BIM-based building energy consumption prediction method and system

Abstract

The present application relates to the technical field of data processing, and more particularly to a building energy consumption prediction method and system based on BIM, which comprises the following steps: obtaining parameters; preliminary screening and partitioning; flow determination; personnel subdivision; equipment subdivision; type aggregation; energy consumption prediction; threshold correction. The present application constructs a spatial partition structure based on a BIM model, organizes and associates multi-source data under unified spatial semantics, constrains the geometric boundaries of each spatial partition, air supply paths and equipment arrangement relationships, forms a corresponding relationship between environmental side parameters and equipment side parameters in the spatial dimension, enables the system to dynamically adjust the determination benchmark according to the spatial diffusion and temporal evolution of abnormalities, thereby forming a closed-loop adjustment mechanism composed of spatial structure constraints, multi-parameter coupling analysis and result feedback correction, and finally improving the accuracy and adaptability of building energy consumption prediction.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a BIM-based method and system for predicting building energy consumption. Background Technology

[0002] With the acceleration of urbanization and the continuous expansion of large public buildings, energy consumption in densely populated areas such as high-speed rail station waiting halls, subway transfer halls, large shopping mall atriums, and convention centers has become a significant component of total social energy consumption. In these semi-enclosed, large spaces, the energy consumption of air conditioning systems is particularly prominent, and their load status is heavily influenced by fluctuations in instantaneous passenger flow density, changes in the intensity of human activity, and the efficiency of equipment coordination and response. However, traditional energy management methods often rely on static setpoints or single-dimensional threshold judgments, making it difficult to accurately capture the coupling relationship between dynamic changes in human activity and environmental response. This leads to frequent problems such as system control lag, energy supply and demand mismatch, and difficulty in tracing abnormal operating conditions. How to achieve accurate prediction of building energy consumption and early identification of abnormal states in complex dynamic environments has become a core challenge that urgently needs to be addressed in the field of green building operation and energy-saving optimization.

[0003] Chinese Patent Application Publication No. CN116955960A discloses a BIM-based energy consumption analysis method, system, and storage medium. The method includes: establishing a 3D model using BIM technology based on building information; performing energy consumption analysis on the established 3D model; installing multiple smart sensors within the building to collect energy consumption data, including room temperature, humidity, wind speed, heat load, cooling load, and energy consumption of all electrical appliances; preprocessing the data before collection by the data collection module; pre-determining the number N of smart sensors and setting up N corresponding sub-data collection modules, each interconnected via a wireless network; pre-defining data groups, each containing N data packets and N dedicated tags, with the data packets and tags corresponding sequentially; the N data packets recording the energy consumption data received by the corresponding sub-data collection module; and the dedicated tags recording whether the corresponding data packet recorded energy consumption data. The smart sensors send the collected energy consumption data to the corresponding sub-data collection module at pre-set time intervals; and the data collection module performs preprocessing. After processing, a sub-data collection module is selected, and an empty data set is sent to it. Energy consumption data is collected through N sub-data collection modules, which store the energy consumption data in the data set. This process continues until the last sub-data collection module stores the stored energy consumption data in the data set, at which point it sends the data set back to the data collection module. The data collection module receives the data set containing the stored data and sends it to the data transmission module. The data transmission module then sends the received data set to a remote server. The remote server receives the data set, checks for any abnormal data, and deletes any abnormal data found. It also checks for missing data and uses mean interpolation to fill in any missing data. The processed data set is then stored in a memory module. The remote server retrieves the 3D model data and combines it with the energy consumption data from the data set in the memory module to perform energy consumption statistical analysis. Based on the analysis results, the relationship between influencing factors and building energy consumption, and the strength of this relationship, is determined, providing data support for building energy-saving solutions.

[0004] Therefore, the existing technology has the following problems: it relies on statically set thresholds and simple data cleaning methods, and the processing of abnormal data is limited to deletion and completion, which easily ignores the dynamic abnormal patterns hidden behind the data, resulting in deviations between the energy consumption analysis results and the actual situation; it relies on ex-post statistical analysis methods to analyze historical energy consumption data, and lacks dynamic monitoring and abnormal early warning mechanisms for real-time operating status, which can easily cause system regulation to lag and fail to respond in a timely manner to load changes caused by instantaneous passenger flow fluctuations. Summary of the Invention

[0005] To address this, the present invention provides a BIM-based building energy consumption prediction method and system, which overcomes the problems of low energy consumption prediction accuracy and delayed anomaly response caused by ignoring dynamic anomaly patterns and lagging regulation in the prior art through a multi-level dynamic anomaly diagnosis mechanism and real-time feedback optimization.

[0006] To achieve the above objectives, in one aspect, the present invention provides a BIM-based building energy consumption prediction method, comprising: Based on the BIM model, the space of a large public building is divided into several spatial zones, and the zone temperature, air volume and air temperature of each spatial zone are obtained in real time. Based on the comparison results of the threshold values ​​of the zone temperature and the air supply volume, several initially screened abnormal zones are selected. The humidity, equipment operating power, and instantaneous passenger flow density of each of the initial screening abnormal zones are acquired in real time. Based on the comparison results of the instantaneous passenger flow density and the preset density threshold, several abnormal personnel load zones are determined according to the coupling change rate of the zone humidity and the air supply temperature, or several abnormal equipment operation zones are determined according to the matching deviation between the equipment operating power and the air supply volume. Based on the temperature and humidity of the abnormal personnel load zones within a preset time-series analysis window, several humidity-heat coupling zones and transient congestion zones are determined. The supply air temperature of each of the abnormal operating zones of the equipment is obtained in real time, and several supply air mismatch zones and efficiency abnormal zones are determined according to the supply air temperature within a preset time delay window and the operating power of the equipment. Several anomaly types are determined based on the heat-humidity coupling zone, the transient congestion zone, the air supply mismatch zone, and the efficiency anomaly zone within the preset prediction period; Energy consumption prediction results are generated based on each of the aforementioned anomaly types; The preset density threshold is adjusted based on the changing trend of the proportion of each anomaly type within the preset observation period after the energy consumption prediction results are generated.

[0007] Furthermore, the process of screening several initially abnormal zones based on the threshold comparison results of the zone temperature and the air supply volume includes: The temperature of each spatial partition is continuously sampled within a preset sampling time, and the temperature deviation of each partition relative to a preset temperature deviation threshold is calculated. The air supply volume of each spatial partition is continuously sampled within the preset sampling time, and the air volume deviation of each air supply volume relative to the preset air volume matching threshold is calculated. Based on the combined distribution characteristics of the temperature deviation and the air volume deviation of each of the spatial partitions, several candidate abnormal partitions that simultaneously meet the preset deviation conditions are selected. Several initial screening abnormal zones are determined based on the temperature deviation and airflow deviation in each of the candidate abnormal zones.

[0008] Furthermore, the process of determining several preliminary screening abnormal zones based on the temperature deviation and airflow deviation in each of the candidate abnormal zones includes: Based on the consistency of the changes in the temperature deviation and the air volume deviation within a preset time window, interference zones caused by instantaneous fluctuations are eliminated to obtain several preliminary screening abnormal zones.

[0009] Furthermore, the process of identifying several zones with abnormal personnel load includes: Based on the comparison results of the instantaneous passenger flow density being greater than the preset density threshold, the humidity-heat coupling deviation amount, which is used to characterize the rate of coupling change, is determined according to the humidity of the zone and the air supply temperature. Based on the threshold comparison results of the humidity-heat coupling deviation amount, several zones with abnormal personnel load are selected from all the initially screened abnormal zones.

[0010] Furthermore, the process of identifying several partitions with abnormal equipment operation includes: Based on the comparison results of the instantaneous passenger flow density being greater than the preset density threshold, the power-air matching deviation amount used to characterize the matching deviation is determined according to the equipment operating power and the air supply volume, and several equipment operation abnormal zones are selected from all the initial screening abnormal zones based on the threshold comparison results of the power-air matching deviation amount.

[0011] Furthermore, the process of determining several heat-dampness coupled zones and transient congestion zones based on the zone temperature and zone humidity within a preset time-series analysis window for the abnormal personnel load zones includes: The temperature and humidity direction consistency is determined based on the zone temperature and the zone humidity. The abnormal personnel load zone is determined to be the heat-humidity coupled zone based on the threshold comparison result of the temperature-humidity direction consistency, or the abnormal personnel load zone is determined to be the transient congestion zone based on the zone temperature, so as to identify several heat-humidity coupled zones and transient congestion zones.

[0012] Furthermore, the process of determining several air supply mismatch zones and efficiency abnormality zones based on the air supply temperature within a preset time delay window and the operating power of the equipment includes: The power-temperature response time difference, used to characterize response consistency, is determined based on the supply air temperature and the equipment operating power. The abnormal equipment operation zone is determined to be the supply air mismatch zone based on the threshold comparison result of the power-temperature response time difference, or the abnormal equipment operation zone is determined to be the efficiency abnormal zone based on the supply air temperature.

[0013] Furthermore, the process of determining several anomaly types based on the moisture-heat coupling zone, the transient congestion zone, the air supply mismatch zone, and the efficiency anomaly zone within a preset prediction period includes: The heat-humidity coupling zone, the transient congestion zone, the air supply mismatch zone, and the efficiency anomaly zone are marked as anomaly zones, and classified according to the zone category of each anomaly zone to obtain several category zones. The duration percentage and frequency percentage of each abnormal partition in each category of partition set within the preset prediction period are statistically analyzed, and the weighted sum of the duration percentage and frequency percentage is calculated to obtain the partition weight corresponding to each abnormal partition. The weights of all partitions within each category partition set are summed to obtain the type index corresponding to each category partition set; Several abnormal types are determined based on the threshold comparison results of each type index.

[0014] Furthermore, the process of correcting the preset density threshold based on the changing trend of the proportion of each of the aforementioned anomaly types within a preset observation period after generating the energy consumption prediction results includes: Calculate the difference sequence of the proportion of the abnormal type at adjacent time points to obtain the cumulative change in proportion used to characterize the trend of change, and correct the preset density threshold based on the threshold comparison result of the cumulative change in proportion.

[0015] On the other hand, the present invention also provides a BIM-based building energy consumption prediction system, comprising: The initial screening unit is used to partition the space of a large public building based on the BIM model to obtain several spatial partitions, and to obtain the partition temperature, air volume and air temperature of each spatial partition in real time. Based on the threshold comparison results of the partition temperature and the air volume, several abnormal partitions are screened out. An abnormal zone determination unit is used to acquire in real time the zone humidity, equipment operating power and instantaneous passenger flow density of each of the initially screened abnormal zones, and based on the comparison result of the instantaneous passenger flow density and the preset density threshold, determine a number of abnormal personnel load zones according to the coupling change rate of the zone humidity and the supply air temperature, or determine a number of abnormal equipment operation zones according to the matching deviation of the equipment operating power and the supply air volume. The first subdivision determination unit is used to determine several heat-humidity coupled zones and transient congestion zones based on the zone temperature and zone humidity of the abnormal personnel load zone within a preset time-series analysis window. The second subdivision determination unit is used to obtain the air supply temperature of each of the abnormal equipment operation zones in real time, and determine a number of air supply mismatch zones and efficiency abnormal zones based on the air supply temperature within a preset time delay window and the operating power of the equipment. The type determination unit is used to determine several abnormality types based on the heat-humidity coupling zone, the transient congestion zone, the air supply mismatch zone, and the efficiency abnormality zone within a preset prediction period. A generation unit is used to generate energy consumption prediction results based on each of the aforementioned anomaly types; The benchmark correction unit is used to correct the preset density threshold based on the changing trend of the proportion of each of the aforementioned anomalies within a preset observation period after the energy consumption prediction results are generated.

[0016] Compared with existing technologies, the beneficial effects of this invention lie in the fact that by constructing a spatial zoning structure based on a BIM model, multi-source data such as zoning temperature, air supply volume, zoning humidity, equipment operating power, and instantaneous passenger flow density are organized and correlated under a unified spatial semantics. The BIM model constrains the geometric boundaries, air supply paths, and equipment layout relationships of each spatial zoning, establishing a spatial correspondence between environmental and equipment parameters. Based on this, the relationship between zoning temperature and air supply volume reflects the heat supply and demand balance; the relationship between zoning humidity and air supply temperature reflects the air heat and humidity exchange process; the relationship between equipment operating power and air supply volume characterizes the matching degree of energy input and transmission capacity; and instantaneous passenger flow density acts as a disturbance source driving the dynamic changes in temperature and humidity parameters. By stratifying the consistency of temperature and humidity change directions, the deviation of power-wind matching, and the time difference of power-temperature response, the system distinguishes between persistent cumulative effects and transient disturbance effects. Furthermore, it characterizes the intensity and stability of anomaly impacts through weighted fusion of duration and frequency of occurrence, enabling unified quantification of various anomalies in both time and intensity dimensions. Simultaneously, relying on the zoning structure in the BIM model, the system achieves spatial aggregation and type aggregation of anomaly zones. The changing trend of anomaly type proportion is introduced to provide feedback correction to the density threshold, allowing the system to dynamically adjust the judgment criteria based on the spatial diffusion and temporal evolution of anomalies. This forms a closed-loop adjustment mechanism consisting of spatial structural constraints, multi-parameter coupling analysis, and result feedback correction, ultimately improving the accuracy and adaptability of building energy consumption prediction. Attached Figure Description

[0017] Figure 1 This is a flowchart of the BIM-based building energy consumption prediction method in this embodiment; Figure 2 This is a flowchart for screening several initially abnormal partitions in this embodiment; Figure 3 This is the logic diagram for determining the interference zone caused by instantaneous fluctuations in this embodiment; Figure 4This is a schematic diagram of the BIM-based building energy consumption prediction system in this embodiment. Detailed Implementation

[0018] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0019] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0020] Please see Figure 1 As shown, this is a flowchart of the BIM-based building energy consumption prediction method of this embodiment. On one hand, this embodiment provides a BIM-based building energy consumption prediction method, including: Based on the BIM model, the space of a large public building is divided into several spatial zones, and the zone temperature, air volume and air temperature of each spatial zone are obtained in real time. Based on the comparison results of the threshold values ​​of the zone temperature and the air supply volume, several initially screened abnormal zones are selected. The humidity, equipment operating power, and instantaneous passenger flow density of each of the initial screening abnormal zones are acquired in real time. Based on the comparison results of the instantaneous passenger flow density and the preset density threshold, several abnormal personnel load zones are determined according to the coupling change rate of the zone humidity and the air supply temperature, or several abnormal equipment operation zones are determined according to the matching deviation between the equipment operating power and the air supply volume. Based on the temperature and humidity of the abnormal personnel load zones within a preset time-series analysis window, several humidity-heat coupling zones and transient congestion zones are determined. The supply air temperature of each of the abnormal operating zones of the equipment is obtained in real time, and several supply air mismatch zones and efficiency abnormal zones are determined according to the supply air temperature within a preset time delay window and the operating power of the equipment. Several anomaly types are determined based on the heat-humidity coupling zone, the transient congestion zone, the air supply mismatch zone, and the efficiency anomaly zone within the preset prediction period; Energy consumption prediction results are generated based on each of the aforementioned anomaly types; The preset density threshold is adjusted based on the changing trend of the proportion of each anomaly type within the preset observation period after the energy consumption prediction results are generated.

[0021] The application objects involved in this embodiment are public building spaces with large volume, open layout and high personnel flow characteristics. Such spaces are usually composed of multiple functional areas, which are geometrically interconnected and have relatively blurred boundaries. The air flow path is complex, and the air supply system is mostly arranged in a centralized or zoned manner. During operation, the personnel density fluctuates significantly over time. Coupled with equipment start-up and shutdown and changes in the external environment, it is easy to cause non-uniform distribution of temperature and humidity in local areas, and further cause mismatch between energy supply and transport. At the same time, due to the large spatial scale and the delayed airflow organization, changes in environmental parameters often have lag propagation and regional coupling characteristics, making it difficult for a single measuring point to reflect the overall state. It is necessary to rely on spatial zoning for systematic analysis in order to more accurately characterize and predict energy consumption changes.

[0022] The process of partitioning the space of a large public building based on a BIM model includes: First, using existing BIM modeling software to create 3D models of the building's structural components, envelope, and HVAC system. The BIM modeling software includes Autodesk Revit, Bentley OpenBuildings, or Graphisoft. Archicad imports architectural drawings during the modeling process or generates components such as walls, floors, ceilings, doors, windows, ducts, and air vents through parametric modeling, assigning spatial coordinates and attribute information to each component. Based on this, according to the building's actual functional division and the air supply coverage of the HVAC system, closed or semi-closed spaces formed by the building envelope are used as basic dividing units. The spaces are further subdivided by considering the location of air vents and the airflow coverage area, dividing areas with relatively consistent airflow organization into the same spatial partition. For open, large spaces, the space is gridded or regionalized according to the duct branch structure and the service range of the air supply terminal equipment, based on the distribution of the air supply path. After partitioning, each spatial partition is assigned a unique identifier, and the association between the partition and corresponding equipment and sensors is established in the BIM model, thus obtaining a spatial partition structure that can be used for subsequent data collection and analysis. Those skilled in the art can implement the above spatial partitioning process based on the spatial division, component attribute definition, and equipment association functions provided by the aforementioned BIM modeling software, combined with HVAC design specifications and conventional modeling procedures.

[0023] In this embodiment, various types of data are acquired through a multi-source sensing and control system deployed on-site. Zoned temperature and humidity are collected in real time by integrated temperature and humidity sensors installed in each spatial zone. These sensors connect to the building automation system via wired or wireless networks and upload data according to a preset sampling period. Air volume is acquired by air volume sensors installed in the air conditioning ducts or by the built-in flow detection module of the variable air volume terminal device. Air supply temperature is acquired by temperature sensors installed in the air supply ducts or at the terminal air outlets. Equipment operating power is monitored and uploaded in real time by the power metering modules or smart meters of the air conditioning unit, fans, and terminal devices. Instantaneous passenger flow density is statistically analyzed using infrared counters deployed at the entrances or key areas of each spatial zone, video surveillance systems combined with target recognition algorithms or millimeter-wave radar, and converted according to unit area. All of the above data are uniformly collected through the building automation system or IoT gateway and mapped and associated with the spatial zoning information in the BIM model, thereby forming a multi-dimensional time-series dataset organized by spatial zoning, providing basic data support for subsequent analysis and processing.

[0024] In this embodiment, the process of generating energy consumption prediction results based on each of the aforementioned anomaly types includes: First, based on the identified humid-heat coupling zones, transient congestion zones, air supply mismatch zones, and efficiency anomaly zones within a preset prediction period, key feature parameters corresponding to each anomaly zone are extracted—for the humid-heat coupling zone, the temperature and humidity coupling deviation and its duration are extracted; for the transient congestion zone, the temperature fluctuation amplitude and peak passenger flow density are extracted; for the air supply mismatch zone, the power-temperature response time difference and power change amplitude are extracted; for the efficiency anomaly zone, the deviation between the air supply temperature change rate and the power ratio is extracted; then, based on historical operating data, a mapping relationship model between each anomaly type and the energy consumption increment per unit time is established, and the feature parameters of each anomaly zone are input into the corresponding model to calculate the energy consumption increment of each anomaly zone. The expected energy consumption increment within the prediction period is calculated. Based on this, the baseline energy consumption of each spatial zone under normal conditions is obtained. This baseline energy consumption is determined based on the average energy consumption value under similar external weather conditions during the same period in history, or based on the theoretical energy consumption value obtained by simulating the current indoor and outdoor environmental parameters using a BIM model. The expected energy consumption increment corresponding to each abnormal zone is superimposed on the baseline energy consumption of the corresponding zone to obtain the corrected predicted energy consumption of each abnormal zone. For spatial zones without abnormalities, their baseline energy consumption is directly used as the predicted energy consumption. Finally, the predicted energy consumption of all spatial zones is accumulated to generate the total predicted energy consumption value of the entire building within the preset prediction period. The prediction confidence interval is calculated based on the fluctuation range of the characteristic parameters of each abnormal zone, and the energy consumption prediction result containing the energy consumption estimate and the confidence interval is output.

[0025] The preset density threshold is the benchmark value for judging the instantaneous passenger flow density within a unit space partition. It depends on the effective area of ​​the space partition, the designed capacity, and the intensity of the disturbance of temperature and humidity environment by personnel activities. By statistically analyzing the temperature and humidity change ranges corresponding to different passenger flow density intervals in historical operating data, the critical density value that causes a significant shift in temperature and humidity parameters is selected as the setting basis. It is usually set between 0.3 and 1.5 people / square meter. In this embodiment, it is set to 0.8 people / square meter, which can trigger anomaly judgment when the personnel density reaches a level that has a significant impact on the environment, while avoiding misjudgment under low density conditions. The preset time series analysis window is the time range used to analyze the synchronicity of temperature and humidity changes in the partition. It depends on the change cycle of environmental parameters in the space and the sampling frequency. It is determined by statistically analyzing the minimum time length required for the temperature and humidity change directions to remain consistent or inconsistent. It is usually set between 5 and 20 minutes. In this embodiment, it is set to 10 minutes, which can ensure the capture of temperature and humidity coupling change trends while suppressing the interference of short-term fluctuations on direction judgment. The preset time delay window is the time range used to analyze the time difference between equipment operating power changes and supply air temperature response. It depends on the air delivery path in the HVAC system. The diameter length, fan response time, and air supply regulation lag characteristics are determined by statistically analyzing the time distribution corresponding to significant responses in air supply temperature after power changes in historical data. This is typically set between 2 and 10 minutes; in this embodiment, it is set to 5 minutes to cover most air supply response processes, thus accurately identifying response lag. The preset prediction period is the time range used to statistically analyze various abnormal zones and calculate type indices. It depends on the statistical period of energy consumption changes and the cumulative effect of abnormal behavior, and is determined by analyzing the correspondence between energy consumption fluctuations and the frequency of abnormal zones in historical data. This is typically set between 30 and 120 minutes; in this embodiment, it is set to 60 minutes to balance short-term fluctuations and medium-term trends, ensuring representativeness in anomaly type determination. The preset observation period is the time range used to analyze the changing trend of anomaly type proportions and correct the density threshold. This depends on the time scale of system state evolution and the stability requirements of threshold adjustment, and is determined by statistically analyzing the time delay by which changes in the proportion of anomaly types affect energy consumption prediction errors in historical data. This is typically set between 2 and 6 hours; in this embodiment, it is set to 4 hours to ensure the reliability of trend judgment while avoiding frequent threshold adjustments, thereby improving overall system stability.

[0026] By constructing a spatial zoning structure based on a BIM model, multi-source data such as zone temperature, air supply volume, zone humidity, equipment operating power, and instantaneous passenger density are organized and correlated under a unified spatial semantics. The BIM model constrains the geometric boundaries, air supply paths, and equipment layout relationships of each spatial zone, establishing a spatial correspondence between environmental and equipment parameters. Based on this, zone temperature and air supply volume reflect the heat supply-demand balance; zone humidity and air supply temperature represent the air heat and moisture exchange process; equipment operating power and air supply volume characterize the matching degree of energy input and transmission capacity; and instantaneous passenger density acts as a disturbance source driving the dynamic changes in temperature and humidity parameters. This is achieved by ensuring consistency in the direction of temperature and humidity changes. The system employs a stratified approach to determine the power-temperature matching deviation and the power-temperature response time difference, distinguishing between persistent cumulative effects and transient disturbance effects. Furthermore, it uses a weighted fusion of duration and frequency of occurrence to characterize the intensity and stability of anomalies, achieving unified quantification of various anomalies in both time and intensity dimensions. Simultaneously, it leverages the zoning structure within the BIM model to achieve spatial aggregation and type aggregation of anomaly zones. The changing trends in the proportion of anomaly types are used to provide feedback correction to the density threshold, enabling the system to dynamically adjust the judgment criteria based on the spatial diffusion and temporal evolution of anomalies. This forms a closed-loop adjustment mechanism comprised of spatial structural constraints, multi-parameter coupled analysis, and result feedback correction, ultimately improving the accuracy and adaptability of building energy consumption prediction.

[0027] Please see Figure 2 As shown, this is a flowchart for screening several initially abnormal zones in this embodiment. In this embodiment, the process of screening several initially abnormal zones based on the threshold comparison results of the zone temperature and the air supply volume includes: The temperature of each spatial partition is continuously sampled within a preset sampling time, and the temperature deviation of each partition relative to a preset temperature deviation threshold is calculated. The air supply volume of each spatial partition is continuously sampled within the preset sampling time, and the air volume deviation of each air supply volume relative to the preset air volume matching threshold is calculated. Based on the combined distribution characteristics of the temperature deviation and the air volume deviation of each of the spatial partitions, several candidate abnormal partitions that simultaneously meet the preset deviation conditions are selected. Several initial screening abnormal zones are determined based on the temperature deviation and airflow deviation in each of the candidate abnormal zones.

[0028] In this embodiment, both the temperature deviation and the airflow deviation are calculated based on the difference between the actual measured value and the corresponding benchmark value. The temperature deviation is the absolute value of the difference between the average value of the zone temperature within a preset sampling period and the reference temperature corresponding to the preset temperature deviation threshold. The airflow deviation is the proportion of the difference between the average value of the supply airflow within a preset sampling period and the reference airflow corresponding to the preset airflow matching threshold to the reference airflow. During the calculation process, the time series data of the zone temperature and supply airflow are denoised and smoothed, and the time averaging method is used to eliminate the influence of instantaneous fluctuations, so that the obtained temperature deviation and airflow deviation can stably reflect the overall deviation of the corresponding spatial zone within the preset sampling period.

[0029] In this embodiment, the process of selecting several candidate abnormal partitions that simultaneously meet preset deviation conditions based on the combined distribution characteristics of temperature deviation and airflow deviation of each spatial partition includes: constructing a two-dimensional deviation feature space with temperature deviation as the first dimension and airflow deviation as the second dimension, mapping the temperature deviation and airflow deviation of each spatial partition to feature points in the two-dimensional deviation feature space; determining the normal distribution area in the two-dimensional deviation feature space based on historical normal operating data, and delineating anomaly judgment areas in the two-dimensional deviation feature space according to preset deviation conditions, wherein the preset deviation conditions are to delineate anomaly judgment areas in the two-dimensional deviation feature space, and the anomaly judgment areas are limited by a joint constraint boundary composed of temperature deviation threshold and airflow deviation threshold. When the temperature deviation and airflow deviation of the spatial partition are simultaneously located outside the joint constraint boundary, it is determined that it meets the preset deviation conditions; traversing the feature points corresponding to each spatial partition, and selecting the spatial partitions that fall into the anomaly judgment areas as candidate abnormal partitions, thereby realizing the identification of the coordinated anomaly of temperature deviation and airflow deviation.

[0030] The preset sampling duration is the time range used to acquire continuous change data of zone temperature and airflow. It depends on the time scale of environmental parameter changes and the system sampling frequency, and is determined by statistically analyzing historical operating data to determine the minimum duration required for zone temperature and airflow to deviate from a stable state and form a clear trend. It is typically set between 5 and 15 minutes; in this embodiment, it is set to 10 minutes to ensure effective trend capture while avoiding interference from instantaneous fluctuations in the deviation calculation results. The preset temperature deviation threshold is the benchmark value for determining whether the zone temperature deviates from the normal operating range. It depends on the design temperature setpoint and the allowable fluctuation range of the ambient temperature. It is determined by statistically analyzing historical operating data to determine the fluctuation range of zone temperature around the setpoint under normal operating conditions, combined with comfort... The suitability standard determines its upper limit range; it is usually set between 1.0 and 3.0℃, and in this embodiment it is set to 2.0℃, which can effectively identify when the temperature deviation reaches a level that affects comfort and energy consumption, while avoiding misjudging small normal fluctuations as abnormalities; the preset air volume matching threshold is the benchmark value for determining whether there is a matching deviation between the supply air volume and the system adjustment requirements, which depends on the design supply air volume and the system adjustment accuracy, and is determined by statistically analyzing the distribution of deviations between the actual supply air volume and the design or target air volume under different load conditions in historical operating data; it is usually set between 10% and 30%, and in this embodiment it is set to 20%, which can trigger anomaly judgment when the air volume supply deviates significantly from the requirements, while taking into account the normal fluctuation range during the system adjustment process.

[0031] By jointly analyzing the deviation of zone temperature and the deviation of air supply volume, the thermal environment and air delivery capacity are coupled and characterized within the same judgment framework. The deviation of zone temperature reflects the degree of heat accumulation or dissipation in the space, while the deviation of air supply volume characterizes the system's ability to regulate and supply the heat change. When both deviate simultaneously, it means that there is an imbalance between heat generation and transport, thus effectively distinguishing between anomalies caused by single sensor fluctuations and real anomalies caused by system regulation imbalances. Furthermore, by combining distribution characteristics to synergistically constrain temperature deviation and air volume deviation, only when there is a consistent deviation between heat change and airflow regulation will the anomaly be included in the candidate anomaly range. The candidate anomalies are then further screened, thereby gradually weakening the impact of random disturbances and short-term fluctuations on the judgment results, improving the stability and reliability of anomaly identification, and ultimately achieving accurate extraction of the initially screened anomaly zones.

[0032] Specifically, the process of determining several preliminary screening abnormal zones based on the temperature deviation and airflow deviation in each of the candidate abnormal zones includes: Based on the consistency of the changes in the temperature deviation and the air volume deviation within a preset time window, interference zones caused by instantaneous fluctuations are eliminated to obtain several preliminary screening abnormal zones.

[0033] Please see Figure 3 As shown, this is the logic diagram for determining interference zones caused by instantaneous fluctuations in this embodiment. In this embodiment, the process of eliminating interference zones caused by instantaneous fluctuations based on the consistency of changes in temperature deviation and airflow deviation within a preset time window includes: acquiring the time series of temperature deviation and airflow deviation within the preset time window; calculating the change direction sequence of the time series of temperature deviation and the change direction sequence of the time series of airflow deviation; calculating the direction consistency ratio of the two change direction sequences within the preset time window to determine the change consistency coefficient; when the change consistency coefficient is less than a preset consistency threshold, the candidate abnormal zone is determined to be an interference zone caused by instantaneous fluctuations and is eliminated.

[0034] The preset time window is the time range used to analyze the consistency of temperature deviation and airflow deviation changes. It depends on the persistence characteristics of environmental parameter changes and the time scale of system adjustment response. It is determined by statistically analyzing the minimum duration for temperature deviation and airflow deviation to maintain the same direction of change in historical operating data. It is usually set between 5 and 15 minutes, and in this embodiment, it is set to 8 minutes, which can effectively filter the interference of short-term fluctuations on consistency judgment while ensuring the capture of stable change trends. The preset consistency threshold is the benchmark ratio value for judging whether the change direction of temperature deviation and airflow deviation is consistent. It depends on the difference in the distribution of the proportion of consistent change direction under real abnormal conditions and random fluctuation conditions in historical data. It is determined by statistically analyzing the probability distribution of the proportion of consistent change direction under different operating conditions and selecting a boundary value that can distinguish between stable anomalies and random disturbances. It is usually set between 0.6 and 0.9, and in this embodiment, it is set to 0.75, which can effectively eliminate interference zones caused by short-term fluctuations while ensuring sensitivity to persistent anomalies.

[0035] By introducing consistency analysis of temperature deviation and airflow deviation over time, the changes in thermal state within a spatial zone are dynamically correlated with the airflow regulation response. Changes in temperature deviation reflect the trend of heat accumulation or release, while changes in airflow deviation reflect the system's regulation direction in response to these thermal changes. When both show consistent changes over time, it indicates a continuous correlation between the system's regulation behavior and environmental changes. Conversely, when their directions of change are inconsistent or frequently reverse, it suggests that the deviation primarily originates from transient disturbances or measurement fluctuations rather than a stable energy imbalance process. By judging the direction of change rather than its amplitude, the interference of short-term spikes and random noise on anomaly identification is effectively reduced. Furthermore, the consistency ratio is used to quantify persistence, making the screening results more stable and reliable. This effectively distinguishes between truly anomalous zones and zones experiencing transient interference, improving the accuracy of initial anomaly zone identification.

[0036] Specifically, the process of identifying several zones with abnormal personnel workload includes: Based on the comparison results of the instantaneous passenger flow density being greater than the preset density threshold, the humidity-heat coupling deviation amount, which is used to characterize the rate of coupling change, is determined according to the humidity of the zone and the air supply temperature. Based on the threshold comparison results of the humidity-heat coupling deviation amount, several zones with abnormal personnel load are selected from all the initially screened abnormal zones.

[0037] In this embodiment, the process of determining the humidity-heat coupling deviation based on zone humidity and supply air temperature includes: acquiring the time series of zone humidity and supply air temperature respectively within a preset analysis period; performing maximum-minimum normalization on the time series of zone humidity and supply air temperature to eliminate dimensional differences; calculating the rate of change sequence of the normalized zone humidity time series and the rate of change sequence of the normalized supply air temperature time series; calculating the difference sequence between humidity change and supply air temperature change based on the two rate of change sequences; calculating the average value of the difference sequence within the preset analysis period as the offset benchmark; calculating the standard deviation of the difference sequence within the preset analysis period as the fluctuation characterization; and weighted fusion of the offset benchmark and the fluctuation characterization to obtain the humidity-heat coupling deviation.

[0038] In this embodiment, the process of selecting several personnel load abnormal zones from all the initial screening abnormal zones based on the threshold comparison result of the humidity-heat coupling deviation includes: when the humidity-heat coupling deviation is greater than the preset coupling deviation threshold, the corresponding initial screening abnormal zone is determined to be a personnel load abnormal zone, so as to select several personnel load abnormal zones.

[0039] The preset coupling deviation threshold is a critical value used to determine whether the amount of humidity-heat coupling deviation reaches the triggering condition of abnormal personnel load. It depends on the statistical distribution characteristics of the difference sequence between the rate of change of humidity and the rate of change of supply air temperature in the historical operation data under different passenger flow density levels. It is determined by dividing the mean and standard deviation of humidity-heat coupling deviation under normal operating conditions and densely populated operating conditions into intervals, and selecting the boundary point that can effectively distinguish the two types of operating conditions. It is usually set between 0.6 and 1.2. In this embodiment, it is set to 0.9, which can ensure sensitivity to the continuous accumulation of humidity and heat caused by dense personnel, while suppressing misjudgments caused by short-term environmental disturbances or local control fluctuations, thereby improving the accuracy of identifying abnormal personnel load zones.

[0040] By using instantaneous passenger flow density as a pre-constraint, subsequent humidity-heat coupling analysis is triggered only when the population density reaches a certain level, ensuring that the interaction between zonal humidity changes and supply air temperature regulation is based on actual load disturbances. Furthermore, by normalizing the rates of change of humidity and supply air temperature and constructing a difference sequence, the offset relationship between the two during the dynamic response process is quantified. The average value reflects the degree of continuous deviation, and the standard deviation reflects the instability of fluctuations. Weighted fusion is then used to comprehensively characterize stable offsets and random fluctuations, thereby effectively distinguishing the different impact paths of continuous personnel load accumulation and short-term environmental disturbances on the system. Based on this, a threshold judgment mechanism is used to screen abnormal zones, ensuring that the anomaly identification results have both physical response consistency and statistical stability.

[0041] Specifically, the process of identifying several partitions with abnormal device operation includes: Based on the comparison results of the instantaneous passenger flow density being greater than the preset density threshold, the power-air matching deviation amount used to characterize the matching deviation is determined according to the equipment operating power and the air supply volume, and several equipment operation abnormal zones are selected from all the initial screening abnormal zones based on the threshold comparison results of the power-air matching deviation amount.

[0042] In this embodiment, the process of determining the power-airflow matching deviation based on the equipment operating power and the air supply volume includes: acquiring the time series of equipment operating power and the time series of air supply volume within a preset analysis period; performing maximum-minimum value normalization on the time series of equipment operating power and the time series of air supply volume; calculating the rate of change of the normalized equipment operating power time series and the rate of change of the normalized air supply volume time series; calculating the ratio of air supply volume to equipment operating power at each time step to obtain the power-airflow ratio sequence; calculating the rate of change of the power-airflow ratio sequence; calculating the difference between the rate of change of the power-airflow ratio and the rate of change of air supply volume at each time step to obtain the power-airflow deviation sequence; calculating the average value of the power-airflow deviation sequence within the preset analysis period as the power-airflow offset; calculating the sum of the absolute values ​​of the differences between adjacent time steps within the preset analysis period as the power-airflow fluctuation; and performing weighted fusion of the power-airflow offset and the power-airflow fluctuation to obtain the power-airflow matching deviation.

[0043] In this embodiment, the process of selecting several abnormal equipment operation partitions from all the initial screening abnormal partitions based on the threshold comparison result of the power-wind matching deviation includes: when the power-wind matching deviation is greater than the preset matching deviation threshold, the corresponding initial screening abnormal partition is determined to be an abnormal equipment operation partition, so as to select several abnormal equipment operation partitions.

[0044] The preset matching deviation threshold is a critical value used to determine whether the power-airflow matching deviation reaches the trigger condition for abnormal equipment operation. It depends on the statistical distribution characteristics of the matching relationship between the equipment's operating power and the airflow volume in historical operating data. By statistically analyzing the mean and fluctuation range of the power-airflow deviation sequence under normal operating conditions, and combining the degree of deviation under abnormal operating conditions, a dividing point that can effectively distinguish between stable matching and mismatch states is selected. It is usually set between 0.5 and 1.0. In this embodiment, it is set to 0.7, which can ensure that the power supply and airflow capacity imbalance during equipment operation is sensitively identified, while suppressing misjudgments caused by short-term adjustment lag or local disturbances, thereby improving the accuracy and stability of the screening of abnormal equipment operation zones.

[0045] By using equipment operating power and air volume as one-to-one corresponding input and output response parameters, the system first normalizes and extracts the rate of change of both, transforming data of different dimensions into comparable dynamic response characteristics. Then, by constructing a power-to-air volume ratio sequence, the system characterizes the air delivery capacity corresponding to unit power, and further analyzes the deviation relationship between its rate of change and the rate of change of air volume. This transforms the matching degree between equipment adjustment behavior and actual air delivery effect into a quantifiable deviation sequence. Among them, the power-to-air volume offset reflects the systematic deviation between energy supply and delivery capacity in long-term operation, while the power-to-air volume fluctuation reflects the instability in short-term adjustment. By weighted fusion of the two, a comprehensive characterization of stable mismatch and transient fluctuation is achieved. Based on this, threshold judgment is used to screen abnormal zones, so that the identification of equipment operation anomalies can not only reflect the coupling relationship between energy input and airflow output, but also take into account the stability characteristics in the time dimension, thereby improving the accuracy and reliability of equipment operation anomaly judgment.

[0046] Specifically, the process of determining several heat-dampness coupled zones and transient congestion zones based on the zone temperature and zone humidity within a preset time-series analysis window for the abnormal personnel load zones includes: The temperature and humidity direction consistency is determined based on the zone temperature and the zone humidity. The abnormal personnel load zone is determined to be the heat-humidity coupled zone based on the threshold comparison result of the temperature-humidity direction consistency, or the abnormal personnel load zone is determined to be the transient congestion zone based on the zone temperature, so as to identify several heat-humidity coupled zones and transient congestion zones.

[0047] In this embodiment, the process of determining the temperature and humidity direction consistency based on zone temperature and zone humidity includes: acquiring the time series of zone temperature and zone humidity respectively within a preset time series analysis window; performing maximum-minimum normalization on the time series of zone temperature and zone humidity; calculating the change direction sequence of the normalized zone temperature time series and the change direction sequence of the zone humidity time series, wherein the change direction sequence is determined by comparing the magnitude relationship of zone temperature or zone humidity at adjacent times, and when the zone temperature or zone humidity at the later time is greater than the corresponding value at the previous time, it is recorded as an increase, and when the zone temperature or zone humidity at the later time is less than the corresponding value at the previous time, it is recorded as a decrease; counting the number of times the two change direction sequences are consistent in direction within the preset time series analysis window, and calculating the proportion of the number of consistent directions to the total number of comparisons, so as to obtain the temperature and humidity direction consistency.

[0048] In this embodiment, the process of determining whether the abnormal personnel load zone is a humidity-heat coupling zone based on the threshold comparison result of temperature and humidity direction consistency, or determining whether the abnormal personnel load zone is a transient congestion zone based on the zone temperature, includes: when the temperature and humidity direction consistency is greater than a preset synchronicity threshold, the corresponding abnormal personnel load zone is determined to be a humidity-heat coupling zone; when the temperature and humidity direction consistency is less than or equal to the preset synchronicity threshold, and the difference between the maximum and minimum values ​​of the zone temperature within a preset time-series analysis window is greater than a preset temperature fluctuation threshold, the corresponding abnormal personnel load zone is determined to be a transient congestion zone; the preset synchronicity threshold is a critical ratio value used to determine whether the consistency between the zone temperature change direction and the zone humidity change direction reaches the humidity-heat coupling determination condition, which depends on the distribution of the direction consistency ratio of the temperature change direction sequence and the humidity change direction sequence under different passenger flow densities and different air conditioning operating conditions in historical operating data. It is determined by statistically analyzing the mean and distribution range of the direction consistency ratio under normal operating conditions and continuous personnel activity conditions, and selecting a boundary that can effectively distinguish between the two types of operating conditions. The temperature fluctuation threshold is typically set between 0.6 and 0.9, and in this embodiment, it is set to 0.75. This ensures sensitive identification of continuous personnel load characterized by synchronous temperature and humidity changes while avoiding misjudgments caused by occasional synchronous fluctuations, thereby improving the accuracy of humidity-heat coupling zoning determination. The preset temperature fluctuation threshold is the critical difference used to determine whether the fluctuation amplitude of the zoning temperature within the preset time-series analysis window reaches the transient congestion determination condition. It depends on the statistical distribution of the difference between the maximum and minimum zoning temperature under different rapid passenger flow change scenarios in historical operation data. It is determined by dividing the temperature fluctuation range under normal stable operating conditions and rapid personnel gathering or evacuation conditions into intervals and selecting a boundary value that can effectively reflect the characteristics of short-term thermal disturbance. It is typically set between 1.5 and 3.5℃, and in this embodiment, it is set to 2.0℃. This ensures sensitive identification of temperature changes caused by instantaneous changes in personnel density while suppressing small fluctuation interference caused by system adjustment lag or environmental noise, thereby improving the reliability of transient congestion zoning determination.

[0049] By analyzing the synchronicity of the changes in zone temperature and humidity within the same time window, the dynamic changes of the two are transformed from continuous numerical values ​​into directional sequences, thus highlighting the synergistic relationship between heat accumulation and water vapor change in the time dimension. When the directions of temperature and humidity change are highly consistent, it indicates that the release of sensible and latent heat caused by human activities is synchronously superimposed in space, causing environmental parameters to exhibit continuous coupled evolution characteristics. When the directional consistency is low, it indicates that the sources of temperature and humidity changes are separate. In this case, further judgment based on the fluctuation amplitude of zone temperature can identify situations where temperature rises and falls rapidly in a short period of time but humidity response is lagging behind as disturbances caused by transient human gathering or evacuation. By jointly judging the directional consistency and temperature fluctuation amplitude, the process of heat and humidity accumulation and the instantaneous crowding effect can be distinguished in terms of mechanism. This reflects the different paths of human activities on the thermal and humid environment and avoids misjudgment caused by a single indicator.

[0050] Specifically, the process of determining several air supply mismatch zones and efficiency abnormal zones based on the air supply temperature within a preset time delay window and the operating power of the equipment includes: The power-temperature response time difference, used to characterize response consistency, is determined based on the supply air temperature and the equipment operating power. The abnormal equipment operation zone is determined to be the supply air mismatch zone based on the threshold comparison result of the power-temperature response time difference, or the abnormal equipment operation zone is determined to be the efficiency abnormal zone based on the supply air temperature.

[0051] In this embodiment, the process of determining the power-temperature response time difference value to characterize response consistency based on the supply air temperature and the equipment operating power includes: acquiring the time series of supply air temperature and the time series of equipment operating power respectively within a preset time delay window; performing maximum-minimum normalization on the time series of supply air temperature and the time series of equipment operating power; calculating the change rate sequence of the normalized equipment operating power time series and the change rate sequence of the supply air temperature time series; determining the time position corresponding to the extreme point in the change rate sequence corresponding to equipment operating power as the power change trigger time; within a preset time delay window after the power change trigger time, determining the time position of the first occurrence of the extreme point in the corresponding change direction in the change rate sequence corresponding to supply air temperature as the temperature response time; and calculating the time difference between the temperature response time and the power change trigger time to obtain the power-temperature response time difference value.

[0052] In this embodiment, the process of determining the abnormal equipment operation zone as the air supply mismatch zone based on the threshold comparison result of the power-temperature response time difference, or determining the abnormal equipment operation zone as the efficiency abnormal zone based on the air supply temperature, includes: when the power-temperature response time difference is greater than a preset delay threshold, the corresponding abnormal equipment operation zone is determined to be an air supply mismatch zone; when the power-temperature response time difference is less than or equal to the preset delay threshold, and the difference between the maximum and minimum values ​​of the air supply temperature change rate within the preset time delay window is less than a preset response amplitude threshold, the corresponding abnormal equipment operation zone is determined to be an efficiency abnormal zone. In this embodiment, when the difference between the power temperature response time and the preset delay threshold is less than or equal to the preset delay threshold and the difference between the maximum and minimum values ​​of the rate of change of the supply air temperature within the preset time delay window is greater than or equal to the preset response amplitude threshold, the original abnormal equipment operation partition type remains unchanged.

[0053] The preset delay threshold is a critical time value used to determine whether the time difference between the change in equipment operating power and the response of the supply air temperature reaches the condition for determining air supply mismatch. It depends on the distribution of the time difference between the trigger time of the change in equipment operating power and the response time of the supply air temperature in historical operating data. By statistically analyzing the response time intervals under different load levels and different duct lengths, and combining the average response time under normal adjustment conditions with the degree of deviation under abnormal lag conditions, a boundary point that can effectively distinguish between timely response and delayed response is selected. It is usually set between 2 and 5 minutes. In this embodiment, it is set to 3 minutes, which can ensure that it has a sensitive identification capability for duct transmission lag or control execution delay while avoiding misjudging short-term response delays in the normal adjustment process as abnormal, thereby improving the determination of air supply mismatch zones. The accuracy of the response amplitude threshold is determined by statistically analyzing the difference between the maximum and minimum values ​​of the air supply temperature change rate under normal and efficiency reduction conditions, and selecting a boundary value that reflects the difference in heat exchange capacity or air supply adjustment capacity. This threshold is typically set between 0.4 and 0.8 °C / min, and in this embodiment, it is set to 0.6 °C / min. This ensures sensitive identification of decreased heat exchange efficiency or insufficient air supply adjustment capacity while suppressing small fluctuations caused by environmental disturbances or measurement noise, thereby improving the reliability of efficiency anomaly zone determination.

[0054] By using changes in equipment operating power as the input trigger signal for system regulation and changes in supply air temperature as the output characterization, a dynamic response relationship between input and output is established using the time positions corresponding to the extreme points of their change rates. This transforms the time transfer process between equipment regulation behavior and air treatment effect into a quantifiable response time difference. When the response time difference increases, it reflects a lag in energy transfer or control execution, causing the air supply effect to fail to keep up with power changes in a timely manner. Conversely, when the response time difference is small but the supply air temperature change amplitude is insufficient, it indicates that although the system has responsiveness, the actual heat exchange or delivery efficiency has decreased. By jointly judging the response time difference and the temperature change amplitude, the timeliness of response in the time dimension and the effectiveness of regulation in the amplitude dimension form a complementary constraint, thereby distinguishing between two different mechanistic problems: air supply mismatch and efficiency anomaly.

[0055] Specifically, the process of determining several anomaly types based on the moisture-heat coupling zone, the transient congestion zone, the air supply mismatch zone, and the efficiency anomaly zone within a preset prediction period includes: The heat-humidity coupling zone, the transient congestion zone, the air supply mismatch zone, and the efficiency anomaly zone are marked as anomaly zones, and classified according to the zone category of each anomaly zone to obtain several category zones. The duration percentage and frequency percentage of each abnormal partition in each category of partition set within the preset prediction period are statistically analyzed, and the weighted sum of the duration percentage and frequency percentage is calculated to obtain the partition weight corresponding to each abnormal partition. The weights of all partitions within each category partition set are summed to obtain the type index corresponding to each category partition set; Several abnormal types are determined based on the threshold comparison results of each type index.

[0056] In this embodiment, the process of classifying abnormal partitions according to their partition categories to obtain several category partition sets includes: obtaining the partition categories of each abnormal partition, where the partition categories include the personnel humidity and heat load category corresponding to the humidity and heat coupling partition, the personnel transient density fluctuation category corresponding to the transient congestion partition, the air supply response mismatch category corresponding to the air supply mismatch partition, and the equipment operating efficiency category corresponding to the efficiency abnormal partition; establishing a category partition set that corresponds one-to-one with each partition category, and initializing each category partition set; traversing all abnormal partitions and adding each abnormal partition to the corresponding category partition set according to its corresponding partition category; during the addition process, recording the spatial partition identifier and the time interval of the abnormal occurrence of each abnormal partition; arranging multiple abnormal time intervals occurring in the same spatial partition within a preset prediction period in chronological order, and merging adjacent time intervals according to a preset time interval threshold to obtain continuous abnormal intervals; retaining the abnormal partitions corresponding to each continuous abnormal interval in the corresponding category partition set; after completing the classification of all abnormal partitions, obtaining the personnel humidity and heat load category partition set, the personnel transient density fluctuation category partition set, the air supply response mismatch category partition set, and the equipment operating efficiency category partition set.

[0057] In this embodiment, during the calculation of the weighted sum of the duration percentage and the frequency percentage, the preset duration percentage weight depends on the degree of continuous impact of the abnormal state on the system's energy consumption. It is determined by statistically analyzing the ratio between the average energy consumption increment of each abnormal partition per unit time and the corresponding duration in historical operating data. It is usually set between 0.5 and 0.8. In this embodiment, it is set to 0.7, which can improve the ability to characterize the cumulative impact of persistent abnormalities on energy consumption, so that abnormal partitions that exist for a long time have a higher weight in the weight calculation. The preset frequency percentage weight depends on the degree of impact of the triggering frequency of the abnormal state on the stability of system operation. It is determined by statistically analyzing the correspondence between the number of abnormal triggers and the energy consumption fluctuation amplitude in historical operating data. It is usually set between 0.2 and 0.5. In this embodiment, it is set to 0.3, which can enhance the ability to identify high-frequency short-term abnormalities, so that frequently occurring abnormal partitions are effectively reflected in the weight calculation.

[0058] In this embodiment, the process of determining several abnormal types based on the threshold comparison results of each type index includes: presetting an abnormal type judgment threshold corresponding one-to-one with each category partition set, including a preset damp heat load threshold, a preset transient density fluctuation threshold, a preset air supply response mismatch threshold, and a preset operating efficiency threshold; comparing the type index corresponding to the personnel damp heat load category partition set with the preset damp heat load threshold, comparing the type index corresponding to the personnel transient density fluctuation category partition set with the preset transient density fluctuation threshold, and comparing the type index corresponding to the air supply response mismatch category partition set with the preset air supply response mismatch threshold; and comparing the... The type index corresponding to the set of equipment operation efficiency category partitions is compared with the preset operation efficiency threshold. When any type index is greater than or equal to the corresponding anomaly type judgment threshold, the partition category corresponding to the type index is determined to constitute the dominant anomaly source in the current preset prediction period, and the dominant anomaly source is determined as the anomaly type. When more than a preset number of type indices are simultaneously greater than or equal to the corresponding anomaly type judgment threshold, they are sorted according to the value of each type index, and the dominant anomaly source corresponding to the preset priority number of partition categories ranked first is determined as the anomaly type. An anomaly type priority list is generated in descending order of type index.

[0059] In this embodiment, when all type indices are less than the corresponding anomaly type determination threshold, it is determined that each category partition set has not formed a dominant anomaly source within the current preset prediction period, and a normal status indicator is output.

[0060] The preset humidity and heat load threshold depends on the degree of deviation between humidity changes caused by personnel activity and the air supply regulation capacity. It is determined by statistically analyzing the distribution range of the average and standard deviation of the difference sequence between humidity change rate and air supply temperature change rate in historical operating data under different passenger flow density ranges. It is usually set between 0.8 and 1.5, and in this embodiment, it is set to 1.0, which can ensure sensitivity to the continuous humidity and heat accumulation effect caused by dense personnel while avoiding misjudging short-term environmental disturbances as abnormal personnel load. The preset transient density fluctuation threshold depends on the degree of influence of rapid gathering or evacuation of personnel in a short period of time on the temperature fluctuation amplitude of the zone. It is determined by statistically analyzing the distribution range of the difference between the maximum and minimum zone temperatures corresponding to high-frequency passenger flow fluctuation periods in historical operating data. It is usually set between 2.0 and 4.0℃, and in this embodiment, it is set to 2.5℃, which can effectively distinguish between continuous heat accumulation and short-term human activity. Temperature fluctuations caused by overcrowding; the preset air supply response mismatch threshold depends on the hysteresis characteristic of the air supply temperature response after a change in equipment operating power, and is determined by statistically analyzing the time difference distribution between the power change trigger time and the air supply temperature change response time in historical operating data; it is usually set between 2 and 5 minutes, and in this embodiment it is set to 3 minutes, which can accurately identify the problem of untimely air supply response caused by duct lag or control adjustment delay; the preset operating efficiency threshold depends on the effective amplitude of the air supply temperature change rate under the condition of equipment operating power change, and is determined by statistically analyzing the difference between the maximum and minimum values ​​of the air supply temperature change rate under normal and abnormal operating conditions in historical operating data; it is usually set between 0.4 and 0.8℃ / min, and in this embodiment it is set to 0.6℃ / min, which can effectively reflect the actual adjustment capability of the equipment under unit power change, thereby distinguishing between efficiency decline and normal response.

[0061] The preset time interval threshold is used to determine whether adjacent abnormal time intervals are continuous when merging multiple abnormal time intervals occurring within a preset prediction period in the same spatial partition. It depends on the system's sensitivity to abnormal continuity determination and the characteristics of cumulative energy consumption effects. It is typically set between 1 and 10 minutes; in this embodiment, it is set to 5 minutes, effectively treating brief interruptions as continuous abnormalities, thus accurately reflecting the persistence of abnormal partitions. The preset judgment quantity is used to control the upper limit of the judgment abnormality type when multiple type indices simultaneously exceed the corresponding abnormality type judgment threshold. It depends on the number of dominant abnormality sources the system focuses on and the decision complexity. It is typically set between 2 and 5; in this embodiment, it is set to 3, ensuring that only the abnormality types with the most significant impact on energy consumption are selected, avoiding excessive interference from multiple major abnormalities in the overall judgment. The preset priority quantity is used to select higher-priority abnormality types for output after sorting by type indices. It depends on the degree of attention paid to the priority of abnormal types and the needs of operation and maintenance resource allocation. It is typically set between 1 and 3; in this embodiment, it is set to 2, highlighting the most critical abnormality types and focusing operation and maintenance and energy consumption control on the main influencing factors.

[0062] By uniformly classifying anomalies from different sources according to their categories, personnel load and equipment operation—two different impact paths—are made comparable within the same analytical framework. The duration and frequency of anomalies are jointly quantified over time, with the duration percentage reflecting the cumulative effect of the anomaly on system energy consumption and the frequency percentage reflecting the degree of disturbance to system operational stability. Weighted fusion achieves a unified representation of two different impact forms: long-term deviations and high-frequency fluctuations. Furthermore, at the category set level, the weights of each region are accumulated to form a type index, transforming spatially discrete local anomalies into statistically dominant overall characteristics, thus establishing a correlation between local physical disturbances and overall energy consumption performance. Based on this, threshold judgment is used to identify the dominant anomaly source, ensuring that the final anomaly type not only reflects the scale effect of the anomaly distribution but also its dominant influence on the overall system operation. This improves the global consistency and decision-making effectiveness of anomaly type determination, ultimately achieving accurate identification of the sources of building energy consumption anomalies.

[0063] Specifically, the process of correcting the preset density threshold based on the changing trend of the proportion of each anomaly type within a preset observation period after generating the energy consumption prediction results includes: Calculate the difference sequence of the proportion of the abnormal type at adjacent time points to obtain the cumulative change in proportion used to characterize the trend of change, and correct the preset density threshold based on the threshold comparison result of the cumulative change in proportion.

[0064] In this embodiment, the process of calculating the difference sequence of adjacent time points of the proportion of anomaly types includes: within a preset observation period, obtaining the proportion of anomaly types corresponding to each sampling time point, and arranging them in chronological order to form a proportion time series; calculating the difference between the proportions of adjacent time points in the proportion time series to obtain a proportion difference sequence. The process of calculating the cumulative change in proportion includes: summing up all the differences in the proportion difference sequence to obtain the cumulative change in proportion. A positive cumulative change in proportion indicates that the proportion of the corresponding abnormal type is on the rise, and a negative cumulative change in proportion indicates that the proportion of the corresponding abnormal type is on the fall. The process of correcting the preset density threshold based on the threshold comparison results of the cumulative change in the proportion includes: when the cumulative change in the proportion is greater than the preset upward trend threshold, the preset density threshold is reduced by the preset adjustment step size. In this embodiment, when the cumulative change in the proportion is less than the preset downward trend threshold, the preset density threshold is increased by the preset adjustment step size; when the cumulative change in the proportion is between the preset upward trend threshold and the preset downward trend threshold, the preset density threshold is kept unchanged.

[0065] The preset upward trend threshold depends on the cumulative change in the proportion of personnel load-related anomalies that significantly deviates from energy consumption prediction results during a continuous increase in the proportion of such anomalies within historical observation periods. This is determined by statistically analyzing the correlation between the cumulative change in proportion and the increment of energy consumption prediction error over multiple observation periods. It is typically set between 0.1 and 0.3; in this embodiment, it is set to 0.15. This setting triggers density threshold adjustment when the proportion of anomalies continues to increase and has a significant impact on energy consumption, while avoiding overreaction to short-term fluctuations. The preset adjustment step size depends on the sensitivity of density threshold changes to the determination of personnel load anomaly zones. This is achieved by comparing and analyzing the changes in the number of anomaly zones under different step sizes in historical data, selecting the appropriate step size. The step size for a stable change in the number of abnormal zones without significant oscillations is typically set between 0.05 and 0.2. In this embodiment, it is set to 0.1, which ensures effective adjustment while avoiding frequent and large fluctuations in the density threshold, thus maintaining the stability of the judgment results. The preset downward trend threshold depends on the cumulative change in the proportion of personnel load-related abnormal types during the continuous decline in the historical observation period. It is determined by statistically analyzing the range of values ​​when the cumulative change in proportion is negative and the system energy consumption tends to stabilize. It is typically set between -0.3 and -0.1. In this embodiment, it is set to -0.15, which allows for timely relaxation of the density threshold when the proportion of abnormal types decreases significantly and the system tends to stabilize, while avoiding frequent adjustments due to slight downward trends.

[0066] The cumulative change in the proportion of anomalies is obtained by summing the time-series differences between adjacent time points. This reflects the trend of personnel load or equipment anomalies within the observation period. When the cumulative change in the proportion is positive and exceeds the preset upward trend threshold, it indicates that the number of anomalies is gradually increasing in the overall building operation. In this case, appropriately reducing the preset density threshold can improve the sensitivity to areas with excessively high personnel density and identify potential load anomalies in a timely manner. When the cumulative change in the proportion is negative and below the preset downward trend threshold, it indicates that the number of anomalies is decreasing in the overall operation. In this case, increasing the preset density threshold can avoid over-responding to occasional low-density fluctuations. When the cumulative change in the proportion is between the upward and downward thresholds, the proportion of anomalies is relatively stable. Maintaining the preset density threshold unchanged can prevent unnecessary adjustments, achieving adaptive control of personnel load anomalies and improving the accuracy and real-time response capability of building energy consumption prediction.

[0067] Please see Figure 4 As shown, this is a schematic diagram of the BIM-based building energy consumption prediction system of this embodiment. Furthermore, this embodiment also provides a BIM-based building energy consumption prediction system, including: The initial screening unit is used to partition the space of a large public building based on the BIM model to obtain several spatial partitions, and to obtain the partition temperature, air volume and air temperature of each spatial partition in real time. Based on the threshold comparison results of the partition temperature and the air volume, several abnormal partitions are screened out. An abnormal zone determination unit is connected to the primary screening unit to obtain the zone humidity, equipment operating power and instantaneous passenger flow density of each of the primary screening abnormal zones in real time. Based on the comparison result of the instantaneous passenger flow density and the preset density threshold, it determines several abnormal personnel load zones according to the coupling change rate of the zone humidity and the supply air temperature, or determines several abnormal equipment operation zones according to the matching deviation of the equipment operating power and the supply air volume. The first subdivision determination unit is connected to the abnormal zone determination unit and is used to determine several humid-heat coupling zones and transient congestion zones based on the zone temperature and zone humidity of the abnormal personnel load zone within a preset time series analysis window. The second subdivision determination unit is connected to the abnormal partition determination unit to obtain the air supply temperature of each abnormal partition of the equipment in real time, and to determine a number of air supply mismatch partitions and efficiency abnormal partitions based on the air supply temperature within a preset time delay window and the operating power of the equipment. A type determination unit, which is connected to the first subdivision determination unit and the second subdivision determination unit respectively, is used to determine several abnormal types based on the heat-humidity coupling zone, the transient congestion zone, the air supply mismatch zone and the efficiency abnormal zone within a preset prediction period. A generation unit, which is connected to a type determination unit, is used to generate energy consumption prediction results based on each of the aforementioned anomaly types; The benchmark correction unit is connected to the generation unit, the type determination unit, and the anomaly partition determination unit, respectively, and is used to correct the preset density threshold according to the changing trend of the proportion of each anomaly type within a preset observation period after generating the energy consumption prediction result.

[0068] By real-time monitoring, anomaly identification, and categorized analysis of multi-dimensional information such as building space zoning, environmental parameters, equipment operating status, and personnel density, and by dynamically adjusting density thresholds based on anomaly type trends, the system achieves refined prediction and adaptive control of building energy consumption changes, thereby improving the accuracy and response speed of energy management.

[0069] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A BIM-based building energy consumption prediction method, characterized in that, include: Based on the BIM model, the space of a large public building is divided into several spatial zones, and the zone temperature, air volume and air temperature of each spatial zone are obtained in real time. Based on the comparison results of the threshold values ​​of the zone temperature and the air supply volume, several initially screened abnormal zones are selected. The humidity, equipment operating power, and instantaneous passenger flow density of each of the initial screening abnormal zones are acquired in real time. Based on the comparison results of the instantaneous passenger flow density and the preset density threshold, several abnormal personnel load zones are determined according to the coupling change rate of the zone humidity and the air supply temperature, or several abnormal equipment operation zones are determined according to the matching deviation between the equipment operating power and the air supply volume. Based on the temperature and humidity of the abnormal personnel load zones within a preset time-series analysis window, several humidity-heat coupling zones and transient congestion zones are determined. The supply air temperature of each of the abnormal operating zones of the equipment is obtained in real time, and several supply air mismatch zones and efficiency abnormal zones are determined according to the supply air temperature within a preset time delay window and the operating power of the equipment. Several anomaly types are determined based on the heat-humidity coupling zone, the transient congestion zone, the air supply mismatch zone, and the efficiency anomaly zone within the preset prediction period; Energy consumption prediction results are generated based on each of the aforementioned anomaly types; The preset density threshold is adjusted based on the changing trend of the proportion of each anomaly type within the preset observation period after the energy consumption prediction results are generated.

2. The BIM-based building energy consumption prediction method according to claim 1, characterized in that, The process of screening several initially abnormal zones based on the comparison results of the threshold values ​​of the zone temperature and the supply air volume includes: The temperature of each spatial partition is continuously sampled within a preset sampling time, and the temperature deviation of each partition relative to a preset temperature deviation threshold is calculated. The air supply volume of each spatial partition is continuously sampled within the preset sampling time, and the air volume deviation of each air supply volume relative to the preset air volume matching threshold is calculated. Based on the combined distribution characteristics of the temperature deviation and the air volume deviation of each of the spatial partitions, several candidate abnormal partitions that simultaneously meet the preset deviation conditions are selected. Several initial screening abnormal zones are determined based on the temperature deviation and airflow deviation in each of the candidate abnormal zones.

3. The BIM-based building energy consumption prediction method according to claim 2, characterized in that, The process of determining several preliminary screening abnormal zones based on the temperature deviation and airflow deviation in each of the candidate abnormal zones includes: Based on the consistency of the changes in the temperature deviation and the air volume deviation within a preset time window, interference zones caused by instantaneous fluctuations are eliminated to obtain several preliminary screening abnormal zones.

4. The BIM-based building energy consumption prediction method according to claim 3, characterized in that, The process of identifying several areas with abnormal personnel workload includes: Based on the comparison results of the instantaneous passenger flow density being greater than the preset density threshold, the humidity-heat coupling deviation amount, which is used to characterize the rate of coupling change, is determined according to the humidity of the zone and the air supply temperature. Based on the threshold comparison results of the humidity-heat coupling deviation amount, several zones with abnormal personnel load are selected from all the initially screened abnormal zones.

5. The BIM-based building energy consumption prediction method according to claim 4, characterized in that, The process of identifying several partitions with abnormal equipment operation includes: Based on the comparison results of the instantaneous passenger flow density being greater than the preset density threshold, the power-air matching deviation amount used to characterize the matching deviation is determined according to the equipment operating power and the air supply volume, and several equipment operation abnormal zones are selected from all the initial screening abnormal zones based on the threshold comparison results of the power-air matching deviation amount.

6. The BIM-based building energy consumption prediction method according to claim 5, characterized in that, The process of determining several heat-humidity coupled zones and transient congestion zones based on the zone temperature and humidity within a preset time-series analysis window for the abnormal personnel load zones includes: The temperature and humidity direction consistency is determined based on the zone temperature and the zone humidity. The abnormal personnel load zone is determined to be the heat-humidity coupled zone based on the threshold comparison result of the temperature-humidity direction consistency, or the abnormal personnel load zone is determined to be the transient congestion zone based on the zone temperature, so as to identify several heat-humidity coupled zones and transient congestion zones.

7. The BIM-based building energy consumption prediction method according to claim 6, characterized in that, The process of determining several air supply mismatch zones and efficiency abnormal zones based on the air supply temperature within a preset time delay window and the operating power of the equipment includes: The power-temperature response time difference, used to characterize response consistency, is determined based on the supply air temperature and the equipment operating power. The abnormal equipment operation zone is determined to be the supply air mismatch zone based on the threshold comparison result of the power-temperature response time difference, or the abnormal equipment operation zone is determined to be the efficiency abnormal zone based on the supply air temperature.

8. The BIM-based building energy consumption prediction method according to claim 7, characterized in that, The process of determining several anomaly types based on the heat-humidity coupling zone, the transient congestion zone, the air supply mismatch zone, and the efficiency anomaly zone within a preset prediction period includes: The heat-humidity coupling zone, the transient congestion zone, the air supply mismatch zone, and the efficiency anomaly zone are marked as anomaly zones, and classified according to the zone category of each anomaly zone to obtain several category zones. The duration percentage and frequency percentage of each abnormal partition in each category of partition set within the preset prediction period are statistically analyzed, and the weighted sum of the duration percentage and frequency percentage is calculated to obtain the partition weight corresponding to each abnormal partition. The weights of all partitions within each category partition set are summed to obtain the type index corresponding to each category partition set; Several abnormal types are determined based on the threshold comparison results of each type index.

9. The BIM-based building energy consumption prediction method according to claim 8, characterized in that, The process of correcting the preset density threshold based on the changing trend of the proportion of each of the aforementioned anomaly types within a preset observation period after generating energy consumption prediction results includes: Calculate the difference sequence of the proportion of the abnormal type at adjacent time points to obtain the cumulative change in proportion used to characterize the trend of change, and correct the preset density threshold based on the threshold comparison result of the cumulative change in proportion.

10. A BIM-based building energy consumption prediction system, constructed based on the BIM-based building energy consumption prediction method according to any one of claims 1-9, characterized in that, include: The initial screening unit is used to partition the space of a large public building based on the BIM model to obtain several spatial partitions, and to obtain the partition temperature, air volume and air temperature of each spatial partition in real time. Based on the threshold comparison results of the partition temperature and the air volume, several abnormal partitions are screened out. An abnormal zone determination unit is used to acquire in real time the zone humidity, equipment operating power and instantaneous passenger flow density of each of the initially screened abnormal zones, and based on the comparison result of the instantaneous passenger flow density and the preset density threshold, determine a number of abnormal personnel load zones according to the coupling change rate of the zone humidity and the supply air temperature, or determine a number of abnormal equipment operation zones according to the matching deviation of the equipment operating power and the supply air volume. The first subdivision determination unit is used to determine several damp-heat coupled zones and transient congestion zones based on the zone temperature and zone humidity of the abnormal personnel load zone within a preset time-series analysis window. The second subdivision determination unit is used to obtain the air supply temperature of each of the abnormal equipment operation zones in real time, and determine a number of air supply mismatch zones and efficiency abnormal zones based on the air supply temperature within a preset time delay window and the operating power of the equipment. The type determination unit is used to determine several abnormality types based on the heat-humidity coupling zone, the transient congestion zone, the air supply mismatch zone, and the efficiency abnormality zone within a preset prediction period. A generation unit is used to generate energy consumption prediction results based on each of the aforementioned anomaly types; The benchmark correction unit is used to correct the preset density threshold based on the changing trend of the proportion of each of the aforementioned anomalies within a preset observation period after the energy consumption prediction results are generated.

Images