Electromechanical comprehensive diagnosis system based on BAS visualization

Abstract

The invention provides an electromechanical comprehensive diagnosis system based on BAS visualization, which is composed of five main parts, namely an equipment data acquisition module, an equipment image acquisition module, a data processing module, an equipment control module and a visualization background, and each module is connected with the data processing module. Wherein the equipment data acquisition module further comprises a PLC data acquisition module, a working condition data acquisition module and an environment data acquisition module which jointly provide comprehensive equipment operation parameters for the data processing module. According to the system, by integrating multi-source data and image information, comprehensive monitoring and intelligent analysis of the running state of the equipment are achieved, and abnormity of the electromechanical equipment can be found in time. And the visual background provides a visual and real-time data interface for management personnel, so that rapid decision making is facilitated. The application of the system significantly improves the equipment maintenance efficiency, reduces the maintenance cost, and provides an effective solution for the intelligent management of electromechanical equipment.

Description

Technical Field

[0001] The present application relates to the field of automatic control, and in particular to a mechanical and electrical integrated diagnosis system based on BAS visualization. Background Art

[0002] The main role of BAS visualization is to improve operational efficiency and equipment safety. It allows managers to quickly identify system anomalies, optimize energy use, extend equipment life, and improve user experience. Through the visual interface, even non-professional technicians can understand complex system behaviors and achieve more efficient building management. In addition, this technology also supports historical data analysis and trend prediction, providing a basis for preventive maintenance.

[0003] BAS visualization technology has significant advantages in the field of electromechanical equipment diagnosis. It provides an intuitive display of equipment status and enables rapid abnormality identification; supports multi-system integration and data correlation analysis; lowers the technical threshold, allowing more people to participate in equipment management; reduces downtime and maintenance costs through predictive maintenance; and supports remote monitoring, improving work efficiency.

[0004] However, BAS visualization technology also has obvious shortcomings. First, data reliability issues are prominent, and sensor failure or improper calibration can lead to misdiagnosis. Second, the complex interactions between devices are difficult to fully present in the visualization interface, which may lead to one-sided diagnosis. In addition, the system usually relies on preset thresholds for abnormal judgment, which makes it difficult to adapt to performance changes caused by equipment aging. The visualization system itself also faces the problem of data overload, and too much information will interfere with decision-making. Lack of technical standardization also restricts the interoperability between equipment from different manufacturers.

[0005] It is particularly noteworthy that BAS visualization has limitations in complex fault diagnosis. When multiple subsystems have interactive faults, pure visualization often fails to present the root cause and propagation path of the fault. In addition, visualization systems rely heavily on the experience of professionals to interpret data and lack in-depth automatic analysis capabilities, which makes it difficult to maintain diagnostic quality in an environment with frequent personnel flow. Moreover, most existing systems focus on anomaly detection rather than root cause analysis, and are unable to provide targeted maintenance recommendations, which increases the complexity of subsequent processing. Summary of the invention

[0006] The present application provides a BAS-based integrated electromechanical diagnostic system, including: Equipment data acquisition module; Equipment image acquisition module; Data processing module; Equipment control module; Visual background; Wherein, the device data acquisition module, the device image acquisition module, the device control module and the visual background data are connected to the data processing module; Among them, the equipment data acquisition module includes a PLC data acquisition module, an operating condition data acquisition module and an environmental data acquisition module, and the PLC data acquisition module, the operating condition data acquisition module and the environmental data acquisition module are data-connected to the data processing module.

[0007] By adopting the above technical solution, the BAS visualization-based electromechanical integrated diagnostic system can realize comprehensive monitoring of the equipment operation status by combining PLC data acquisition, working condition data and environmental data acquisition modules; at the same time, combined with the image acquisition module, it provides an intuitive visualization of the equipment status, and integrates various types of data and image data for intelligent analysis, so that equipment abnormalities can be discovered in time; the visualization background provides managers with an intuitive and timely data interface, which facilitates quick decision-making; improves equipment maintenance efficiency and reduces maintenance costs.

[0008] Optionally, the BAS visualization-based electromechanical integrated diagnosis system further includes a visualization diagnosis strategy, including the following steps: A1, obtaining preset background data of the operating status of each electromechanical device and target operating status data in the visualization background; A2, acquiring the operating status image data of each electromechanical device through the device image acquisition module; A3, determining the corresponding device state consistency according to the operating state background data of each electromechanical device, the target operating state data and the corresponding operating state image data using a preset device state consistency algorithm; A4, if the device status consistency of the electromechanical device is false, the corresponding electromechanical device is defined as an abnormal operation device; A5, obtaining device PLC data, device operating condition data and device environment data of the abnormally operating device through the device data acquisition module; A6, generating comprehensive data on abnormal equipment status based on equipment PLC data, equipment operating condition data, equipment environment data and corresponding equipment type information; A7, based on the comprehensive data of abnormal equipment status, the corresponding estimated fault cause is determined through the pre-trained electromechanical fault identification model.

[0009] By adopting the above technical solution, the BAS visualization-based electromechanical integrated diagnosis system can combine the display data of the visualization background with the real-time image data by adopting a multi-angle data cross-validation mechanism, and compare them through the equipment status consistency algorithm; when abnormal equipment is found, PLC data, operating condition data and environmental data are collected for the abnormal equipment, and then the pre-trained electromechanical fault recognition model is used to perform fault diagnosis based on various data, which not only improves the accuracy and timeliness of the diagnosis, but also reduces the false alarm rate, reduces the workload of maintenance personnel, and improves the reliability of the overall system.

[0010] Optionally, step A3 includes the following steps: A301, generating corresponding running status vector data according to the running status background data; A302, generating corresponding target state vector data according to the target operation state data; A303, calculating the corresponding device state Euclidean distance according to the running state vector data and the target state vector data; A304, calculating a corresponding modulus according to the target state vector data and defining it as the target state vector modulus; A305, calculates the corresponding running state deviation according to the Euclidean distance of the equipment state and the quotient of the target state vector; A306, if the running state deviation is greater than the preset equipment running deviation threshold, the equipment state consistency is defined as false; A307, if the running state deviation is less than or equal to the equipment running deviation threshold, then generate corresponding equipment start / stop state recognition data according to the running state image data through the pre-trained equipment state recognition model; A308, determining corresponding target start / stop state data according to the target operation state data; A309, if the device start / stop state identification data is not equal to the target start / stop state data, the device state consistency is defined as false; A310, if the device start / stop state identification data is equal to the target start / stop state data, the device state consistency is defined as true.

[0011] By adopting the above technical scheme, the electromechanical integrated diagnosis system based on BAS visualization can vectorize the equipment operation status data and target status data, and use the Euclidean distance method to calculate the deviation to make the anomaly detection more accurate and quantitative, and combine the dual verification mechanism of numerical threshold judgment and image recognition. It not only considers the degree of deviation of the background data, but also uses the pre-trained image model to intelligently identify the actual start and stop status of the equipment; improves the system's sensitivity and accuracy to equipment anomalies, effectively avoids the misjudgment that may be caused by a single data source, and provides a more reliable trigger basis for subsequent fault diagnosis, thereby realizing comprehensive and accurate monitoring of the operation status of electromechanical equipment.

[0012] Optionally, step A4 includes the following steps: A401, if the device status consistency of the electromechanical device is false, the corresponding background start / stop status data is determined according to the running status background data; A402: If the background start / stop status data is equal to the target start / stop status data, and the device start / stop status identification data is not equal to the target start / stop status data, the corresponding electromechanical device is defined as an abnormal operation device; A403: If the background start / stop status data is not equal to the target start / stop status data, and the device start / stop status identification data is equal to the target start / stop status data, the corresponding electromechanical device is defined as a data link abnormal device.

[0013] By adopting the above technical solution, the BAS visualization-based electromechanical integrated diagnosis system can cross-validate by combining background data with image recognition results. When the background data appears normal but the image recognition result is abnormal, the actual operation abnormality of the equipment can be quickly identified; conversely, when the background data is abnormal but the image recognition result appears normal, the system can intelligently determine that it is a data link problem; this not only improves the accuracy of fault location, but also shortens the troubleshooting time, avoids unnecessary on-site inspections by maintenance personnel, and enables a more reasonable allocation of maintenance resources, thereby significantly improving the operation and maintenance efficiency and reliability of the entire system.

[0014] Optionally, step A7 includes the following steps: A701, determining the corresponding device type value in a preset device value code comparison table according to the device type information; A702, generating corresponding PLC feature data, working condition feature data and environmental feature data through a preset feature extraction algorithm according to the equipment PLC data, equipment working condition data and equipment environment data; A703, generating abnormal equipment dimension reduction feature data according to the PLC feature data, the working condition feature data and the environmental feature data by using a preset feature dimension reduction algorithm; A704, generating abnormal operation equipment feature data according to the combination of equipment type value and abnormal equipment dimension reduction feature data; A705, based on the characteristic data of the abnormally operating equipment, the corresponding estimated fault cause is determined through the electromechanical fault identification model.

[0015] By adopting the above technical scheme, the BAS visualization-based electromechanical integrated diagnosis system can standardize the equipment type through the equipment numerical code comparison table to ensure that the fault judgment of different types of equipment is targeted; then, the key feature information is extracted from the PLC data, working condition data and environmental data through the feature extraction algorithm, and then the data dimension is efficiently compressed through the feature dimension reduction algorithm, retaining the core features while improving the calculation efficiency; finally, the equipment type value and the dimension reduction feature data are intelligently integrated, and the cause of the fault is accurately inferred through the pre-trained electromechanical fault identification model, which not only improves the accuracy of fault diagnosis, but also realizes personalized diagnosis for different equipment types, shortens the fault response time, and reduces equipment maintenance costs.

[0016] Optionally, the electromechanical integrated diagnosis system based on BAS visualization further includes the following steps for training an electromechanical fault identification model: B1, obtaining corresponding historical operating fault electromechanical equipment data in a preset historical electromechanical database according to preset historical electromechanical equipment fault records; B2, determine the equipment type value of each electromechanical equipment with historical operating failures in the equipment value code comparison table; B3, determining the corresponding equipment fault type value in a preset fault value code comparison table according to the fault cause information of each historical running fault electromechanical equipment in the historical electromechanical equipment fault record; B4, according to each historical running faulty electromechanical equipment, obtain the corresponding faulty equipment PLC data, faulty equipment working condition data and faulty equipment environment data from the historical running faulty electromechanical equipment data; B5, generating corresponding fault PLC feature data, fault condition feature data and fault environment feature data through feature extraction algorithm according to fault device PLC data, fault device working condition data and fault device environment data; B6, generating fault equipment dimension reduction feature data according to the fault PLC feature data, the fault working condition feature data and the fault environment feature data by using a feature dimension reduction algorithm; B7, generating faulty equipment feature data according to the equipment type value corresponding to the historical faulty electromechanical equipment and the faulty equipment dimension reduction feature data combination; B8, generating a fault identification training set based on the fault equipment characteristic data of all historically faulty electromechanical equipment and the corresponding equipment fault type numerical combination; B9, train the preset recognition model through the fault recognition training set to generate an electromechanical fault recognition model.

[0017] By adopting the above technical scheme, the BAS visualization-based electromechanical integrated diagnosis system can realize structured data processing through standardized equipment type and fault type numerical coding; the system uses feature extraction and dimensionality reduction algorithms consistent with diagnosis to process historical PLC data, operating condition data and environmental data, ensuring the consistency of model training and application scenarios; and combines the equipment type value with the dimensionality reduction feature data, so that the model can identify faults of different types of equipment, improves the versatility of the identification model, and provides technical support for equipment predictive maintenance and processing decisions.

[0018] Optionally, the BAS-visualized electromechanical integrated diagnosis system further comprises the following steps: C1, obtain the device identification information corresponding to the data link abnormal device; C2, determining the corresponding link electromechanical equipment list according to the equipment identification information and the preset equipment network topology data; C3, according to each electromechanical device in the link electromechanical device list, obtain the corresponding device PLC data, device operating condition data, device environment data and corresponding device type information and combine them to generate comprehensive data of link device status; C4, based on the comprehensive data of the status of each link device, try to identify and generate the corresponding estimated fault cause through the electromechanical fault identification model.

[0019] By adopting the above technical solution, the BAS-visualized electromechanical integrated diagnostic system can track the related equipment on the entire data transmission link of the data link abnormal device through the combination of equipment identification information and network topology data, and perform fault identification and diagnosis on each device on the link, thereby improving the efficiency of link fault troubleshooting. It not only speeds up the fault location speed in the link, but also can discover the mutual influence and reaction between equipment, providing a scientific basis for system optimization.

[0020] Optionally, the BAS-visualized electromechanical integrated diagnosis system further comprises the following steps: D1, obtaining the historical operating status image data corresponding to each electromechanical device from the historical electromechanical database according to the preset inspection cycle; D2, using the preset pre-order sampling time window and post-order sampling time window to sequentially acquire the adjacent pre-order time window historical state image data and post-order time window historical state image data from the historical operation state image data of each electromechanical device; D3, generating corresponding equipment previous start / stop state recognition data through equipment state recognition model recognition according to previous time window historical state image data; D4, generating corresponding equipment subsequent start / stop state recognition data through equipment state recognition model recognition according to the subsequent time window historical state image data; D5, if the previous start / stop state identification data of the device is inconsistent with the subsequent start / stop state identification data of the device, the corresponding device state conversion time is determined according to the previous sampling time window and the subsequent sampling time window; D6, according to the equipment state conversion time, obtain the corresponding equipment PLC data, equipment condition data, equipment environment data in the historical electromechanical database and combine them with the corresponding equipment type information to generate the state conversion equipment state comprehensive data; D7, based on the comprehensive data of the state transition equipment state, attempts to identify and generate the corresponding estimated fault cause through the electromechanical fault identification model.

[0021] By adopting the above technical solution, the BAS visualization-based electromechanical integrated diagnosis system can capture the equipment state transition points by regularly checking historical image data and adopting the pre-order and post-order time window analysis method; after determining the state transition moment, the system immediately conducts in-depth analysis of the PLC data, operating condition data and environmental data at that time point to construct a complete state transition comprehensive data set; and then analyzes it through the electromechanical fault identification model to achieve early detection of potential faults in the state transition process; this method of actively capturing state changes and predicting faults not only improves the system's sensitivity to fault signs, but also provides early warnings before the fault fully manifests, providing sufficient response time for preventive maintenance and significantly reducing the risk of equipment damage and maintenance costs.

[0022] In summary, the present application includes at least one of the following beneficial technical effects: 1. Comprehensive monitoring of the equipment's operating status can be achieved by combining PLC data acquisition, working condition data and environmental data acquisition modules; at the same time, combined with the image acquisition module, it provides an intuitive visual display of the equipment status, and integrates various types of data and image data for intelligent analysis, so that equipment abnormalities can be discovered in time; the visual background provides managers with an intuitive and timely data interface, which facilitates quick decision-making; improves equipment maintenance efficiency and reduces maintenance costs.

[0023] 2. By adopting a multi-angle data cross-validation mechanism, the display data of the visualization background can be combined with the real-time image data, and compared through the equipment status consistency algorithm; when abnormal equipment is found, PLC data, operating data and environmental data are collected for the abnormal equipment, and then the pre-trained electromechanical fault identification model is used to diagnose the fault according to various data. This not only improves the accuracy and timeliness of the diagnosis, but also reduces the false alarm rate, reduces the workload of maintenance personnel, and improves the reliability of the overall system.

[0024] 3. The equipment operation status data and target status data can be vectorized, and the Euclidean distance method can be used to calculate the deviation to make anomaly detection more accurate and quantitative. In combination with the dual verification mechanism of numerical threshold judgment and image recognition, not only the degree of deviation of the background data is considered, but also the actual start and stop status of the equipment is intelligently identified through the pre-trained image model. This improves the system's sensitivity and accuracy to equipment anomalies, effectively avoids the misjudgment that may be caused by a single data source, and provides a more reliable trigger basis for subsequent fault diagnosis, thereby achieving comprehensive and accurate monitoring of the operating status of electromechanical equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a schematic diagram of the principle of a BAS-visualized electromechanical integrated diagnosis system of the present invention.

[0026] Figure 2 It is a process schematic diagram of a BAS-based visualized electromechanical integrated diagnosis system of the present invention. DETAILED DESCRIPTION

[0027] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0028] The embodiments of the present application are further described in detail below in conjunction with the drawings in the specification.

[0029] refer to Figure 1 and Figure 2 The present invention provides a BAS-based integrated electromechanical diagnosis system for automated fault diagnosis of electromechanical equipment under the BAS system, including: Equipment data acquisition module 10; Device image acquisition module 20; Data processing module 30; Equipment control module 40; Visual background 50; Wherein, the device data acquisition module 10, the device image acquisition module 20, the device control module 40 and the visualization background 50 are data-connected to the data processing module; The equipment data acquisition module 10 includes a PLC data acquisition module 11 , an operating condition data acquisition module 12 and an environmental data acquisition module 13 , and the PLC data acquisition module 11 , the operating condition data acquisition module 12 and the environmental data acquisition module 13 are data-connected to the data processing module 30 .

[0030] The equipment data acquisition module 10 is mainly used to collect various working parameters of the electromechanical equipment when it is working.

[0031] The PLC data acquisition module 11 is mainly used to collect device parameters on the PLC of the electromechanical equipment. For example, it can connect and read the data on the PLC of the electromechanical equipment through the general MODBUS-TCP protocol to analyze whether there is an abnormality or failure in the electromechanical equipment; The working condition data acquisition module 12 is mainly used to collect data related to working conditions when the electromechanical equipment is working, such as the temperature, current, voltage, operating power, etc. of each main component, which can be used to analyze abnormal conditions of components of the electromechanical equipment; The environmental data acquisition module 13 is mainly used to collect the surrounding external environmental data of the electromechanical equipment when it is working, such as external temperature, humidity, vibration, noise, magnetic field, etc., which can be used to analyze the interference or influence of external factors of the electromechanical equipment on the operation of the equipment.

[0032] The device image acquisition module 20 is mainly used to collect image data of electromechanical equipment. On the one hand, it is used for staff to monitor the electromechanical equipment so that they can respond quickly when an abnormality occurs. On the other hand, it can identify and analyze the working status of the electromechanical equipment in the image through recognition models or algorithms, and combine the data collected by the device data acquisition module to make abnormal judgments and fault predictions on the electromechanical equipment.

[0033] The data processing module 30 is mainly used to collect various collected data for unified processing; The device control module 40 is mainly used to control various electromechanical devices, so that when an electromechanical device with abnormal operation is determined, the abnormal electromechanical device can be remotely shut down or paused to reduce the impact of the abnormal device.

[0034] The visualization background 50 is mainly used to visualize various working parameter data of electromechanical equipment to facilitate viewing and supervision by staff.

[0035] Through the above technical scheme, the BAS visualization-based electromechanical integrated diagnosis system can realize comprehensive monitoring of the equipment operation status by combining PLC data acquisition, working condition data and environmental data acquisition modules; at the same time, combined with the image acquisition module, it provides an intuitive visualization of the equipment status, and integrates various types of data and image data for intelligent analysis, so that equipment abnormalities can be discovered in time; the visualization background provides managers with an intuitive and timely data interface, which facilitates quick decision-making; improves equipment maintenance efficiency and reduces maintenance costs.

[0036] Furthermore, the BAS visualization-based electromechanical integrated diagnosis system further includes a visualization diagnosis strategy, including the following steps: A1, obtaining preset background data of the operating status of each electromechanical device and target operating status data in the visualization background 50; The operating status background data is data on the operating status of the electromechanical equipment displayed on the visualization background 50, for example, data on the rotation speed of the fan equipment displayed on the visualization background 50, data on the brightness of the lighting equipment displayed on the visualization background 50, etc.; The target operating state data is the pre-set operating state data required for the electromechanical equipment to reach, for example, the speed that the fan equipment is currently required to reach, the brightness that the lighting equipment is required to reach, and other data.

[0037] A2, acquiring the operating status image data of each electromechanical device through the device image acquisition module 20; The operating status image data is the image or video content data of the on-site working status of each electromechanical device during actual operation, captured by the device image acquisition module 20 .

[0038] A3, determining the corresponding device state consistency according to the operating state background data of each electromechanical device, the target operating state data and the corresponding operating state image data using a preset device state consistency algorithm; The device status consistency algorithm is a pre-set algorithm used to determine whether the working status of the electromechanical equipment reflected by the operating status background data, target operating status data and operating status image data of the electromechanical equipment is consistent.

[0039] A4, if the device status consistency of the electromechanical device is false, the corresponding electromechanical device is defined as an abnormal operation device; Abnormally operating equipment is electromechanical equipment whose equipment status consistency is false, that is, electromechanical equipment that may have abnormal operating status and needs further analysis and judgment.

[0040] A5, obtaining the device PLC data, device operating condition data and device environment data of the abnormally operating device through the device data acquisition module 10; The equipment PLC data is the equipment data read from the PLC of the electromechanical equipment by the PLC data acquisition module 11; The equipment operating condition data is the internal state data of the electromechanical equipment collected from the electromechanical equipment by the operating condition data collection module 12; The equipment environment data is the working environment data collected outside the electromechanical equipment by the environment data collection module 13 .

[0041] A6, generating comprehensive data on abnormal equipment status based on equipment PLC data, equipment operating condition data, equipment environment data and corresponding equipment type information; The equipment type information is the pre-entered equipment model information corresponding to the electromechanical equipment, which is used to identify and determine the type, type, model and other information corresponding to the electromechanical equipment, so that subsequent fault identification can be correctly analyzed based on the model category; The comprehensive data on the status of equipment with abnormal operation is a collection of data collected from various aspects of electromechanical equipment.

[0042] A7, based on the comprehensive data of abnormal equipment status, the corresponding estimated fault cause is determined through the pre-trained electromechanical fault identification model.

[0043] The electromechanical fault identification model is a fault identification model generated by pre-training with training sample data, and is used to identify possible faults of electromechanical equipment based on various operating data of the electromechanical equipment; The estimated fault cause is the possible fault cause generated by the electromechanical fault identification model.

[0044] Through the above steps, the BAS visualization-based electromechanical integrated diagnosis system can combine the display data of the visualization background with the real-time image data by adopting a multi-angle data cross-validation mechanism, and compare them through the equipment status consistency algorithm; when abnormal equipment is found, PLC data, operating condition data and environmental data are collected for the abnormal equipment, and then the pre-trained electromechanical fault recognition model is used to perform fault diagnosis based on various data, which not only improves the accuracy and timeliness of diagnosis, but also reduces the false alarm rate, reduces the workload of maintenance personnel, and improves the reliability of the overall system.

[0045] Furthermore, the step A3 comprises the following steps: A301, generating corresponding running status vector data according to the running status background data; The running status vector data is a vector generated by the running status background data.

[0046] A302, generating corresponding target state vector data according to the target operation state data; The target state vector data is a vector generated from the target operating state data.

[0047] A303, calculating the corresponding device state Euclidean distance according to the running state vector data and the target state vector data; The device state Euclidean distance is the Euclidean distance between the running state vector data and the target state vector data.

[0048] A304, calculating a corresponding modulus according to the target state vector data and defining it as the target state vector modulus; The target state vector modulus is the modulus of the target state vector data.

[0049] A305, calculates the corresponding running state deviation according to the Euclidean distance of the equipment state and the quotient of the target state vector; The operating state deviation is the quotient of the Euclidean distance of the equipment state and the modulus of the target state vector.

[0050] A306, if the running state deviation is greater than the preset equipment running deviation threshold, the equipment state consistency is defined as false; The equipment operation deviation threshold is a pre-set reference value used to determine whether the operation state deviation of the electromechanical equipment is too large; The value of the device state consistency can be represented by a Boolean value, that is, 0 or 1, true or false.

[0051] A307, if the running state deviation is less than or equal to the equipment running deviation threshold, then generate corresponding equipment start / stop state recognition data according to the running state image data through the pre-trained equipment state recognition model; The equipment state recognition model is a pre-trained model used to recognize the operating state of the electromechanical equipment based on the image or video in the operating state image data; The equipment start / stop state identification data is the value output by the equipment state identification model, which can be a Boolean value and is used to determine the start / stop state of the electromechanical equipment.

[0052] A308, determining corresponding target start / stop state data according to the target operation state data; The target start / stop state data is part of the target operation state data, which is used to reflect the start or stop state that the device needs to reach, and can be a Boolean value; A309, if the device start / stop state identification data is not equal to the target start / stop state data, the device state consistency is defined as false; The equipment start / stop status identification data is not equal to the target start / stop status data, that is, the current start / stop status of the electromechanical equipment does not match the expected setting.

[0053] A310, if the device start / stop state identification data is equal to the target start / stop state data, then the device state consistency is defined as true; The equipment start / stop state identification data is equal to the target start / stop state data, that is, the current start / stop state of the electromechanical equipment is consistent with the expected setting; For example, for a certain fan, the start / stop status displayed on the visualization background 50 is started, but the data obtained by actual measurement shows that it is stopped, then the device status consistency can be determined to be false; if the start / stop status of the fan displayed on the visualization background 50 is started, and the data obtained by actual measurement shows that it is started, then the device status consistency can be determined to be true.

[0054] Through the above steps, the BAS visualization-based electromechanical integrated diagnosis system can vectorize the equipment operation status data and target status data, use the Euclidean distance method to calculate the deviation to make the anomaly detection more accurate and quantitative, and combine the dual verification mechanism of numerical threshold judgment and image recognition. It not only considers the degree of deviation of the background data, but also uses the pre-trained image model to intelligently identify the actual start and stop status of the equipment; improves the system's sensitivity and accuracy to equipment anomalies, effectively avoids the misjudgment that may be caused by a single data source, and provides a more reliable trigger basis for subsequent fault diagnosis, thereby realizing comprehensive and accurate monitoring of the operation status of electromechanical equipment.

[0055] Furthermore, the step A4 comprises the following steps: A401, if the device status consistency of the electromechanical device is false, the corresponding background start / stop status data is determined according to the running status background data; The background start / stop status data is part of the running status background data, which is used to reflect the start or stop status of the device displayed in the visualization background 50 and can be a Boolean value.

[0056] A402: If the background start / stop status data is equal to the target start / stop status data, and the device start / stop status identification data is not equal to the target start / stop status data, the corresponding electromechanical device is defined as an abnormal operation device; When the background start-stop status data is equal to the target start-stop status data, it means that the data between the electromechanical equipment and the visualization background 50 are consistent, indicating that the data path between them should be normal. If the equipment start-stop status identification data is not equal to the target start-stop status data, it means that the electromechanical equipment cannot reach the expected set start-stop status, which means that the electromechanical equipment may have abnormal operation.

[0057] A403: If the background start / stop status data is not equal to the target start / stop status data, and the device start / stop status identification data is equal to the target start / stop status data, the corresponding electromechanical device is defined as a data link abnormal device. When the background start-stop status data is not equal to the target start-stop status data, it means that the data between the electromechanical device and the visualization background 50 is inconsistent, and the device start-stop status identification data is equal to the target start-stop status data, which means that the electromechanical device can reach the expected set start-stop status, indicating that there may be an abnormality in the data path between the electromechanical device and the visualization background 50, or there is an abnormality in the data processing of the visualization background 50 itself.

[0058] Through the above steps, the BAS visualization-based electromechanical integrated diagnosis system can cross-validate by combining background data with image recognition results. When the background data appears normal but the image recognition result is abnormal, the actual operation abnormality of the equipment can be quickly identified; conversely, when the background data is abnormal but the image recognition result appears normal, the system can intelligently determine it as a data link problem; this not only improves the accuracy of fault location, but also shortens the troubleshooting time, avoids unnecessary on-site inspections by maintenance personnel, and enables a more reasonable allocation of maintenance resources, thereby significantly improving the operation and maintenance efficiency and reliability of the entire system.

[0059] Furthermore, the step A7 comprises the following steps: A701, determining the corresponding device type value in a preset device value code comparison table according to the device type information; The equipment numerical code comparison table is a pre-set comparison table used to determine the equipment numerical code of each electromechanical device so as to distinguish different types of electromechanical devices; The device type value is the numerical code corresponding to the device type information in the device value code comparison table. For example, the device type value of a fan may be 050, the device type value of a lighting fixture may be 060, and so on.

[0060] A702, generating corresponding PLC feature data, working condition feature data and environmental feature data through a preset feature extraction algorithm according to the equipment PLC data, equipment working condition data and equipment environment data; The feature extraction algorithm is a pre-set algorithm used to extract corresponding feature data based on the collected data; PLC characteristic data is characteristic data corresponding to the device PLC data; The operating condition characteristic data is characteristic data corresponding to the equipment operating condition data; The environmental characteristic data is characteristic data corresponding to the device environmental data.

[0061] A703, generating abnormal equipment dimension reduction feature data according to the PLC feature data, the working condition feature data and the environmental feature data by using a preset feature dimension reduction algorithm; The feature dimension reduction algorithm is a pre-set dimension reduction algorithm used to reduce the dimension of PLC feature data, working condition feature data and environmental feature data to reduce the amount of data while retaining sufficient features; The dimension reduction feature data of abnormal equipment is the feature data after dimension reduction.

[0062] A704, generating abnormal operation equipment feature data according to the combination of equipment type value and abnormal equipment dimension reduction feature data; The feature data of abnormal operation equipment is a collection of equipment type values ​​and dimensionality reduction feature data of abnormal equipment.

[0063] A705, according to the characteristic data of the abnormally operating equipment, the corresponding estimated fault cause is determined by using the electromechanical fault identification model; The characteristic data is input into the electromechanical fault identification model to try to identify the corresponding fault cause.

[0064] Through the above steps, the BAS visualization-based electromechanical integrated diagnosis system can standardize the equipment type through the equipment numerical code comparison table to ensure that the fault judgment of different types of equipment is targeted; then, the key feature information is extracted from the PLC data, working condition data and environmental data through the feature extraction algorithm, and then the data dimension is efficiently compressed through the feature dimension reduction algorithm, retaining the core features while improving the calculation efficiency; finally, the equipment type value and the reduced dimension feature data are intelligently integrated, and the cause of the fault is accurately inferred through the pre-trained electromechanical fault identification model, which not only improves the accuracy of fault diagnosis, but also realizes personalized diagnosis for different equipment types, shortens the fault response time, and reduces equipment maintenance costs.

[0065] Furthermore, the electromechanical integrated diagnosis system based on BAS visualization also includes the following steps for training the electromechanical fault identification model: B1, obtaining corresponding historical operating fault electromechanical equipment data in a preset historical electromechanical database according to preset historical electromechanical equipment fault records; The historical electromechanical equipment fault record is the pre-recorded electromechanical equipment fault information that occurred in the past and the cause of the fault was determined; The historical electromechanical database is a pre-set database storing the historical operation data of electromechanical equipment, including the equipment PLC data, equipment working condition data, equipment environment data and operation status image data collected in the past; The historical operation fault electromechanical equipment data is the equipment PLC data, equipment operating condition data, equipment environment data and operation status image data etc. collected when the electromechanical equipment that failed in the past failed.

[0066] B2, determine the equipment type value of each electromechanical equipment with historical operating failures in the equipment value code comparison table; Get the equipment type value of each electromechanical equipment with historical operation failure.

[0067] B3, determining the corresponding equipment fault type value in a preset fault value code comparison table according to the fault cause information of each historical running fault electromechanical equipment in the historical electromechanical equipment fault record; The fault cause information is the classification information of the mechanical and electrical equipment fault; The fault numerical code comparison table is a pre-set comparison table used to digitize the fault type for data processing; The device fault type value is the value corresponding to the fault cause information.

[0068] B4, according to each historical running faulty electromechanical equipment, obtain the corresponding faulty equipment PLC data, faulty equipment working condition data and faulty equipment environment data from the historical running faulty electromechanical equipment data; The PLC data of the faulty equipment is the PLC data of the equipment collected when the faulty electromechanical equipment in historical operation occurs; The faulty equipment operating condition data is the equipment operating condition data collected when the faulty electromechanical equipment in historical operation fails; The faulty equipment environment data is the equipment environment data collected when the historical faulty electromechanical equipment fails.

[0069] B5, generating corresponding fault PLC feature data, fault condition feature data and fault environment feature data through feature extraction algorithm according to fault device PLC data, fault device working condition data and fault device environment data; The fault PLC characteristic data is the characteristic data corresponding to the PLC data of the faulty device; The fault condition characteristic data is characteristic data corresponding to the fault equipment condition data; The fault environment characteristic data is characteristic data corresponding to the fault device environment data.

[0070] B6, generating fault equipment dimension reduction feature data according to the fault PLC feature data, the fault working condition feature data and the fault environment feature data by using a feature dimension reduction algorithm; The dimension-reduced feature data of the faulty device is the feature data of the faulty device after dimension reduction.

[0071] B7, generating faulty equipment feature data according to the equipment type value corresponding to the historical faulty electromechanical equipment and the faulty equipment dimension reduction feature data combination; The faulty equipment characteristic data is a data collection of equipment type values ​​corresponding to the historically faulty electromechanical equipment and the reduced dimension characteristic data of the faulty equipment.

[0072] B8, generating a fault identification training set based on the fault equipment characteristic data of all historically faulty electromechanical equipment and the corresponding equipment fault type numerical combination; The fault identification training set is a collection of the faulty equipment feature data of all historically faulty electromechanical equipment and the corresponding equipment fault type values, which is used to train the corresponding model.

[0073] B9, training a preset recognition model through a fault recognition training set to generate an electromechanical fault recognition model; The fault identification training set is used as training data to train the corresponding model to generate an electromechanical fault identification model.

[0074] Through the above steps, the BAS visualization-based electromechanical integrated diagnosis system can achieve structured data processing through standardized equipment type and fault type numerical coding; the system uses feature extraction and dimensionality reduction algorithms consistent with diagnosis to process historical PLC data, operating condition data and environmental data, ensuring the consistency of model training and application scenarios; and combines the equipment type value with the dimensionality reduction feature data, so that the model can identify faults of different types of equipment, improves the versatility of the identification model, and provides technical support for equipment predictive maintenance and processing decisions.

[0075] Furthermore, the electromechanical integrated diagnosis system based on BAS visualization also includes the following steps: C1, obtain the device identification information corresponding to the data link abnormal device; Determine the device identification information of the device with abnormal data link.

[0076] C2, determining the corresponding link electromechanical equipment list according to the equipment identification information and the preset equipment network topology data; The equipment network topology data is the pre-collected topology structure data of the connection network between various electromechanical equipments; The link electromechanical equipment list is a list of all electromechanical equipment on the data path between the data link abnormal device and the visualization background 50 .

[0077] C3, according to each electromechanical device in the link electromechanical device list, obtain the corresponding device PLC data, device operating condition data, device environment data and corresponding device type information and combine them to generate comprehensive data of link device status; The comprehensive data of link device status is comprehensive status data of each electromechanical device on the data path between the data link abnormal device and the visualization background 50 .

[0078] C4, based on the comprehensive data of the status of each link device, try to identify and generate the corresponding estimated fault cause through the electromechanical fault identification model; The electromechanical fault identification model attempts to identify faults of each electromechanical device on the link to determine whether other electromechanical devices on the link may have abnormalities or faults, so as to facilitate fault troubleshooting.

[0079] Through the above steps, the BAS visualization-based electromechanical integrated diagnosis system can track the related equipment on the entire data transmission link of the data link abnormal device through the combination of equipment identification information and network topology data, and perform fault identification and diagnosis on each device on the link, thereby improving the efficiency of link fault troubleshooting. It not only speeds up the fault location speed in the link, but also can discover the mutual influence and reaction between equipment, providing a scientific basis for system optimization.

[0080] Furthermore, the electromechanical integrated diagnosis system based on BAS visualization also includes the following steps: D1, obtaining the historical operating status image data corresponding to each electromechanical device from the historical electromechanical database according to the preset inspection cycle; The inspection cycle is a pre-set cycle used to determine the time span for obtaining historical data. For example, the inspection cycle can be set to 48 hours, and historical data within the past 48 hours will be obtained; The historical operating status image data is the historical image data of the electromechanical equipment during operation, and may be image or video content.

[0081] D2, using the preset pre-order sampling time window and post-order sampling time window to sequentially acquire the adjacent pre-order time window historical state image data and post-order time window historical state image data from the historical operation state image data of each electromechanical device; The preceding sampling time window is a pre-set time window for collecting data within a corresponding time range; The post-order sampling time window is a pre-set time window for collecting data within the corresponding time range, wherein the time range of the post-order sampling time window is after the pre-order sampling time window; for example, the pre-order sampling time window and the post-order sampling time window are set to 1 minute respectively, then at the sampling time T, the corresponding pre-order sampling time window is [T-1min,T], and the post-order sampling time window is [T,T+1min]; The previous time window historical state image data is the image data corresponding to the previous sampling time window in the historical operation state image data; The subsequent time window historical state image data is the image data corresponding to the subsequent sampling time window in the historical operation state image data.

[0082] D3, generating corresponding equipment previous start / stop state recognition data through equipment state recognition model recognition according to previous time window historical state image data; The preceding start / stop state identification data of the device is the start / stop state identification data of the device determined based on the preceding time window historical state image data.

[0083] D4, generating corresponding equipment subsequent start / stop state recognition data through equipment state recognition model recognition according to the subsequent time window historical state image data; The subsequent start / stop state identification data of the equipment is the start / stop state identification data of the equipment determined based on the subsequent time window historical state image data.

[0084] D5, if the previous start / stop state identification data of the device is inconsistent with the subsequent start / stop state identification data of the device, the corresponding device state conversion time is determined according to the previous sampling time window and the subsequent sampling time window; If the equipment start / stop status identification data before and after are inconsistent, it means that the operation status of the electromechanical equipment has changed; The device state transition moment is the approximate moment when the operating state of the electromechanical equipment changes, which can be estimated based on the preceding sampling window and the following sampling window. For example, for the preceding sampling window [T-1min, T] and the following sampling window [T, T+1min], if the operation of the equipment changes, T can be regarded as the device state transition moment.

[0085] D6, according to the equipment state conversion time, obtain the corresponding equipment PLC data, equipment condition data, equipment environment data in the historical electromechanical database and combine them with the corresponding equipment type information to generate the state conversion equipment state comprehensive data; The state transition device state comprehensive data is the historical working state comprehensive data of the electromechanical equipment corresponding to the device state transition moment.

[0086] D7, based on the comprehensive data of the state of the state conversion device, attempts to identify and generate the corresponding estimated fault cause through the electromechanical fault identification model; Input the comprehensive data of the state transition equipment status into the electromechanical fault identification model to generate possible fault causes for reference by maintenance personnel; Usually, electromechanical equipment operates continuously. When the working state of the electromechanical equipment changes, an abnormality or failure may occur. By obtaining historical data when the working state changes, electromechanical equipment with potential abnormalities or failures can be discovered through the historical data.

[0087] Through the above steps, the BAS visualization-based electromechanical integrated diagnosis system can capture the equipment state transition points by regularly checking historical image data and adopting the pre-order and post-order time window analysis method; after determining the state transition moment, the system immediately conducts in-depth analysis of the PLC data, operating condition data and environmental data at that time point to construct a complete state transition comprehensive data set; and then analyzes it through the electromechanical fault identification model to achieve early detection of potential faults in the state transition process; this method of actively capturing state changes and predicting faults not only improves the system's sensitivity to fault signs, but also provides early warnings before the fault fully manifests, providing sufficient response time for preventive maintenance and significantly reducing equipment damage risks and maintenance costs.

[0088] The above are all preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Any feature disclosed in this specification (including the abstract and drawings), unless otherwise stated, can be replaced by other equivalent or alternative features with similar purposes. That is, unless otherwise stated, each feature is only an example of a series of equivalent or similar features.

Claims

1. A BAS-based integrated electromechanical diagnostic system, characterized in that: include: Equipment data acquisition module; Equipment image acquisition module; Data processing module; Equipment control module; Visual background; wherein, the equipment data acquisition module, the equipment image acquisition module, the equipment control module and the visual background data are connected to the data processing module; wherein, the equipment data acquisition module includes a PLC data acquisition module, an operating condition data acquisition module and an environmental data acquisition module, and the PLC data acquisition module, the operating condition data acquisition module and the environmental data acquisition module data are connected to the data processing module.

2. The electromechanical integrated diagnosis system based on BAS visualization according to claim 1 is characterized in that: The electromechanical integrated diagnosis system based on BAS visualization further includes a visualization diagnosis strategy, including the following steps: A1, obtaining the preset background data of the operating status and target operating status data of each electromechanical device in the visualization background; A2, obtaining the operating status image data of each electromechanical device through the device image acquisition module; A3, determining the corresponding device state consistency according to the operating status background data, target operating status data and corresponding operating status image data of each electromechanical device using a preset device state consistency algorithm; A4, if the device state consistency of the electromechanical device is false, the corresponding electromechanical device is defined as an abnormal operation device; A5, obtaining the device PLC data, device operating condition data and device environment data of the abnormal operation device through the device data acquisition module; A6, generating the abnormal operation device state comprehensive data according to the device PLC data, device operating condition data, device environment data and corresponding device type information; A7, determining the corresponding estimated fault cause according to the abnormal operation device state comprehensive data through the pre-trained electromechanical fault recognition model.

3. The electromechanical integrated diagnosis system based on BAS visualization according to claim 2 is characterized in that: Step A3 includes the following steps: A301, generating corresponding running state vector data according to running state background data; A302, generating corresponding target state vector data according to target running state data; A303, calculating corresponding device state Euclidean distance according to running state vector data and target state vector data; A304, calculating corresponding modulus according to target state vector data and defining it as target state vector modulus; A305, calculating corresponding running state deviation according to quotient of device state Euclidean distance and target state vector modulus; A306, if running state deviation is greater than preset device running state vector data, A307, if the running state deviation is less than or equal to the equipment running deviation threshold, the corresponding equipment start-stop state recognition data is generated according to the running state image data through the pre-trained equipment state recognition model recognition; A308, the corresponding target start-stop state data is determined according to the target running state data; A309, if the equipment start-stop state recognition data is not equal to the target start-stop state data, the equipment state consistency is defined as false; A310, if the equipment start-stop state recognition data is equal to the target start-stop state data, the equipment state consistency is defined as true.

4. The electromechanical integrated diagnosis system based on BAS visualization according to claim 3 is characterized in that: Step A4 includes the following steps: A401, if the device status consistency of the electromechanical equipment is false, then determine the corresponding background start-stop status data according to the running status background data; A402, if the background start-stop status data is equal to the target start-stop status data, and the device start-stop status identification data is not equal to the target start-stop status data, then define the corresponding electromechanical equipment as an abnormal running device; A403, if the background start-stop status data is not equal to the target start-stop status data, and the device start-stop status identification data is equal to the target start-stop status data, then define the corresponding electromechanical equipment as a data link abnormal device.

5. The electromechanical integrated diagnosis system based on BAS visualization according to claim 4 is characterized in that: Step A7 includes the following steps: A701, determining the corresponding equipment type value in the preset equipment numerical code comparison table according to the equipment type information; A702, generating corresponding PLC feature data, operating condition feature data and environmental feature data through a preset feature extraction algorithm according to the equipment PLC data, equipment operating condition data and equipment environment data; A703, generating abnormal equipment dimensionality reduction feature data according to the PLC feature data, operating condition feature data and environmental feature data using a preset feature dimension reduction algorithm; A704, generating abnormal operation equipment feature data according to a combination of the equipment type value and the abnormal equipment dimensionality reduction feature data; A705, determining the corresponding estimated fault cause according to the abnormal operation equipment feature data through an electromechanical fault identification model.

6. The electromechanical integrated diagnosis system based on BAS visualization according to claim 5 is characterized in that: The method further includes the following steps for training the electromechanical fault identification model: B1, obtaining corresponding historical operation fault electromechanical equipment data in a preset historical electromechanical database according to preset historical electromechanical equipment fault records; B2, determining the equipment type value of each historical operation fault electromechanical equipment in the equipment value code comparison table; B3, determine the corresponding equipment fault type value in the preset fault value code comparison table according to the fault cause information of each historical running fault electromechanical equipment in the historical running fault electromechanical equipment fault record; B4, obtain the corresponding fault equipment PLC data, fault equipment working condition data and fault equipment environment data from the historical running fault electromechanical equipment data according to each historical running fault electromechanical equipment; B5, generate the corresponding fault PLC feature data, fault working condition feature data and fault environment feature data through the feature extraction algorithm according to the fault equipment PLC data, fault equipment working condition data and fault equipment environment data; B6, generate the fault equipment dimension reduction feature data according to the fault PLC feature data, fault working condition feature data and fault environment feature data by the feature dimension reduction algorithm; B7, generate the fault equipment feature data according to the equipment type value corresponding to the historical running fault electromechanical equipment and the fault equipment dimension reduction feature data combination; B 8. Generate a fault identification training set based on the faulty equipment characteristic data of all historically faulty electromechanical equipment and the corresponding equipment fault type numerical combination; B9. Train a preset identification model through the fault identification training set to generate an electromechanical fault identification model.

7. The electromechanical integrated diagnosis system based on BAS visualization according to claim 6 is characterized in that: Further comprising the steps of: C1, obtain the device identification information corresponding to the data link abnormal device; C2, determining the corresponding link electromechanical equipment list according to the equipment identification information and the preset equipment network topology data; C3, according to each electromechanical equipment in the link electromechanical equipment list, obtain the corresponding equipment PLC data, equipment operating condition data, equipment environment data and corresponding equipment type information and combine them to generate comprehensive data on the link equipment status; C4, according to the comprehensive data on the status of each link equipment, try to identify and generate the corresponding estimated fault cause through the electromechanical fault identification model.

8. The electromechanical integrated diagnosis system based on BAS visualization according to claim 7 is characterized in that: Further comprising the steps of: D1, according to the preset inspection cycle, obtain the historical operation status image data corresponding to each electromechanical device from the historical electromechanical database; D2, using the preset pre-order sampling time window and post-order sampling time window to obtain the adjacent pre-order time window historical status image data and post-order time window historical status image data from the historical operation status image data of each electromechanical device in sequence; D3, according to the historical state image data of the previous time window, the corresponding equipment previous start-stop state identification data is generated through the equipment state identification model; D4, according to the historical state image data of the subsequent time window, the corresponding equipment subsequent start-stop state identification data is generated through the equipment state identification model; D5, if the equipment previous start-stop state identification data is inconsistent with the equipment subsequent start-stop state identification data, the corresponding equipment state conversion time is determined according to the previous sampling time window and the subsequent sampling time window; D6, according to the equipment state conversion time, the corresponding equipment PLC data, equipment operating condition data, equipment environment data are obtained from the historical electromechanical database and combined with the corresponding equipment type information to generate the state conversion equipment state comprehensive data; D7, according to the state conversion equipment state comprehensive data, the electromechanical fault identification model is used to try to identify and generate the corresponding estimated fault cause.

Images