parameter package is enabled to reduce the daily average energy consumption on the premise of ensuring that the PMV-PPD comfort level meets the standard by embedding a Pareto optimal solution screening mechanism and an equipment mutual exclusion rule verification. Description of the Drawings

[0040] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0041] Figure 1 It is a flowchart of an intelligent building energy-saving management method based on digital twin technology.

[0042] Figure 2 It is a schematic diagram for generating a twin database.

[0043] Figure 3 It is a schematic diagram of energy consumption benchmark modeling and simulation.

[0044] Figure 4 It is a schematic diagram for generating multi-objective optimization control parameters. Specific Embodiments

[0045] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following will give a detailed description of the specific embodiments of the present invention with reference to the accompanying drawings of the specification.

[0046] In the following description, many specific details are set forth to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0047] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that can be included in at least one implementation manner of the present invention. The "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it an individual or selectively exclusive embodiment with other embodiments.

[0048] Referring to Figures 1 to 4 , it is an embodiment of the present invention. This embodiment provides an intelligent building energy-saving management method based on digital twin technology, including the following steps:

[0049] S1. Use building information model technology to extract the structural drawings of the target large building from the building engineering project document library.

[0050] Among them, the structural drawings include beam-column structure dimension markings, space segmentation floor plans, and door and window installation positioning information.

[0051] Use a 3D laser scanner to perform panoramic scanning on the structural drawings of a target large building, generating point cloud data containing wall curvature data, pipeline routing data, and the positions of mechanical and electrical cable trays; import the point cloud data into 3D modeling software, identify and remove noise points such as residual points of construction scaffolding and interference points of temporary equipment through the RANSAC algorithm, and retain the point cloud data of the building's main structure; perform multi-station point cloud registration on the point cloud data after removing noise points, and use the target ball positioning method to unify the point cloud data of the scanning stations into the building's global coordinate system.

[0052] Input the registered point cloud data into MeshLab software for Poisson surface reconstruction: set the sampling depth and the number of smoothing iterations to generate a 3D building model containing the triangular mesh of beam-column structures and the topological structure of door and window openings; establish a communication connection with building automation equipment through the OPC protocol, and poll the communication channel to collect the device status data of the real-time switch status of each floor's lighting circuit, the operating parameters of air handling units, and the data stream of the elevator group control system.

[0053] Among them, the operating parameters of the air handling unit include the data of IoT monitoring points such as fan speed, cooling water valve opening, and the operating status of the fresh air handling unit;

[0054] Import the CAD wiring diagram of the lighting circuit into the ETAP electrical design software, perform an association mapping between the lamp IDs in the CAD wiring diagram of the lighting circuit and the real-time switch status data of the lighting circuit collected by the building automation equipment, and establish a lighting component model containing the coordinate positioning of the lamps in the 3D building model.

[0055] Perform non-linear regression on the condensation temperature and evaporation temperature of the compressor through the Levenberg-Marquardt algorithm to generate the compressor COP curve; based on the refrigeration capacity range output by the COP curve obtained by regression, perform cubic spline interpolation on the discrete fan speed and air delivery volume data to generate a continuous air volume; establish a mapping relationship between the fan power and the continuous air volume in the form of a cubic polynomial to obtain real-time flow data; use the real-time flow data and the inlet water temperature to perform bilinear interpolation in the heat exchanger efficiency table to dynamically correct the heat exchanger efficiency under different working conditions, and perform multi-parameter coupling of the compressor COP curve, the mapping relationship between the fan power and the continuous air volume, and the corrected heat exchanger efficiency in the simulation platform to generate a heating, ventilation, and air conditioning component model.

[0056] Configure the event log parser of AnyLogic simulation software to parse the car call record data transmitted by the elevator group control system, generate a vertical transportation component model containing the elevator's docking operation completed within a fixed period, and the vertical transportation component model outputs an energy consumption pulse signal corresponding to each door machine action.

[0057] Import a 3D building model file in the Unity3D engine scene editor. The 3D building model file contains vertex coordinate data of the beam-column structure and the path of the material texture map. Define the initial positions of the lighting component model, the HVAC component model, and the vertical transportation component model in the coordinate system of the 3D building model file through an XML configuration file, and perform least squares optimization calculations based on the Levenberg-Marquardt algorithm to minimize the deviation between the coordinates of the lighting fixture model and the corresponding coordinates of the 3D building model.

[0058] Configure the WebSocket data interface of the Unity3D engine, continuously receive updated device status data, and drive the lighting brightness parameter of the lighting component model, the fan blade rotation angle parameter of the HVAC component model, and the elevator car position parameter of the vertical transportation component model to achieve dynamic synchronization through the keyframe interpolation algorithm; write the synchronized 3D building model coordinate data and device status data into the time series database, and record the vertex coordinate data of the building components and the device tag values to form a twin database containing the building geometric features and the real-time status of the devices.

[0059] S2. Extract the LED lighting fixture specification parameters from the lighting component model exported from the ETAP electrical design software. The specification parameters include the dimming level-power correspondence table of the lighting fixture model, and record the non-linear power curve corresponding to the dimming range in the dimming level-power correspondence table.

[0060] Further explain that the dimming level-power correspondence table is stored as a numerical matrix of row × column. The first column is the dimming level percentage value, and the second column is the corresponding power value. The power curve is generated by cubic polynomial fitting.

[0061] Parse the historical lighting circuit dimming records. The dimming records include the historical data of the lighting fixtures maintaining the dimming level at fixed time periods every day. Convert the dimming records into a three-dimensional array. Each element in the three-dimensional array stores the dimming level value of the corresponding lighting fixture at a specific minute period on a specific date. The dimming level value is extracted through a JSON parser and filled into the array index position.

[0062] Perform matrix multiplication operations on the dimming level-power correspondence table and the three-dimensional dimming level array. Use the dimming level value as an index to query the corresponding column of the dimming level-power correspondence table to generate the three-dimensional actual power value of the lighting fixtures.

[0063] Further explain that the expression for generating the three-dimensional actual power value of the lighting fixtures is:

[0064] ;

[0065] where, is the th lighting fixture on the th day at the The actual power value in minutes, is the lamp number, is the date sequence, is the minute period, is the th lamp at the th day and the th minute of the dimming level corresponding power value, is the th lamp at the th day and the th minute of the dimming level, , , are the coefficients of the cubic polynomial fitting, is the constant term.

[0066] Construct a three-dimensional operation duration tensor in MATLAB. The three dimensions of the three-dimensional operation duration tensor correspond to the lamp number, date sequence, and minute period respectively. The three-dimensional operation duration tensor is filled with the actual on-time data of each lamp per minute of each day. The on-time data is based on the historical record of the real-time switch state of the lighting circuit. When a switch closure signal is detected, a 1 (on state) is written at the corresponding lamp, date, and minute index positions, otherwise it remains 0 (off state).

[0067] Perform the Hadamard product operation on the three-dimensional actual power value of the lamp and the three-dimensional operation duration tensor to generate an energy consumption tensor. The expression is:

[0068] ;

[0069] where, is the energy consumption tensor, is the three-dimensional operation duration tensor;

[0070] Accumulate and calculate the energy consumption tensor along the time dimension to generate the daily lighting energy consumption benchmark value of the office area lamps.

[0071] Retrieve the model parameters of the HVAC components in the TRNSYS simulation platform.

[0072] Among them, the model parameters of the HVAC components include the compressor COP curve, the mapping relationship between the fan power and the continuous air volume, and the corrected heat exchanger efficiency; the compressor COP curve records the COP value operating points corresponding to the condensation temperature, the mapping relationship between the fan power and the continuous air volume stores the power corresponding to the air volume; the heat exchanger efficiency table registers the heat exchange efficiency of the plate heat exchanger at the cold water inlet temperature.

[0073] Associate with the historical temperature and humidity data provided by the local meteorological data center. The outdoor temperature and humidity data includes the peak temperature and humidity records for each day. Substitute the fan speed in the operating parameters of the air conditioner unit into the mapping relationship between fan power and continuous air volume to obtain the real-time power. Read the opening value recorded by the cooling water valve opening sensor, and perform bilinear interpolation calculation in combination with the heat exchanger efficiency to obtain the cooling capacity output. The expression is:

[0074] ;

[0075] Among them, is the cooling capacity output, is the heat exchange efficiency at the cooling water valve opening and the temperature difference , is the cooling water valve opening, is the temperature difference between the cold water inlet temperature and the set temperature, is the density of the cooling water, is the specific heat capacity of water, is the cooling water flow rate;

[0076] The expression of the heat exchange efficiency is:

[0077] ;

[0078] Among them, is the adjacent interpolation point index, is the contribution weight of each adjacent interpolation point to the actual efficiency, is the discrete efficiency value in the heat exchanger efficiency, is the row and column index of the adjacent interpolation point.

[0079] Convert the cooling capacity to the actual power consumption through the compressor COP curve, and generate the hourly cooling load peak for a fixed period according to the historical temperature and humidity data and the outdoor temperature and humidity correction coefficient.

[0080] Further explanation, the outdoor temperature and humidity correction coefficient is used to quantify the additional impact of outdoor temperature and humidity on the cooling load of the air conditioning system, convert the additional heat load caused by outdoor high temperature and high humidity into a correction value for the cooling load, and obtain the outdoor temperature and humidity correction coefficient.

[0081] Import the elevator energy consumption pulse signal output by the vertical transportation component model into the simulation software. The energy consumption pulse signal records the characteristic waveform data of the energy consumption generated by each elevator stop. Statistically analyze the car call record data of the building automation equipment during the past historical peak periods. The car call record data shows the average number of stops per day;

[0082] Construct a Markov chain state transition probability matrix. The row and column dimensions of the state transition probability matrix correspond to the elevator stop floor numbers. The element values in the state transition probability matrix are obtained by counting the floor transition frequencies between adjacent stop events. The state transition probability matrix shows the transition probabilities between floors.

[0083] Input the state transition probability matrix into a Monte Carlo simulation and set the number of simulations. Each simulation generates an operation scenario including the number of stops, running distance, and door operation times. The simulation results of the operation scenario show the statistical distribution of the daily average power consumption. Integrate the daily average lighting energy consumption benchmark value, the hourly peak cooling load of the air conditioning system, and the statistical distribution of the elevator daily average power consumption to generate a building energy consumption benchmark report containing the sub-item energy consumption curves of each component, and mark the energy consumption pattern differences between weekdays and holidays.

[0084] S3. Deploy network cameras with the function of counting the number of people at the ceiling positions of each building zone. The network cameras output H.264 encoded video stream data in TCP / IP protocol. The video stream data is transmitted to the edge computing device through a power supply and communication integrated device. The edge computing device loads an object detection algorithm improved based on the YOLOv5 architecture. The object detection algorithm performs frame-by-frame human body recognition on the video stream, counts the number of the largest human bounding boxes recognized in each frame, and generates the personnel density data updated periodically for each area.

[0085] It should be noted that the network cameras in this embodiment are only used for counting the number of people and do not involve privacy information.

[0086] Uniformly arrange temperature and humidity composite sensors at the top of the building functional spaces according to the principle of symmetry about the geometric center of the space. The temperature and humidity composite sensors are built with environmental detection chips. The environmental detection chips cover the preset temperature and humidity measurement ranges, and monitor nodes are configured according to the standard to form the spatial distribution topology of the environmental monitoring points.

[0087] It should be noted that in this embodiment, the preset temperature range is 0 - 50 °C, the preset environmental humidity is 0 - 100%, and the standard for configuring monitoring nodes is to install 1 node per 200 square meters to cover all ranges.

[0088] Configure a low-power wireless transmission gateway in the weak current room of the building equipment. The low-power wireless transmission gateway receives the measurement data sent by the temperature and humidity composite sensors.

[0089] Among them, the measurement data includes device identification, temperature measurement value, humidity measurement value, and acquisition time, and the transmission period is set to a fixed time interval.

[0090] Install a meteorological parameter acquisition device on the periphery of the building. The meteorological parameter acquisition device is configured with a wind speed detection element, a temperature detection element, and a radiation detection element. The wind speed detection element covers a preset range, the temperature detection element covers a preset temperature range, and the radiation detection element measures the solar radiation intensity;

[0091] It should be noted that the preset range of the wind speed detection element here is 0 - 60 m / s, and the range of the temperature detection element is -40°C to +85°C.

[0092] Connect to the CO2 concentration transmitter of the existing building control system through a bus. The CO2 concentration transmitter outputs an analog signal, and use a communication protocol to regularly read the CO2 concentration value;

[0093] Create a message queue topic in the data processing device. The message queue topic receives the personnel density data from the edge computing device, the temperature and humidity measurement values from the low-power wireless transmission gateway, the wind speed, temperature, and solar radiation intensity measured by the outdoor meteorological parameter acquisition device, and the CO2 concentration value read by the communication protocol; perform timestamp calibration in fixed time units to generate an original data set with a unified time identifier; perform data cleaning on the original data set: data with temperature measurement values exceeding the sensor range are marked as invalid, data with relative humidity measurement values exceeding the physical saturation state are marked as abnormal, and sudden changes in CO2 concentration values are regarded as interference noise.

[0094] Among them, the range of the temperature measurement value is based on the working temperature range in the sensor factory calibration certificate, the abnormal value determination standard is set according to the sensor technical specifications, and a sudden change in the CO2 concentration value exceeding 500 ppm / minute is interference noise, which is set according to the indoor air quality standard.

[0095] Perform inverse distance weighted spatial interpolation calculation on the cleaned original data set:

[0096] Set the spatial interpolation search radius, calculate the temperature prediction value for the area where no sensor is arranged, and generate an environmental parameter distribution map covering the entire building.

[0097] Integrate the personnel density data, the environmental parameter distribution map, the outdoor meteorological acquisition parameters, and the CO2 concentration value to construct a building environmental state matrix including fields of personnel quantity density, environmental parameters, outdoor meteorological acquisition parameters, and CO2 concentration value in the air quality field, and establish a spatio-temporal mapping relationship with the building energy consumption benchmark report.

[0098] S4. Divide the target building into functional areas such as office areas, commercial areas, and equipment areas according to the building function planning drawings, and form a spatial partition database including regional boundary coordinates and attribute labels.

[0099] Among them, the functional area division standard is based on the spatial use and service object characteristics;

[0100] Define the three - level threshold of personnel density for each type of functional area in the spatial partition database: The first - level threshold corresponds to the lowest personnel density state, the second - level threshold corresponds to the medium personnel density state, and the third - level threshold corresponds to the highest personnel density state.

[0101] Furthermore, the setting of the three - level threshold is based on whether the area of the spatial capacity region in the 3D building model matches the fire evacuation passage setting corresponding to the threshold. For example:

[0102] Threshold setting for the office area:

[0103] The first - level threshold: 0.05 persons / ㎡ (the lowest personnel density, corresponding to non - working hours);

[0104] The second - level threshold: 0.15 persons / ㎡ (medium personnel density, normal office state);

[0105] The third - level threshold: 0.30 persons / ㎡ (the highest personnel density, meeting / concentrated office state);

[0106] Threshold setting for the commercial area:

[0107] The first - level threshold: 0.10 persons / ㎡ (during business preparation);

[0108] The second - level threshold: 0.35 persons / ㎡ (normal passenger flow);

[0109] The third - level threshold: 0.60 persons / ㎡ (during peak promotion);

[0110] Threshold setting for the equipment area:

[0111] The first - level threshold: 0.02 persons / ㎡ (during daily inspection);

[0112] The second - level threshold: 0.05 persons / ㎡ (during equipment debugging);

[0113] The third - level threshold: 0.10 persons / ㎡ (the highest safety capacity, over - limit prohibited).

[0114] Define the CO2 concentration classification standard: The first - level standard is the excellent air quality level, the second - level standard is the good air quality level, and the third - level standard is the air quality warning level.

[0115] Furthermore, the classification standard is obtained by verifying the real - time monitoring data of the CO2 concentration sensor in the building automation equipment. For example:

[0116] The first - level (excellent air quality level): CO2 concentration ≤ 800 ppm;

[0117] Second level (good air quality level): 800 ppm < CO2 concentration ≤ 1000 ppm;

[0118] Third level (air quality warning level): CO2 concentration > 1000 ppm.

[0119] Define the allowable temperature difference range as the reasonable fluctuation range of the indoor and outdoor temperature difference.

[0120] Further explanation, in this embodiment, the range of the reasonable fluctuation range is as follows:

[0121] Office area: The allowable range of indoor and outdoor temperature difference is ±1.5°C; Commercial area: The allowable range of indoor and outdoor temperature difference is ±2.0°C; Equipment area: The allowable range of indoor and outdoor temperature difference is ±0.5°C.

[0122] Based on the building energy consumption benchmark report and historical operation data, formulate equipment control rules for the corresponding grade threshold combinations of each type of functional area, synchronously generate the equipment linkage relationship matrix, and construct a control strategy library containing multi-condition combinations.

[0123] Further explanation, the equipment control rules are as follows:

[0124] Generate a mapping table of the air conditioner unit numbers and lighting circuit codes in the same area according to the coordinate information and space segmentation floor plan in the 3D building model;

[0125] Retrieve the constraint terms for the interlocking operation of the fresh air valve and the fan (for example, when the opening of the cooling water valve > 70%, the opening of the fresh air valve ≤ 50%);

[0126] Combine the statistical results of historical equipment status mutually exclusive events in the time series database (for example, the conflict probability between the lighting off command and the personnel presence detection is 4.7%);

[0127] The matrix dimensions in the equipment linkage relationship matrix include the air conditioning area, lighting circuit, and fresh air valve; the assignment logic includes that when the lighting circuit in the area equipment room is closed, the air conditioning temperature is forcibly set; when the dimming level of the lighting circuit in the area display area is increased, the opening of the fresh air valve is increased in linkage.

[0128] Establish a mapping relationship between the working conditions and control commands in the control strategy library: when the personnel density in the commercial area reaches the second-level threshold, trigger an increase in the fresh air volume and adjust the lighting brightness, and at the same time check the coordination relationship between the fresh air valve and the lighting circuit in the equipment linkage relationship matrix; when the CO2 concentration in the office area reaches the third-level standard, trigger the air conditioner unit to increase the fresh air ratio and start the emergency ventilation mode, and prohibit the dimming operation of the lighting in the area from being executed simultaneously according to the equipment linkage relationship matrix;

[0129] The policy matching engine receives real-time updated personnel density data, carbon dioxide concentration over-limit warning signals, and indoor-outdoor temperature difference abnormal alarm information in the building environment status matrix. The personnel density data reflects the personnel distribution density in each area through color gradients.

[0130] The policy matching engine conducts a spatial overlay analysis of the real-time personnel density data and the functional area division results, identifies the spatial locations and the corresponding functional categories of the areas exceeding the corresponding level thresholds, and simultaneously calls the equipment linkage relationship matrix to exclude the areas where lighting and air-conditioning adjustment cannot be executed simultaneously due to equipment mutual exclusion. At the same time, it scans the CO2 concentration values and marks the monitoring points exceeding the grading standard. It conducts a working condition matching search for the identified monitoring points exceeding the grading standard.

[0131] Further explanation, the working condition matching search retrieves the corresponding equipment control rules from the control strategy library according to the functional area type, the category of the monitoring points exceeding the grading standard, and the over-limit level.

[0132] Among them, the equipment control rules include the adjustment range of the air-conditioning set temperature, the list of lighting circuit opening and closing combinations, and the target value of the fresh air valve opening.

[0133] Convert the equipment control rules obtained from the matching search into an executable instruction set, and correct the specific parameters in the instruction set according to the equipment linkage matrix.

[0134] For example, when there is a synergistic relationship between the regional lighting circuit dimming instruction and the air-conditioning instruction, an additional compensation amount is added to the downward adjustment value of the air-conditioning set temperature; when there is a mutually exclusive relationship between the regional fresh air valve opening instruction and the lighting circuit instruction, the fresh air valve adjustment instruction is frozen.

[0135] Conduct conflict detection and priority sorting on the executable instruction set. The conflict detection is based on the equipment linkage relationship matrix. When there is a conflict between the lighting circuit closing instruction and the personnel presence detection result, the priority of the personnel presence detection result is automatically increased; the priority sorting divides the execution order according to the severity of the environmental parameter over-limit, and generates a preliminary control strategy set including the timestamp, area number, equipment type, and control parameters.

[0136] Conduct a secondary verification of the preliminary control strategy set with the cooperation / mutual exclusion rules in the equipment linkage relationship matrix to ensure that the lighting enhancement instruction and the air-conditioning load reduction instruction in the area take effect synchronously, and finally output the optimized preliminary control strategy set.

[0137] S5. Extract the air-conditioning set temperature modification instructions, lighting circuit dimming control instructions, and fresh air valve opening adjustment instructions in the optimized preliminary control strategy set, and construct an instruction mapping table; convert the air-conditioning set temperature modification instructions into HVAC system set point parameters through the instruction mapping table, convert the lighting circuit dimming control instructions into lighting power density adjustment coefficients, and convert them into the input file format of the EnergyPlus building energy consumption simulation software; load the converted EnergyPlus input file in the digital twin environment, configure the simulation time span, set the time step, activate the building envelope heat conduction calculation model and the internal heat calculation model, start the dynamic energy consumption simulation process, and record the instantaneous values of lighting energy consumption in each zone, the power consumption of the air-conditioning unit compressor, and the cumulative value of elevator operation energy consumption in real time. After the simulation ends, output a predicted energy consumption report containing sub-item energy consumption curves, and mark the predicted daily total energy consumption value after the implementation of the strategy in the predicted energy consumption report.

[0138] Synchronously start the computational fluid dynamics simulation. The computational fluid dynamics simulation generates a structured grid based on the three-dimensional building model. The structured grid is locally refined in the personnel activity area, and the boundary conditions are set as the air velocity distribution at the air-conditioning supply outlet, the pressure condition at the return air outlet, and the surface temperature of the building envelope; import the air-conditioning operation parameters in the predicted energy consumption report into the computational fluid dynamics simulation. The operation parameters include supply air temperature, wind speed, and fresh air ratio, and combine with the real-time personnel density distribution data in the building environment state matrix to calculate the air velocity field, temperature field, and CO2 concentration field distribution in each zone.

[0139] Based on the computational fluid dynamics simulation results, intercept the horizontal section data at the personnel activity location for thermal comfort evaluation: calculate the predicted mean vote value and the expected percentage of dissatisfaction according to the PMV-PPD standard, and mark the supercooled area where the PMV value is lower than the comfort lower limit and the exceeded area where the CO2 concentration exceeds the air quality warning level.

[0140] Establish a multi-objective optimization function. The optimization objectives include two optimization objectives: the energy consumption reduction rate and the comfort compliance rate. The energy consumption reduction rate is calculated by comparing the predicted energy consumption value with the baseline value in the building energy consumption baseline report, and the comfort compliance rate is obtained by statistically analyzing the proportion of the space where the PMV value is within the comfort interval.

[0141] Use the non-dominated sorting genetic algorithm for multi-objective optimization. The initial population of the non-dominated sorting genetic algorithm is generated by Gaussian mutation of the control instruction parameters in the preliminary control strategy set. Each individual gene encoding uses real number encoding, and each gene corresponds to a combination parameter of the air-conditioning temperature setting value, lighting dimming level, and fresh air valve opening.

[0142] During the iterative optimization process, dynamic energy consumption simulation and computational fluid dynamics thermal comfort evaluation are re-executed in each iteration. The candidate strategies that meet both low energy consumption and a high comfort compliance rate in each generation of the population are recorded, and the candidate strategies are prioritized according to the Pareto front sorting rule.

[0143] When the optimal solutions of multiple consecutive generations of the population remain stable, the optimization process is terminated, and the strategy combination with the highest comprehensive score is selected from the Pareto optimal solution set. The strategy combination includes the optimized parameters of the adjusted air-conditioning temperature setting value, lighting dimming level, and fresh air valve opening degree.

[0144] All the optimized parameters are re-encoded into the format of device-executable instructions. The device-executable instruction format includes a triple of device address code, parameter type, and target value, and an optimized control parameter package containing control instructions is generated. Each instruction in the optimized control parameter package is marked with an effective time window and a spatial scope of action.

[0145] This embodiment also provides an intelligent building energy-saving management system based on digital twin technology, including:

[0146] A modeling and fusion module that establishes a three-dimensional building model based on the building information model and three-dimensional laser scanning point cloud data, and generates a twin database through multi-source device data fusion.

[0147] An energy consumption benchmark module that performs energy consumption simulation modeling based on the lamp dimming parameters, air-conditioning performance curves, and elevator operation logs in the twin database, and generates an energy consumption benchmark report.

[0148] An environment perception module that collects the personnel density and environmental parameters, and generates a building environmental state matrix mapped to the twin database after spatio-temporal calibration.

[0149] A strategy generation module that establishes a hierarchical threshold control strategy library based on the building energy consumption benchmark report and the building environmental state matrix, and generates an optimized control strategy set through the device linkage relationship matrix.

[0150] A simulation and optimization module that inputs the optimized control strategy set into the digital twin simulation environment for multi-objective optimization calculation, generates Pareto optimal solutions through the non-dominated sorting genetic algorithm, and finally outputs an executable optimized control parameter package.

[0151] This embodiment also provides a computer device applicable to the intelligent building energy-saving management method based on digital twin technology, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the intelligent building energy-saving management method proposed in the above embodiment.

[0152] The computer device can be a terminal, which includes a processor, a memory, a communication interface, a display screen, and an input device connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner. The wireless manner can be achieved through WIFI, carrier networks, NFC (Near Field Communication), or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads provided on the housing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0153] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the intelligent building energy-saving management method based on digital twin technology as proposed in the above embodiment. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, abbreviated as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, abbreviated as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, abbreviated as EPROM), programmable read-only memory (Programmable Red-Only Memory, abbreviated as PROM), read-only memory (Read-Only Memory, abbreviated as ROM), magnetic memory, flash memory, magnetic disks, or optical discs.

[0154] In summary, the present invention generates a global environmental parameter distribution map through the inverse distance weighted interpolation algorithm, and dynamically adjusts the priority of the control strategy in combination with the functional area division standard; synchronously improves the fresh air volume and lighting brightness through the device linkage relationship matrix, avoiding the local thermal environment imbalance caused by traditional single-device adjustment; when using the non-dominated sorting genetic algorithm for multi-objective optimization, by embedding the Pareto optimal solution screening mechanism and device mutual exclusion rule verification, the optimized control parameter package can reduce the daily average energy consumption on the premise of ensuring that the PMV-PPD comfort level meets the standard.

[0155] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. An intelligent building energy-saving management method based on digital twin technology, characterized in that: including establishing a 3D building model based on building information model and 3D laser scanning point cloud data, and generating a digital twin database through multi-source device data fusion; conducting energy consumption simulation modeling according to the lighting dimming parameters, air-conditioning performance curves and elevator operation logs in the digital twin database, and generating an energy consumption baseline report; collecting personnel density and environmental parameters, and generating a building environmental state matrix mapped to the digital twin database after spatio-temporal calibration, including the following steps processing the video stream of network cameras to output personnel density data for each area; performing anomaly marking and noise filtering on the heterogeneous data of temperature and humidity sensors, CO2 concentration transmitters and meteorological parameter collection devices, generating a global environmental parameter distribution map through inverse distance weighted interpolation, forming an environmental state matrix in combination with personnel density data, and establishing a spatio-temporal mapping relationship with the building energy consumption baseline report; establishing a hierarchical threshold control strategy library based on the building energy consumption baseline report and the building environmental state matrix, and generating an optimized control strategy set through the device linkage relationship matrix; the establishment of the hierarchical threshold control strategy library includes the following steps defining the three-level threshold of personnel density, the grading standard of CO2 concentration and the allowable range of indoor-outdoor temperature difference according to the functional area; formulating the coordination rules for air-conditioning temperature adjustment, lighting dimming and fresh air valve based on the device linkage relationship matrix, and generating a multi-condition combined control strategy; identifying the over-limit areas through the strategy matching engine, performing conflict detection and priority sorting in combination with the device mutual exclusion rules, and outputting an optimized preliminary control strategy set; the matrix dimensions in the device linkage relationship matrix include the air-conditioning area, lighting circuit and fresh air valve; inputting the optimized control strategy set into the digital twin simulation environment for multi-objective optimization calculation, generating Pareto optimal solutions through the non-dominated sorting genetic algorithm, and finally outputting an executable optimized control parameter package.

2. The intelligent building energy-saving management method based on digital twin technology according to claim 1, characterized in that: the generation of the digital twin database includes the following steps scanning the point cloud data and performing RANSAC algorithm denoising, multi-site point cloud registration and Poisson surface reconstruction to generate a 3D building model; integrating the lighting circuit switch status, air-conditioning unit operation parameters and elevator group control data stream of building automation equipment through the OPC protocol, establishing a lighting component model, a heating, ventilation and air-conditioning component model and a vertical transportation component model associated with the coordinates of the 3D building model, and performing least squares optimization in the Unity3D engine to generate a digital twin database.

3. The intelligent building energy-saving management method based on digital twin technology according to claim 1, wherein: the generation of the building energy consumption baseline report includes the following steps extracting the specification parameters of LED lamps, generating a power curve through cubic polynomial fitting, combining historical dimming records to construct a 3D tensor of the actual power value and 3D operation duration of the lamps, and generating the daily average lighting energy consumption baseline value; retrieving the model parameters of the heating, ventilation and air-conditioning components of the TRNSYS simulation platform, and generating the instantaneous peak cooling load in combination with historical temperature and humidity data of the same period; constructing a Markov chain state transition probability matrix based on the elevator energy consumption pulse signal, and generating the statistical distribution of the daily average power consumption of the elevator through Monte Carlo simulation; integrating the daily average lighting energy consumption baseline value, the hourly peak cooling load of the air-conditioning system and the statistical distribution of the daily average power consumption of the elevator to generate a building energy consumption baseline report.

4. The intelligent building energy-saving management method based on digital twin technology according to claim 1, characterized in that: the output of the executable optimized control parameter package includes the following steps Convert the optimized preliminary control strategy set into an EnergyPlus input file, perform dynamic energy consumption simulation and computational fluid dynamics analysis, and evaluate thermal comfort; Construct a multi-objective function of energy consumption reduction rate and comfort compliance rate, and use the non-dominated sorting genetic algorithm to generate a Pareto optimal solution set; Select the strategy combination with the highest comprehensive score based on the Pareto optimal solution set, and encode it into an optimized control parameter package in the format of device executable instructions.

5. The intelligent building energy-saving management method based on digital twin technology according to claim 4, characterized in that: The device executable instruction format includes a triple of device address code, parameter type, and target value.

6. An intelligent building energy-saving management system based on digital twin technology, based on the intelligent building energy-saving management method based on digital twin technology according to any one of claims 1 to 5, characterized in that: Including, A modeling fusion module that establishes a three-dimensional building model based on building information models and three-dimensional laser scanning point cloud data, and generates a twin database through multi-source device data fusion; An energy consumption benchmark module that performs energy consumption simulation modeling based on the lamp dimming parameters, air-conditioning performance curves, and elevator operation logs in the twin database, and generates an energy consumption benchmark report; An environmental perception module that collects personnel density and environmental parameters, and generates a building environmental state matrix mapped to the twin database after spatio-temporal calibration; A strategy generation module that establishes a hierarchical threshold control strategy library based on the building energy consumption benchmark report and the building environmental state matrix, and generates an optimized control strategy set through a device linkage relationship matrix; A simulation optimization module that inputs the optimized control strategy set into a digital twin simulation environment for multi-objective optimization calculation, generates Pareto optimal solutions through the non-dominated sorting genetic algorithm, and finally outputs an executable optimized control parameter package.

7. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the intelligent building energy-saving management method based on digital twin technology according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the intelligent building energy-saving management method based on digital twin technology according to any one of claims 1 to 5.

Images