Building equipment operation state monitoring system and method

Abstract

The invention discloses a building equipment operation state monitoring system and method, and relates to the technical field of equipment monitoring, and the system comprises a background management device which is used for carrying out the management, data processing and analysis operation of building equipment, supervising the data information of a plurality of collection devices in real time, and employing an abnormality detection model based on an attention mechanism to carry out the monitoring of the operation state of the building equipment. Building equipment abnormity identification and maintenance operation are realized; the dynamic acquisition device is used for acquiring dynamic images and physical quantity data; the static acquisition device is used for acquiring static images and physical quantity data; and the user side is used for receiving the equipment operation state information, the abnormal alarm and the processing suggestion. Through cooperation of the movable inspection platform and the fixed detection point of the static acquisition device, image data, physical quantity parameters and equipment operation logs are synchronously acquired, multi-dimensional monitoring of vision, physical quantity and historical data is realized, a whole scene of fixed equipment and an inspection area is covered, a data blind area is avoided, and waste of more manpower is not needed.

Description

Technical Field

[0001] This invention relates to the field of equipment monitoring technology, specifically a building equipment operation status monitoring system and method. Background Technology

[0002] With the acceleration of urbanization, the scale and complexity of modern buildings are constantly increasing. Building equipment, as an important component of buildings, plays a crucial role in the normal operation, energy consumption, comfort, and safety of buildings. Building equipment refers to various devices used in residential and commercial buildings, encompassing multiple important systems such as environmental and energy systems, security and monitoring systems, and intelligent systems. Effective monitoring of the operating status of building equipment is a key link in achieving efficient building management, energy conservation and emission reduction, and ensuring personnel safety. Existing methods for monitoring the operating status of building equipment mostly rely on manual inspections or decentralized management models. However, given the large number and wide distribution of building equipment, manual inspections are time-consuming and labor-intensive, easily affected by subjective factors, and result in limited data acquisition, leading to missed or incorrect inspections. In the decentralized management model, each equipment system operates independently without effective integration and analysis, making it difficult to obtain overall equipment operating status and energy consumption information, resulting in monitoring blind spots and the inability to monitor the operating status of some equipment in real time.

[0003] Based on this, a building equipment operation status monitoring system and method are provided, which can eliminate the drawbacks of existing technical solutions. Summary of the Invention

[0004] The purpose of this invention is to provide a building equipment operation status monitoring system and method to solve the problem that manual inspection or decentralized management modes in the background art cannot effectively monitor the operation status of building equipment.

[0005] To achieve the above objectives, the present invention provides the following technical solution: A building equipment operation status monitoring system, comprising: The back-end management device is used to manage, process, and analyze building equipment, monitor data from multiple acquisition devices in real time, and use an attention-based anomaly detection model to identify and maintain abnormalities in building equipment. Dynamic data acquisition device is used to collect dynamic images and physical quantity data of building equipment in the inspection area and in buildings where fixed monitoring is inconvenient; Static data acquisition devices are used to collect static images and physical quantity data of fixed-installation building equipment during operation; The user terminal is used to receive equipment operation status information, abnormal alarms and handling suggestions pushed by the background management device, and to send query instructions and operation requests to the background management device. The background management device, dynamic acquisition device, static acquisition device and user terminal communicate with each other via a network. The back-end management device includes: The database is used to collect, store, and provide real-time data, historical data, initial operating parameter data, and equipment maintenance knowledge of building equipment, and to divide the data in the corresponding information into several groups according to time, type, and parameter characteristics; The analysis module determines whether there are any abnormalities in the operating status of building equipment based on the collected data. The processor is used to receive and output information results from the background management device. Memory is used to receive coordinated instructions from the processor and to control information according to those instructions. The display panel is used to present building equipment operation data to administrators in a visual format.

[0006] Preferably, the dynamic data acquisition device includes several mobile inspection platforms. The inspection platforms are used to move inside the building according to a predetermined inspection path to replace inspection personnel in performing inspection operations. Each inspection platform is equipped with a communication module, a voice interaction module, and a positioning module. The voice interaction module and the positioning module are both connected to the back-end management device through the communication module. Each of the inspection platforms is equipped with a first camera device, a first environmental sensing device, and a first edge computing device. The first camera device is used to capture images and video data in real time during the inspection process; The first environmental sensing device is used to sense various physical quantities during the inspection process. The first environmental sensing device includes several temperature and humidity sensors, gas sensors, smoke sensors, light sensors and noise sensors. The first edge computing device is used to preprocess the collected data and to receive and output image, video and physical quantity data.

[0007] Preferably, the static acquisition device includes several second camera devices fixedly installed at various detection points within the building. Second environmental sensing devices and second edge computing devices are provided on both the left and right sides of the second camera devices. The building equipment is equipped with corresponding sensing and detection devices inside. The second camera device is used to collect images and video data from fixed monitoring points within the building in real time; The second environmental sensing device is used to collect various physical quantities at fixed monitoring points within the building in real time. The second environmental sensing device includes several temperature and humidity sensors, gas sensors, smoke sensors, light sensors, noise sensors, and vibration sensors. The second edge computing device is used to preprocess the collected data and receive and output image, video and physical quantity data. The second edge computing device has the same structural features as the first edge computing device. The sensing and detection equipment is used to monitor the operating status of various equipment in the building in real time and collect the corresponding operating parameters of the equipment itself. The operating parameters include the voltage, current and power factor of the power distribution equipment, the air outlet temperature and compressor pressure of the air conditioner, the running speed, car position and door lock status of the elevator, and the pipe pressure, flow rate and pump speed of the water supply and drainage equipment.

[0008] Preferably, the analysis module includes: The data preprocessing module is used to perform preprocessing operations on the collected equipment operation data. The preprocessing operations include data cleaning, data transformation, data normalization, and data compression. The feature extraction module is used to extract key building equipment operation features and corresponding attribute values ​​from preprocessed equipment operation data; The fusion module is used to merge the operational characteristics of key building equipment attributes to construct a comprehensive set of building equipment operational status characteristics. The comparison module is used to compare the extracted and fused building equipment operation characteristics with the historical characteristic data of the building equipment during normal operation and its own standard operation parameter data thresholds in the database in turn to determine whether there are any abnormal characteristics that exceed the normal range. The decision module is used to determine the anomaly level and generate corresponding handling suggestions based on the analysis results of the comparison module, combined with the historical processing records of anomaly features in the database and equipment maintenance knowledge. The output module is used to convert the analysis results of the decision module into standard format suggestion data and store it in the database, and synchronously push the standard format suggestion data to the display panel and the user terminal of maintenance personnel.

[0009] Preferably, the user terminal is a mobile phone, tablet or computer device, and supports login operations through a visual interface of APP, mini program or web page. The user group of the user terminal includes administrators, inspection personnel, maintenance personnel and building equipment users.

[0010] Preferably, both the background management device and the inspection platform are equipped with audible and visual alarms on one side, and the audible and visual alarms communicate with the processor and communication module through a network.

[0011] Preferably, both the first and second camera devices adopt a dual-image acquisition mode combining a high-precision camera and an infrared thermal imager.

[0012] Preferably, the attention mechanism employs a hybrid attention mechanism that combines channel attention and spatial attention.

[0013] Preferably, the anomaly detection model is based on a deep learning framework and includes a multimodal data processing layer, a hybrid attention mechanism layer, and an anomaly decision layer. The input features of the anomaly detection model are multimodal data, and the output features are anomaly identification results.

[0014] A method for monitoring the operational status of building equipment, the specific steps of which are as follows: Step S1: Collect images and physical quantity data inside the building using dynamic acquisition devices and static acquisition devices respectively, and transmit the collected building equipment operation data to the background management device. Step S2: After receiving the data, the background management device groups and stores the data in the database according to time, type and parameter characteristics. The analysis module extracts the pre-processed building equipment operation characteristics for comparative analysis to determine whether the equipment operation status is abnormal. Step S3: If abnormal features exist, the abnormal features and abnormal attribute values ​​are marked. The background management device generates an abnormal alarm and handling suggestions, and pushes them to the user terminals of maintenance personnel and administrators. If no abnormal features exist, the data is marked as normal and stored in the database according to the grouping rules.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention designs a building equipment operation status monitoring system. By cooperating a mobile inspection platform with a dynamic acquisition device and a fixed detection point with a static acquisition device, it synchronously collects image data, physical parameters, and equipment operation logs, realizing multi-dimensional monitoring of visual, physical, and historical data. It covers the entire scene of fixed equipment and inspection areas, avoiding data blind spots and requiring minimal manpower. The anomaly detection model is based on a deep learning framework and adopts a hybrid mechanism combining channel attention and spatial attention. It can automatically focus on key features, improving the accuracy of anomaly identification. Combined with historical data and maintenance knowledge in the database, it can identify equipment anomalies, achieve early maintenance, and extend the equipment life cycle. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the overall architecture of the monitoring system.

[0017] Figure 2 This is a schematic diagram of the back-end management device of the monitoring system.

[0018] Figure 3 This is a schematic diagram of the analysis module of the monitoring system.

[0019] Figure 4This is a schematic diagram of the dynamic acquisition device for the monitoring system.

[0020] Figure 5 This is a schematic diagram of the inspection platform of the monitoring system.

[0021] Figure 6 This is a schematic diagram of the static data acquisition device for the monitoring system.

[0022] Figure 7 This is a schematic diagram of the steps in the method of the present invention.

[0023] Figure reference numerals: Background management device 100, database 110, analysis module 120, data preprocessing module 121, feature extraction module 122, fusion module 123, comparison module 124, decision module 125, output module 126, processor 130, memory 140, display panel 150, dynamic acquisition device 200, inspection platform 210, communication module 211, voice interaction module 212, positioning module 213, first camera device 220, first environmental perception device 230, first edge computing device 240, static acquisition device 300, second camera device 310, second environmental perception device 320, second edge computing device 330, sensing and detection device 340, user terminal 400. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0025] In this embodiment, as Figures 1-7 As shown, a building equipment operation status monitoring system includes: The background management device 100 is used to manage, process and analyze building equipment, monitor data information from multiple acquisition devices in real time, and use an attention-based anomaly detection model to identify and maintain abnormalities in building equipment. Specifically, the attention mechanism adopts a hybrid attention mechanism that combines channel attention and spatial attention. The anomaly detection model is based on a deep learning framework and includes a multimodal data processing layer, a hybrid attention mechanism layer, and an anomaly decision layer. The input features of the anomaly detection model are multimodal data, and the output features are anomaly identification results. Specifically, the multimodal data processing layer integrates convolutional neural network units and recurrent neural network units. The convolutional neural network units are used to extract features from the image data collected by the first camera device 220 of the dynamic acquisition device 200 and the second camera device 310 of the static acquisition device 300 to generate spatial feature vectors of the equipment appearance and operating status. The recurrent neural network units are used to process the time-series physical quantity data collected by the first environmental sensing device 230, the second environmental sensing device 320 and the sensing and detection device 340, such as the voltage or current of the power distribution equipment, the air outlet temperature or compressor pressure of the air conditioner, the running speed or car position of the elevator, etc., to generate time-series feature vectors of the equipment operating parameters. The hybrid attention mechanism layer embeds a channel attention submodule and a spatial attention submodule. The channel attention submodule is used to assign weights to the importance of different parameter channels in the feature vector output by the multimodal data processing layer, strengthening the channel features that are highly correlated with device anomalies. The hybrid attention mechanism assigns weights to each feature. The spatial attention submodule is used to focus on the spatial distribution features of key device regions and physical quantity data in the image data, highlighting the local features of the anomaly location. The anomaly decision layer compares the key feature vector output by the hybrid attention mechanism layer with the historical feature data of normal operation of building equipment stored in the database 110, its own initial operating parameter thresholds, and equipment maintenance knowledge. It calculates the feature deviation through fully connected network units. When the deviation exceeds the preset threshold, it is determined to be an abnormal state. It also generates the corresponding anomaly level and maintenance suggestions by combining historical anomaly handling records and outputs them to the processor 130 and the user terminal 400. The anomaly levels include minor anomaly, moderate anomaly, and severe anomaly. The maintenance suggestions include component inspection, parameter calibration, and emergency shutdown. The feature deviation is calculated as follows: for each feature i, calculate its current value. Compared with normal benchmark The difference between (historical mean or standard threshold) is expressed by the following formula: ,in, The average of historical normal data or a preset standard threshold, The standard deviation of historical normal data. For single feature deviation, a hybrid attention mechanism is used to assign weights to each feature. The weighted deviations of all features are summed and then mapped to the [0,1] range using the Sigmoid activation function to obtain the final feature deviations: n is the total number of features. Whether it is abnormal is determined based on the deviation threshold. The deviation threshold is adjusted according to actual needs. Specifically, the anomaly detection model is pre-trained using historical data. The input features include multimodal data, which includes image features, physical quantity features, and historical correlation features. Image features include dynamic or static images and video data of the equipment collected by the first camera device 220 of the dynamic acquisition device 200 and the second camera device 310 of the static acquisition device 300, including the appearance of the equipment, the scene of operation, and the thermal distribution features of key components. Physical quantity features include environmental parameters such as temperature and humidity, gas concentration, smoke value, light intensity, noise decibels, and vibration frequency collected by the first environmental sensing device 230 and the second environmental sensing device 320, as well as the equipment's own operating parameters such as voltage or current of the power distribution equipment, air conditioner outlet temperature, or compressor pressure collected by the sensing and detection device 340. Historical correlation features include historical operating data of building equipment, its own initial operating parameter thresholds, and equipment maintenance knowledge data stored in the database 110. The output features are the anomaly identification results, including the anomaly judgment of the equipment's operating status and the specific description of the anomaly features. For example, the elevator car position sensor data fluctuates abnormally. The dynamic acquisition device 200 is used to acquire dynamic images and physical quantity data of building equipment in the inspection area and in buildings where fixed monitoring is inconvenient; Specifically, such as Figure 1 , Figure 4 and Figure 5 As shown, the dynamic data acquisition device 200 includes several mobile inspection platforms 210, such as the inspection vehicle device in the prior art. The inspection platform 210 is used to move inside the building according to the established inspection route to replace the inspection personnel in performing inspection operations. The inspection platform 210 is equipped with a communication module 211, a voice interaction module 212 and a positioning module 213. The voice interaction module 212 and the positioning module 213 are both connected to the background management device 100 through the communication module 211. Each inspection platform 210 is equipped with a first camera device 220, a first environmental sensing device 230, and a first edge computing device 240. The first camera device 220 is used to capture images and video data in real time during the inspection process; The first environmental sensing device 230 is used to sense various physical quantities during the inspection process. The first environmental sensing device 230 includes several temperature and humidity sensors, gas sensors, smoke sensors, light sensors and noise sensors. The first edge computing device 240 is used to preprocess the collected data and to receive and output image, video and physical quantity data. Specifically, it compensates for the coverage blind spots of the static acquisition device 300, and flexibly monitors mobile devices, hidden areas or devices that require regular inspection. The voice interaction module 212 enables remote collaboration between inspection personnel and the back-end. The first edge computing device 240 completes data preprocessing locally, reducing the amount of data transmitted to the back-end and reducing network load. The static data acquisition device 300 is used to acquire static images and physical quantity data of fixed-installation building equipment during operation. Specifically, such as Figure 1 and Figure 6 As shown, the static acquisition device 300 includes several second camera devices 310 fixedly installed at various detection points in the building. Each detection point is assigned a fixed number, so that the static acquisition data is uploaded with the number information, which facilitates the timely detection of abnormal locations. Second environmental sensing devices 320 and second edge computing devices 330 are set on the left and right sides of the second camera devices 310. The building equipment is equipped with corresponding sensing and detection devices 340. The second camera device 310 is used to collect images and video data at fixed monitoring points inside the building in real time. Both the first camera device 220 and the second camera device 310 adopt a dual-image acquisition mode combining a high-precision camera and an infrared thermal imager. The second environmental sensing device 320 is used to collect various physical quantities at fixed monitoring points in the building in real time. The second environmental sensing device 320 includes several temperature and humidity sensors, gas sensors, smoke sensors, light sensors, noise sensors and vibration sensors. The second edge computing device 330 is used to preprocess the collected data and receive and output image, video and physical quantity data. The second edge computing device 330 has the same structural features and functions as the first edge computing device 240. The sensing and detection device 340 is used to monitor the operating status of various equipment in the building in real time and collect the corresponding operating parameters of the equipment itself. The operating parameters include the voltage, current and power factor of the power distribution equipment, the air outlet temperature and compressor pressure of the air conditioner, the running speed, car position and door lock status of the elevator, and the pipeline pressure, flow rate and pump speed of the water supply and drainage equipment. It also includes the operating parameters of other types of equipment such as charging equipment and lighting equipment. Specifically, the system performs 24-hour uninterrupted monitoring of long-term fixed equipment. The core parameters of the equipment are collected by the sensing and detection device 340, and combined with the environmental data of the second environmental sensing device 320, the correlation between the equipment's operating status and the environment is analyzed. The infrared thermal imaging function of the second camera device 310 can capture abnormal heating inside the equipment, making up for the limitation of physical sensors that can only detect surface parameters. User terminal 400 is used to receive device operation status information, abnormal alarms and handling suggestions pushed by background management device 100, and send query instructions and operation requests to background management device 100. Background management device 100, dynamic acquisition device 200, static acquisition device 300 and user terminal 400 communicate with each other through networks such as Wi-Fi, 5G, etc. Specifically, such as Figure 1 As shown, the user terminal 400 can be a mobile phone, tablet, or computer device, and supports login via a visual interface on an APP, mini-program, or web page. The user group of the user terminal 400 includes administrators, inspection personnel, maintenance personnel, and building equipment users. Administrators can use the user terminal 400 to view the overall operating status of all building equipment in real time, receive anomaly summary reports, assign permissions, and issue management instructions. Inspection personnel can receive specific inspection task pushes, upload equipment data and on-site conditions collected during the inspection process, and obtain key inspection prompts from the background. Maintenance personnel can receive equipment anomaly alarms and corresponding handling suggestions, record the maintenance process and results, and query equipment historical maintenance records to assist in maintenance. Building equipment users can use the user terminal to query the usage status of relevant equipment, realizing information sharing and collaborative work.

[0026] Among them, such as Figure 1 - Figure 3 As shown, the background management device 100 includes: Database 110 is used to collect, store, and provide real-time data, historical data, initial operating parameter data, and equipment maintenance knowledge of building equipment. It divides the data in the corresponding information into several groups according to time, type, and parameter characteristics. For example, grouping by time, such as storing elevator speed frequently exceeding the ±5% threshold from 9:00 to 10:00 on August 8, 2025, stores the hourly operating speed and car position data of the elevator; grouping by equipment type, such as storing voltage and current data of power distribution equipment and compressor pressure data of air conditioners; and grouping by parameter characteristics, such as storing elevator door lock status and operating speed separately, supporting targeted queries. Analysis module 120 determines whether there are any abnormalities in the operating status of building equipment based on the collected data; Specifically, such as Figure 2 As shown, the analysis module 120 includes: The data preprocessing module 121 is used to preprocess the collected equipment operation data. The preprocessing operations include data cleaning, data conversion, data normalization and data compression operations. It removes blurry images caused by light interference and parameter jump values ​​caused by faults, compresses the dynamically collected video data frame by frame, and retains key frames. The feature extraction module 122 is used to extract key attributes of building equipment operation features and corresponding attribute values ​​from preprocessed equipment operation data. Examples include: extracting "highest temperature in the compressor area" and "temperature gradient change rate" from air conditioning infrared thermal imaging data; extracting "acceleration time" and "car horizontal vibration frequency" from elevator operation parameters; and extracting features such as "terminal color change" and "cable slack" from power distribution equipment images. The fusion module 123 is used to merge the building equipment operation features of key attributes to construct a comprehensive building equipment operation status feature set, avoiding misjudgment caused by a single feature. For example, it merges the "door lock status signal" and "door rail deformation image features" of the elevator to determine whether the door lock is abnormal due to door rail deformation, and merges the "current fluctuation features" and "surround gas concentration" of the power distribution equipment to determine whether there is partial discharge caused by poor contact. The comparison module 124 is used to compare the extracted and fused building equipment operation characteristics with the historical characteristic data of the building equipment during normal operation and the threshold data of its own standard operation parameters in the database 110 in turn to determine whether there are abnormal characteristics that exceed the normal range. For example, the current elevator "car vibration frequency 2.5Hz" is compared with the historical normal data of 1 to 2Hz and is determined to be abnormal. The threshold data of its own standard operation parameters can be adjusted according to the actual environmental requirements and different equipment parameters. The decision module 125 is used to determine the level of abnormality and generate corresponding processing suggestions based on the analysis results of the comparison module 124, combined with the historical processing records of abnormal features in the database 110 and equipment maintenance knowledge. For example, if the elevator "door rail deformation" is 1mm, historical data shows that ≤0.5mm is normal. Combined with the records in the database that "1~2mm deformation corresponds to 'door jamming' fault", it is determined to be "moderate abnormality" and it is recommended to "adjust the door rail parallelism". Output module 126 is used to convert the analysis results of decision module 125 into standard format suggestion data and store it in database 110. It also pushes the standard format suggestion data to display panel 150 and user terminal 400 of maintenance personnel. The standard format suggestion data for abnormal information is: {Equipment type: elevator; Location: Unit 1, Building 3; Abnormal feature: door rail deformation 1.2mm; Level: moderate; Suggestion: adjust door rail parallelism}. It is also stored in database 110 and pushed to the mobile phone of maintenance personnel. Normal data is marked as: {Equipment type: air conditioner; Status: normal; Characteristic value: compressor pressure Xbar}. It is grouped by time and stored in historical records. For serious abnormalities, in addition to text information, an audible and visual alarm is triggered. Processor 130 is used to receive and output information results from background management device 100; The memory 140 is used to receive coordination instructions from the processor 130 and to perform information control according to the instructions; Display panel 150 is used to display building equipment operation data to the administrator in a visual form; Among them, such as Figure 2 As shown, both the background management device 100 and the inspection platform 210 are equipped with audible and visual alarms on one side, and the audible and visual alarms communicate with the processor 130 and the communication module 211 through the network. Specifically, for example: when the static acquisition device 300 detects that the current of the power distribution equipment on a certain floor continuously exceeds the threshold, and the dynamic acquisition device 200 captures an image of the equipment's wiring terminals being overheated and red, the abnormal detection model of the background management device 100 focuses on the "current parameter channel" and the "wiring terminal area" through a hybrid attention mechanism, determines it as a "serious abnormality," and immediately triggers the following operations: the instruction processor 130 activates the audible and visual alarm, retrieves the historical short-circuit fault records of the equipment through the database 110, generates a "power outage + terminal replacement" handling suggestion, simultaneously pushes the alarm to the user terminal 400 of the administrator and maintenance personnel, and marks the fault location and real-time data curve on the display panel 150.

[0027] A method for monitoring the operational status of building equipment, the specific steps of which are as follows: Step S1: Collect images and physical quantity data inside the building through the dynamic acquisition device 200 and the static acquisition device 300 respectively, and transmit the collected building equipment operation data to the background management device 100. Specifically, the inspection platform 210 of the dynamic acquisition device 200 moves along a preset path, and the first camera device 220 acquires dynamic images and infrared thermal imaging data of the equipment. The first environmental sensing device 230 simultaneously acquires environmental physical quantities such as temperature, humidity, and gas concentration. The first edge computing device 240 performs preliminary cleaning of the data and uploads it to the background management device 100 through the communication module 211. The second camera device 310 of the static acquisition device 300 captures static images of fixed detection points. The second environmental sensing device 320 acquires physical quantities such as vibration frequency. The sensing and detection device 340 records equipment parameters such as power distribution equipment voltage and elevator running speed. The second edge computing device 330 performs preliminary cleaning of the data and uploads it. Both types of acquisition devices carry timestamps and location information when uploading data. Dynamic acquisition includes data from the positioning module 213, and static acquisition includes fixed detection point numbers, ensuring that the data is accurately associated with the equipment location and time, which facilitates timely detection of abnormal locations. Step S2: After receiving the data, the background management device 100 groups and stores the data according to time, type and parameter characteristics. The analysis module 120 extracts the pre-processed building equipment operation characteristics for comparative analysis to determine whether the equipment operation status is abnormal. Specifically, the content in database 110 can also be grouped and stored according to dynamic data or static data, image data or physical quantity data, real-time data or historical data, and associated with tags such as device number and acquisition time. The data preprocessing module 121 of the analysis module 120 performs secondary processing on the uploaded data, including image denoising, physical quantity data compression, and elimination of residual noise not processed by edge computing. The feature extraction module 122 extracts key features from the preprocessed data. Examples of key features are: high temperature area of ​​infrared image and fluctuation frequency of current data. The fusion module 123 merges the multi-dimensional features of the same device into comprehensive features to construct a comprehensive building equipment operating status feature set. An example is the vibration data of elevator + door track image features. Step S3: If abnormal features exist, the abnormal features and abnormal attribute values ​​are marked. The background management device 100 generates an abnormal alarm and handling suggestions and pushes them to the user terminal 400 of maintenance personnel and administrators. If no abnormal features exist, the data is marked as normal and stored in the database 110 according to the grouping rules. Specifically, the multimodal data processing layer of the anomaly detection model processes the feature set of the operating status of the integrated building equipment, the hybrid attention mechanism layer focuses on key features and strengthens the anomaly correlation features, the comparison module 124 compares the processed features with the historical feature data of the building equipment during normal operation and its own standard operating parameter data threshold in the database 110, calculates the deviation, if the deviation is ≤ the threshold, it is marked as "normal" and stored according to the grouping rules, if the deviation is > the threshold, it is judged as an anomaly, the decision module 125 determines the anomaly level by combining the historical processing records, generates maintenance suggestions, and the output module 126 converts the anomaly information, level and suggestions into a standard format and pushes them synchronously to the display panel 150 and the user terminal 400, triggering the sound and light alarm of the background management device 100 and the inspection platform 210; Specifically, maintenance personnel receive abnormal information through the user terminal 400, navigate to the abnormal location using the data location information, communicate with the backend in real time through the user terminal 400, and record the processing process through the first camera device 220. After the processing is completed, the maintenance personnel submit the results on the user terminal 400, which include the cause of the abnormality and the handling measures, and archive them to the "historical maintenance records" group.

[0028] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A building equipment operation status monitoring system, characterized in that, include: The background management device (100) is used to manage, process and analyze building equipment, monitor the data information of multiple acquisition devices in real time, and use an attention-based anomaly detection model to realize the identification and maintenance of building equipment anomalies. The dynamic acquisition device (200) is used to acquire dynamic images and physical quantity data of building equipment in the inspection area and inconvenient fixed monitoring areas; A static data acquisition device (300) is used to acquire static images and physical quantity data of fixedly installed building equipment during operation; The user terminal (400) is used to receive equipment operation status information, abnormal alarms and handling suggestions pushed by the background management device (100), and to send query instructions and operation requests to the background management device (100); The background management device (100), dynamic acquisition device (200), static acquisition device (300) and user terminal (400) communicate with each other via a network; The background management device (100) includes: The database (110) is used to collect, store and provide real-time data, historical data, initial operating parameter data and equipment maintenance knowledge of building equipment, and to divide the data in the corresponding information into several groups according to time, type and parameter characteristics; The analysis module (120) determines whether there are any abnormalities in the operating status of building equipment based on the collected data; The processor (130) is used to receive and output information results from the background management device (100); The memory (140) is used to receive coordination instructions from the processor (130) and to perform information control according to the instructions; Display panel (150) is used to present building equipment operation data to the administrator in a visual format.

2. The building equipment operation status monitoring system according to claim 1, characterized in that, The dynamic acquisition device (200) includes several movable inspection platforms (210). The inspection platform (210) is used to move inside the building according to a specified inspection path to replace the inspection personnel in performing inspection operations. The inspection platform (210) is equipped with a communication module (211), a voice interaction module (212), and a positioning module (213). The voice interaction module (212) and the positioning module (213) are both connected to the background management device (100) through the communication module (211). Each of the inspection platforms (210) is equipped with a first camera device (220), a first environmental sensing device (230), and a first edge computing device (240). The first camera device (220) is used to capture images and video data in real time during the inspection process; The first environmental sensing device (230) is used to sense various physical quantities during the inspection process. The first environmental sensing device (230) includes several temperature and humidity sensors, gas sensors, smoke sensors, light sensors and noise sensors. The first edge computing device (240) is used to preprocess the collected data and to receive and output image, video and physical quantity data.

3. The building equipment operation status monitoring system according to claim 2, characterized in that, The static acquisition device (300) includes several second camera devices (310) fixedly installed at various monitoring points in the building. The second camera devices (310) are equipped with second environmental sensing devices (320) and second edge computing devices (330) on both the left and right sides. The building equipment is equipped with corresponding sensing and detection devices (340). The second camera device (310) is used to collect images and video data from fixed monitoring points within the building in real time; The second environmental sensing device (320) is used to collect various physical quantities at fixed monitoring points in the building in real time. The second environmental sensing device (320) includes several temperature and humidity sensors, gas sensors, smoke sensors, light sensors, noise sensors and vibration sensors. The second edge computing device (330) is used to preprocess the collected data and receive and output image, video data and physical quantity data. The second edge computing device (330) has the same structural features as the first edge computing device (240). The sensing and detection device (340) is used to monitor the operating status of various equipment in the building in real time and collect the operating parameters of the corresponding equipment. The operating parameters include the voltage, current and power factor of the power distribution equipment, the air outlet temperature and compressor pressure of the air conditioner, the running speed, car position and door lock status of the elevator, and the pipeline pressure, flow rate and pump speed of the water supply and drainage equipment.

4. The building equipment operation status monitoring system according to claim 1, characterized in that, The analysis module (120) includes: The data preprocessing module (121) is used to perform preprocessing operations on the collected equipment operation data. The preprocessing operations include data cleaning, data conversion, data normalization and data compression. The feature extraction module (122) is used to extract key attributes of building equipment operation features and corresponding attribute values ​​from the preprocessed equipment operation data; The fusion module (123) is used to merge the building equipment operation characteristics of key attributes to construct a comprehensive building equipment operation status feature set; The comparison module (124) is used to compare the extracted and fused building equipment operation features with the historical feature data of the building equipment during normal operation and the threshold data of its own standard operation parameters in the database (110) in turn to determine whether there are any abnormal features that exceed the normal range. The decision module (125) is used to determine the anomaly level and generate corresponding processing suggestions based on the analysis results of the comparison module (124), combined with the historical processing records of anomaly features in the database (110) and equipment maintenance knowledge; The output module (126) is used to convert the analysis results of the decision module (125) into standard format suggestion data and store it in the database (110), and synchronously push the standard format suggestion data to the display panel (150) and the user terminal (400) of the maintenance personnel.

5. The building equipment operation status monitoring system according to claim 1, characterized in that, The user terminal (400) is a mobile phone, tablet or computer device, and supports login operations via a visual interface of APP, mini program or web page. The user group of the user terminal (400) includes administrators, inspection personnel, maintenance personnel and building equipment users.

6. The building equipment operation status monitoring system according to claim 2, characterized in that, Both the background management device (100) and the inspection platform (210) are equipped with audible and visual alarms on one side, and the audible and visual alarms communicate with the processor (130) and the communication module (211) through the network.

7. The building equipment operation status monitoring system according to claim 3, characterized in that, Both the first camera device (220) and the second camera device (310) adopt a dual-image acquisition mode combining a high-precision camera and an infrared thermal imager.

8. The building equipment operation status monitoring system according to claim 1, characterized in that, The attention mechanism described employs a hybrid attention mechanism that combines channel attention and spatial attention.

9. The building equipment operation status monitoring system according to claim 8, characterized in that, The anomaly detection model is based on a deep learning framework and includes a multimodal data processing layer, a hybrid attention mechanism layer, and an anomaly decision layer. The input features of the anomaly detection model are multimodal data, and the output features are anomaly identification results.

10. A method for monitoring the operating status of building equipment according to any one of claims 1-9, characterized in that, The specific usage steps are as follows: Step S1: Collect images and physical quantity data inside the building through the dynamic acquisition device (200) and the static acquisition device (300), and transmit the collected building equipment operation data to the background management device (100). Step S2: After the background management device (100) receives the data, the database (110) groups and stores the data according to time, type and parameter characteristics. The analysis module (120) extracts the pre-processed building equipment operation characteristics for comparative analysis to determine whether the equipment operation status is abnormal. Step S3: If there are abnormal features, mark the abnormal features and abnormal attribute values. The background management device (100) generates an abnormal alarm and handling suggestions and pushes them to the user terminal (400) of maintenance personnel and administrators. If there are no abnormal features, mark the data as normal and store it in the database (110) according to the grouping rules.

Images