Low-voltage electric work safety intelligent early warning method and system based on multi-dimensional data

Abstract

The application discloses a low-voltage electric work safety intelligent early warning method and system based on multidimensional data, relates to the low-voltage electric work safety early warning field, and the method comprises the following steps: in response to a starting instruction, a three-dimensional layout map of a low-voltage electric work area is acquired; the positions of each type of equipment in the low-voltage electric work area are determined based on the three-dimensional layout map, and the equipment data of each type of equipment is acquired in real time through a plurality of types of sensors; whether there is abnormal operation is determined through the equipment data; in the case that there is abnormal operation, alarm information is sent out, and the type of equipment corresponding to the abnormal operation is located; and the alarm information is sent to the type of equipment, so that the type of equipment can visualize the alarm information. The application can effectively solve the problem that safety hazards in the low-voltage electric work process cannot be identified and warned in time.

Description

TECHNICAL FIELD

[0001] The embodiment of the present application relates to the field of low-voltage electrical work safety early warning, and particularly relates to a low-voltage electrical work safety intelligent early warning method and system based on multi-dimensional data. BACKGROUND

[0002] In the practical operation link of the low-voltage electrical work examination room, it is crucial to monitor the safety hazards in the operation process of the examinee and to alarm and prompt the irregular operation. However, the current abnormality identification in the examination room is often realized through manual patrol. However, the manual patrol has certain limitations. The experience level of the evaluators or invigilators is a key factor in the abnormality identification link of the examination room, which directly and significantly affects their ability to identify irregular operations or safety hazards. The experienced evaluators can more sensitively perceive the subtle abnormalities in the operation of the examinee, and are more likely to discover and handle these problems at the first time. However, the less experienced evaluators may miss some key details due to lack of sufficient practical training and experience accumulation when facing complex examination room environment and variable examinee operation, which leads to that the potential safety hazards cannot be discovered and handled in time, and the abnormality identification of the examination room is inaccurate. Due to the inaccuracy of the abnormality identification, the safety hazards are brought to the examinee and the evaluator.

[0003] At present, there is no good solution to the above problem. SUMMARY

[0004] The embodiment of the present application provides a low-voltage electrical work safety intelligent early warning method and system based on multi-dimensional data, which is used to solve the problem that the safety hazards in the low-voltage electrical work process cannot be identified and warned in time, and improve the accuracy of the abnormality identification of the examination room.

[0005] To achieve the above purpose, the embodiments of the present application adopt the following technical solutions:

[0006] In a first aspect, a low-voltage electrical work safety intelligent early warning method based on multi-dimensional data is provided, which is applied to a low-voltage electrical work safety examination monitoring system, the low-voltage electrical work safety examination monitoring system comprising a plurality of types of sensors, a camera and a wearable device, and the method comprises:

[0007] In response to a start instruction, a three-dimensional layout map of a low-voltage electrical work area is acquired;

[0008] Based on the three-dimensional layout map, the positions of each type of device in the low-voltage electrical work area are determined, and device data of each type of device is acquired in real time through the plurality of types of sensors;

[0009] Whether there is an abnormal operation is determined through the device data;

[0010] In the case of abnormal operation, an alarm information is sent out, and the work type equipment corresponding to the abnormal operation is located;

[0011] The alarm information is sent to the work type equipment, so that the work type equipment can visualize the alarm information;

[0012] The abnormal operation includes device state abnormality of the work type equipment and electrical parameter abnormality of the work type equipment.

[0013] In a possible implementation manner of the first aspect, the multi-type sensor includes an electrical parameter sensor and a device state sensor, and the device data includes electrical parameter data and device data. The method for determining whether there is abnormal operation through the device data includes:

[0014] The electrical parameter set of the work type equipment and the device data set of the work type equipment are respectively acquired through the multi-type sensor;

[0015] Features of the electrical parameter set and the device data set are respectively extracted, to obtain an electrical parameter feature set and a device data feature set;

[0016] Based on the electrical parameter feature set and the device data feature set, it is determined whether there is device state abnormality or electrical parameter abnormality.

[0017] In a possible implementation manner of the first aspect, the low-voltage electrical work safety examination monitoring system further includes a camera and a wearable device, and the abnormal operation further includes personnel operation abnormality of an operator. The method further includes:

[0018] Biological feature data set of the operator is acquired through the camera and the wearable device;

[0019] It is determined whether there is personnel operation abnormality through the biological feature data set.

[0020] In a possible implementation manner of the first aspect, the electrical parameter abnormality includes grounding monitoring abnormality, three-phase phase absence abnormality and safety voltage abnormality. The feature of the electrical parameter set is extracted to obtain an electrical parameter feature set, including:

[0021] Three-phase voltage, grounding current value and grounding resistance value of the work type equipment are acquired in real time through the electrical parameter sensor;

[0022] Based on the three-phase voltage, the three-phase phase absence abnormality is determined;

[0023] The resistance average value and the resistance standard deviation in a preset time period are calculated through the grounding resistance value;

[0024] calculate a current average value and a current standard deviation in a preset time period through the ground current value;

[0025] extract features of the resistance average value and the resistance standard deviation, determine a ground resistance abnormal state feature and a ground resistance stable state feature of the type of equipment;

[0026] determine a ground monitoring abnormality through the ground resistance abnormal state feature and the ground resistance stable state feature;

[0027] extract features of the current average value and the current standard deviation, determine a ground current abnormal state feature and a ground current stable state feature of the type of equipment;

[0028] determine a safety voltage abnormality through the ground current abnormal state feature and the ground current stable state feature.

[0029] In a possible implementation manner of the first aspect, the three-phase voltage includes a first-phase voltage, a second-phase voltage, and a third-phase voltage, and the determining the three-phase open-phase abnormality based on the three-phase voltage includes:

[0030] delaying the first-phase voltage, the second-phase voltage, and the third-phase voltage by a preset phase angle respectively by using a preset phase-shifting circuit to obtain a first-delayed-phase voltage, a second-delayed-phase voltage, and a third-delayed-phase voltage;

[0031] adding the first-phase voltage and the second-delayed-phase voltage, adding the second-phase voltage and the third-delayed-phase voltage, and adding the third-phase voltage and the first-delayed-phase voltage to obtain a first voltage difference, a second voltage difference, and a third voltage difference respectively;

[0032] rectifying the first voltage difference, the second voltage difference, and the third voltage difference to obtain a first rectified voltage, a second rectified voltage, and a third rectified voltage;

[0033] judging whether the first rectified voltage, the second rectified voltage, and the third rectified voltage are zero;

[0034] if at least one of the first rectified voltage, the second rectified voltage, and the third rectified voltage is zero, determining the three-phase open-phase abnormality.

[0035] In a possible implementation manner of the first aspect, the issuing an alarm information in the case of an abnormal operation includes:

[0036] In a case where the abnormal operation is any one of the personnel operation abnormality, the equipment state abnormality and the electrical parameter abnormality, a first-level alarm information is sent out;

[0037] In a case where the abnormal operation is any two of the personnel operation abnormality, the equipment state abnormality and the electrical parameter abnormality, a second-level alarm information is sent out;

[0038] In a case where the abnormal operation is all of the personnel operation abnormality, the equipment state abnormality and the electrical parameter abnormality, a third-level alarm information is sent out.

[0039] In a possible implementation manner of the first aspect, the locating the work-type equipment corresponding to the abnormal operation comprises:

[0040] acquiring a time when the abnormal operation occurs;

[0041] performing time matching on the time when the abnormal operation occurs, the equipment data and the biological feature data of the operating personnel to obtain a time matching result;

[0042] performing space matching on the equipment data and the biological feature data of the operating personnel to obtain a space matching result;

[0043] combining the time matching result and the space matching result to determine the work-type equipment corresponding to the abnormal operation.

[0044] In a possible implementation manner of the first aspect, the performing time matching on the time when the abnormal operation occurs, the equipment data and the biological feature data of the operating personnel to obtain a time matching result comprises:

[0045] acquiring time stamps of the biological feature data set and the equipment data;

[0046] taking the time stamp when the abnormal operation occurs as a master clock, and taking the time stamps of the biological feature data set and the equipment data as slave clocks;

[0047] sending the time stamp of the master clock to the slave clocks by using a preset network time synchronization protocol;

[0048] acquiring a time stamp of network delay;

[0049] adjusting the time stamp of the slave clock by the time stamp of the master clock and the time stamp of the network delay to obtain a time matching result.

[0050] In a possible implementation manner of the first aspect, the performing space matching on the equipment data and the biological feature data of the operating personnel to obtain a space matching result comprises:

[0051] Taking any one of the work type equipment as an origin, a coordinate system is constructed;

[0052] The multiple types of sensors, the camera, the wearable device and the operator are mapped into the coordinate system;

[0053] A preset coordinate of a preset QR code in the coordinate system is acquired;

[0054] A measurement coordinate of the preset QR code in the coordinate system is measured by the camera, and the preset coordinate is corrected by the measurement coordinate to obtain a corrected coordinate system;

[0055] The device data and the biological feature data of the operator are converted into the corrected coordinate system;

[0056] In the corrected coordinate system, a spatial synchronization processing is performed on the device data and the biological feature data of the operator by using a preset spatial registration algorithm to obtain a spatial matching result.

[0057] In a second aspect, the present application provides an electronic device, comprising:

[0058] a memory configured to store instructions; and

[0059] a processor configured to call the instructions from the memory and capable of realizing the above-mentioned low-voltage electrical work safety intelligent early warning method based on multi-dimensional data when executing the instructions.

[0060] Through the above technical solution, the device data of each work type equipment is acquired in real time by the multiple types of sensors, which can comprehensively and accurately reflect the running state and electrical parameters of the equipment. When an abnormal operation is found, an alarm information can be immediately sent, and the specific work type equipment can be located, so that the management personnel or the evaluator can respond quickly. Moreover, the position of the work type equipment is determined by using the three-dimensional layout diagram, the work area is intuitively displayed, and the management personnel or the evaluator can understand the work environment. Through the visual alarm information, the work type equipment can intuitively display the abnormal state, reminding the examinee or the operator to pay attention to safety, which helps to reduce the safety accidents caused by human negligence, improves the safety management level of the examination room, effectively solves the problem that the safety hidden danger in the low-voltage electrical work process cannot be identified and warned in time, and further improves the accuracy of the examination room abnormal identification. By judging the operation process of the examinee, objective and accurate data support is provided for the examination evaluation, the operation skills and safety awareness of the examinee can be evaluated, targeted guidance is provided for training and education.

[0061] Other features and advantages of the embodiments of the present application will be described in detail in the following specific implementation part. BRIEF DESCRIPTION OF DRAWINGS

[0062] Figure 1 A flowchart of a low-voltage electrical work safety intelligent early warning method based on multi-dimensional data provided by an embodiment of the present application is shown.

[0063] Figure 2 A waveform diagram of a first phase voltage and a second phase voltage before and after phase shifting is provided by an embodiment of the present application.

[0064] Figure 3 A waveform diagram of a first phase voltage and a third phase voltage before and after phase shifting is provided by an embodiment of the present application. DETAILED DESCRIPTION

[0065] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described below in conjunction with the accompanying drawings in the embodiments of the present application. It should be understood that the specific implementation manners described herein are merely used to explain and illustrate the embodiments of the present application, and are not used to limit the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative effort fall within the scope of protection of the present application.

[0066] It should be noted that if the embodiments of the present application involve directional indications (such as up, down, left, right, front, back, etc.), the directional indications are merely used to explain the relative positional relationship, movement condition, etc. between components in a certain posture (as shown in the drawings), and if the certain posture changes, the directional indications also change accordingly.

[0067] In addition, if the embodiments of the present application involve descriptions such as “first”, “second”, etc., the descriptions of “first”, “second”, etc. are merely for description purposes, and cannot be understood as indicating or implying the relative importance of the indicated technical features or implicitly indicating the number of the indicated technical features. Therefore, the features limited by “first”, “second” can explicitly or implicitly include at least one of the features. In addition, the technical solutions of the various embodiments can be combined with each other, but it must be based on the fact that a person of ordinary skill in the art can implement it, and when the combination of technical solutions contradicts each other or cannot be implemented, it should be considered that the combination of technical solutions does not exist, and is not within the scope of protection claimed by the present application.

[0068] Figure 1 A flowchart of a low-voltage electrical work safety intelligent early warning method based on multi-dimensional data according to an embodiment of the present application is schematically shown. As shown in the figure, Figure 1As shown, the embodiment of the application provides a low-voltage electrical work safety intelligent early warning method based on multi-dimensional data, which is applied to a low-voltage electrical work safety examination monitoring system. The low-voltage electrical work safety examination monitoring system includes a plurality of types of sensors, a camera and a wearable device. The method can include the following steps.

[0069] S110, in response to a starting instruction, acquiring a three-dimensional layout map of a low-voltage electrical work area;

[0070] S120, determining the position of each type of device in the low-voltage electrical work area based on the three-dimensional layout map, and acquiring device data of each type of device in real time through the plurality of types of sensors;

[0071] S130, determining whether there is an abnormal operation through the device data;

[0072] S140, in the case of an abnormal operation, issuing an alarm information and positioning the type of device corresponding to the abnormal operation;

[0073] S150, sending the alarm information to the type of device to make the type of device visualize the alarm information;

[0074] S160, wherein the abnormal operation includes device state abnormality of the type of device and electrical parameter abnormality of the type of device.

[0075] Firstly, in response to a starting instruction, a three-dimensional layout map of a low-voltage electrical work area is acquired, that is, when an examiner triggers a starting instruction, the system starts to enter a monitoring and early warning state. After responding to the starting instruction, the three-dimensional layout map of the low-voltage electrical work area is acquired, which can be called from a preset database. The three-dimensional layout map can exist in the form of a digital model and contains the position information of all types of devices, facilities and obstacles in the work area.

[0076] Subsequently, the position of each type of equipment in the low-voltage electrical work area is determined based on the three-dimensional layout map, and the equipment data of each type of equipment is obtained in real time through multiple types of sensors. In this embodiment, the three-dimensional layout map provides spatial information of the low-voltage electrical work area, including the precise positions of various types of equipment, facilities, obstacles, etc. Specifically, the three-dimensional layout map constructed in advance or obtained in real time is loaded into the system, and the layout map is parsed through a spatial parsing algorithm to extract the position information of the type of equipment. In this embodiment, the spatial parsing algorithm can be a computer vision algorithm, that is, the position of each type of equipment in the three-dimensional layout map is determined through the computer vision algorithm. After determining the position of each type of equipment, the state data and electrical parameters of the equipment are obtained in real time through multiple types of sensors. When the position of each type of equipment is determined, the state data and electrical parameters of the equipment can be obtained in real time using multiple types of sensors. The multiple types of sensors can be current sensors, voltage sensors, temperature sensors, vibration sensors, etc., for monitoring the current, voltage, temperature, vibration, etc. of the equipment. Specifically, first, corresponding sensors are deployed near each type of equipment in the work area and on each type of equipment, and the sampling frequency, accuracy, etc. of the sensors are configured according to the characteristics and monitoring requirements of the equipment. Subsequently, the state data and electrical parameters of the equipment are collected in real time by the sensors, and the data is transmitted to the system, which processes and analyzes the collected data to extract key indicators and patterns for subsequent abnormal monitoring and early warning.

[0077] Next, it is determined whether there is an abnormal operation based on the equipment data. In this embodiment, the abnormal operation includes equipment state abnormality of the type of equipment, electrical parameter abnormality of the type of equipment, and personnel operation abnormality, that is, in the low-voltage electrical work safety examination monitoring, various data collected from the equipment (such as operating parameters, state indicators, operation records, etc.) are used to analyze and judge whether the equipment is running in a normal and expected manner, or whether the operator has non-normal operation behavior that does not conform to the conventional operation mode, which may indicate a fault or safety hazard. Specifically, first, various sensors, monitoring systems or data recording devices are needed to collect the operating data of the equipment in real time or periodically. These data may include temperature, pressure, speed, current, voltage, vibration frequency, etc. physical parameters, as well as the on-off state of the equipment, operation instructions, etc. logical information. Subsequently, the collected data is sent to the data analysis system for processing and analysis to identify abnormal patterns or values deviating from the normal range in the data. Based on the results of data analysis, the system determines whether the equipment has an abnormal operation. Once an abnormal operation is detected, the system triggers an alarm mechanism to notify the relevant personnel through sound, light, SMS, email, etc. The operator or maintenance personnel will take appropriate measures to investigate, diagnose and solve the problem according to the alarm information.

[0078] In the event of abnormal operation, an alarm message is issued, and the corresponding work equipment is located. Specifically, firstly, various types of sensors, cameras, and wearable devices are used to collect real-time operating data of the equipment, such as temperature, pressure, current, and vibration frequency. Then, the collected data is analyzed to identify abnormal patterns. Once an abnormal operation is detected, the system should immediately trigger the alarm mechanism. After issuing the alarm message, the abnormal data is analyzed to determine the location of the abnormal operation on the equipment. Layout diagrams and structural diagrams of the equipment can be used to further pinpoint the exact location of the abnormal operation. Based on the type and purpose of the equipment, the work type associated with the abnormal operation is determined.

[0079] Next, alarm information is sent to the relevant equipment to make the alarm information visible. First, it is necessary to determine which equipment needs to receive alarm information. This can be done via wired communication (such as Ethernet, serial communication, etc.). If the equipment supports wireless communication (such as Wi-Fi, Bluetooth, LoRa, etc.), alarm information can be sent wirelessly. After receiving the alarm information, the equipment needs to parse and process it, extract key information, and display it on the visual interface. To attract the operator's attention, the equipment can emit sound or light signals when receiving alarm information. Simultaneously, a confirmation mechanism should be provided, allowing the operator to disable the notification signals after confirming the alarm information.

[0080] In this embodiment, abnormal operations include abnormal equipment status and abnormal electrical parameters of the equipment. Abnormal equipment status refers to a significant difference between the physical or operational state of the equipment and its normal state during operation. This abnormality may manifest as decreased equipment performance, component damage, unstable operation, or inability to function properly. For example: decreased equipment performance manifests as lower-than-normal production capacity, processing accuracy, or operating efficiency; component damage manifests as wear, breakage, or deformation of one or more components; unstable operation manifests as vibration, noise, or temperature increases during operation; and inability to function properly manifests as complete cessation of operation or inability to perform predetermined tasks. Abnormal electrical parameters refer to significant differences between the electrical parameters (such as voltage, current, and frequency) of the equipment and its normal set values ​​or standard ranges during operation. This abnormality may indicate a fault or potential risk in the equipment's electrical system. For example: abnormal voltage manifests as excessively high or low input or output voltage, exceeding the normal range; abnormal current manifests as excessively high or low current, inconsistent with normal operating conditions; and abnormal frequency manifests as a mismatch between the equipment's operating frequency and the power supply frequency, causing the equipment to malfunction.

[0081] By acquiring real-time equipment data for each type of workstation using multiple types of sensors, the system can comprehensively and accurately reflect the equipment's operating status and electrical parameters. When abnormal operation is detected, alarm messages can be issued immediately, pinpointing the specific workstation equipment, enabling managers or assessors to respond quickly. Furthermore, the use of 3D layout maps to determine the location of each workstation provides a visual display of the work area, facilitating understanding of the work environment for managers or assessors. Visualized alarm messages allow for intuitive display of abnormal equipment conditions, reminding candidates or operators to pay attention to safety, helping to reduce safety accidents caused by human negligence and improving the safety management level of the examination site. By analyzing the candidates' operational processes, objective and accurate data support is provided for examination assessment, and candidates' operational skills and safety awareness can also be evaluated, providing targeted guidance for training and education.

[0082] In one embodiment of this invention, the multiple types of sensors include electrical parameter sensors and equipment status sensors. By using these multiple types of sensors, a set of electrical parameters and a set of equipment data for each type of equipment are acquired. The equipment data is then used to determine whether any abnormal operation exists. This method includes:

[0083] S210. By using multiple types of sensors, the electrical parameter set and equipment data set of the equipment for each type of work are obtained respectively;

[0084] S220. Extract the features from the electrical parameter set and the equipment data set respectively to obtain the electrical parameter feature set and the equipment data feature set;

[0085] S230. Based on the electrical parameter feature set and the equipment data feature set, determine whether there is an abnormal equipment status or an abnormal electrical parameter.

[0086] In this embodiment, multiple types of sensors include electrical parameter sensors and equipment status sensors. Electrical parameter sensors can be voltage sensors, current transformers, etc. Voltage sensors are used to monitor the input voltage of electrical equipment; current transformers are used to measure the current of high-current electrical equipment. Equipment status sensors can be vibration sensors and acoustic emission sensors. A vibration sensor is a sensor capable of measuring the vibration parameters of an object; it can convert vibration signals into electrical signals or other forms of signals for subsequent analysis and processing. In equipment status monitoring, vibration sensors are mainly used to monitor the vibration of equipment and assess its operating status and health condition. An acoustic emission sensor is a sensor capable of receiving acoustic emission signals and converting them into electrical signals. Acoustic emission signals are transient elastic wave signals generated by changes in the internal microstructure of materials during stress (such as crack propagation, plastic deformation, etc.); by capturing these signals, acoustic emission sensors can monitor internal damage and potential faults in equipment.

[0087] Multiple types of sensors are used to acquire sets of electrical parameters and equipment data for various types of equipment. Specifically, electrical parameter sensors, such as voltage and current sensors, acquire sets of voltage and current parameters; equipment status sensors, such as vibration and acoustic emission sensors, acquire sets of acoustic emission and vibration signal data. Furthermore, PLC or microcontroller programming can be integrated to implement status indication and feedback. The motor's running direction can be displayed on the operating interface, and LED indicators (green for forward rotation, red for reverse rotation) can monitor the motor's direction. An audible and visual alarm (such as a buzzer or flashing warning light) for abnormal rotation can be integrated with a direction sensor or current direction detection module to monitor the motor's rotation in real time. If the direction deviates from the set direction, an alarm is triggered and the power is automatically cut off, providing intuitive status feedback. In addition, intelligent alarms can monitor the grounding status of low-voltage electrical equipment and lines. When a grounding failure is detected, audible and visual alarms are immediately used to alert operators, ensuring equipment operation safety and personnel safety, and preventing electrical fires and other safety accidents caused by grounding faults.

[0088] After obtaining the electrical parameter set and the equipment data set of the workpiece, features are extracted from both sets to obtain the electrical parameter feature set and the equipment data feature set. In this embodiment, the electrical parameter feature set is a data set extracted from the original electrical parameter set that reflects the key characteristics of the equipment's electrical performance. The equipment data feature set is a data set extracted from the original equipment data set that reflects the equipment's operating status and mechanical health characteristics. Specifically, based on the equipment's characteristics and testing requirements, key parameters reflecting the equipment's electrical performance, such as voltage, current, and frequency, are selected. Subsequently, representative features, such as mean, variance, peak value, and harmonic content, are extracted from the selected parameters. The electrical parameter feature set can contain statistical features and frequency domain features. Statistical features can include the mean and variance of voltage, and the peak and RMS values ​​of current; frequency domain features can include power spectral density and harmonic content, used to analyze the frequency components of electrical signals. Next, based on the equipment type and monitoring requirements, key parameters reflecting the equipment's operating status, such as vibration and acoustic emission, are selected, and representative features, such as the spectral characteristics of vibration signals and the energy distribution of acoustic emission signals, are extracted from the selected parameters. Vibration characteristics can include vibration frequency, amplitude, phase, etc., which are used to assess the mechanical health of equipment; acoustic emission characteristics can include the energy, frequency, duration, etc. of acoustic emission signals, which are used to detect internal damage to equipment.

[0089] Based on the electrical parameter feature set and the equipment data feature set, it is determined whether there are any abnormal equipment conditions or electrical parameter anomalies. Specifically, the electrical parameter feature set and the equipment data feature set are collected, including key parameters such as voltage, current, power, frequency, vibration, and temperature. Representative features, such as mean, variance, peak value, and spectral characteristics, are extracted from the preprocessed data, and key features are selected for analysis according to the characteristics of the equipment and monitoring requirements. Pre-set reasonable threshold ranges are compared with the electrical parameter feature set and the equipment data feature set to identify abnormal fluctuations or deviations from normal ranges. Identifying abnormal fluctuations or deviations from normal ranges can be achieved through cluster analysis, anomaly detection algorithms, etc., to identify abnormal patterns in the data, determine whether there are any abnormal equipment conditions or electrical parameter anomalies, and assess the degree of abnormality in the equipment condition or electrical parameters based on the anomaly detection results.

[0090] By using multiple types of sensors to acquire sets of electrical parameters and equipment data for various types of equipment, and to determine whether there are any abnormalities in equipment status or electrical parameters, the electrical performance of the equipment can be more comprehensively reflected, the operating status and mechanical health characteristics of the equipment can be intuitively displayed, potential abnormalities or faults can be detected in a timely manner, measures can be taken to eliminate safety hazards, and the safe operation of the equipment can be ensured.

[0091] In one embodiment of this invention, the low-voltage electrical work safety examination monitoring system further includes a camera and a wearable device. Abnormal operation also includes abnormal operation by the operator. The method further includes:

[0092] S310. Acquire a set of biometric data of the operator through cameras and wearable devices;

[0093] S320. Determine whether there are any abnormalities in personnel operation through the biometric data set.

[0094] In this embodiment, the low-voltage electrical work safety examination monitoring system also includes a camera and wearable devices. Specifically, the camera can capture real-time images of the examination site, and the wearable devices include smart bracelets and similar devices. The camera and wearable devices acquire the operator's biometric data set. In this embodiment, the biometric data set includes electrocardiogram (ECG) signal characteristics, eye-tracking characteristics, and motion trajectory characteristics. ECG signal characteristics refer to the waveform of electrical conductance generated by heart activity. In biometric identification, ECG signals can also reflect the health status of the heart, used for health monitoring and disease prevention, preventing health problems and potential accidents for the operator. Eye-tracking technology tracks eye movements by measuring the gaze point or the movement of the eyeballs relative to the head. Eye-tracking characteristics can reflect the operator's attention allocation, visual search strategies, and other cognitive processes. In safety assessments, eye-tracking characteristics can be used to determine whether the operator is concentrating and following the correct operating procedures. Motion trajectory refers to the path of movement of the body or a part of the body during an action, possessing characteristics in terms of form, direction, and amplitude. Motion trajectory characteristics can reflect the operator's movement habits and skill level. In safety assessments, by analyzing the characteristics of operators' movement trajectories, it can be determined whether their operations are standardized and whether there are potential safety hazards.

[0095] Subsequently, the presence of abnormal human operation is determined using a biometric dataset. Specifically, the biometric dataset contains various biometric information of the operator, reflecting their physiological state, attention allocation, and movement habits. Key information reflecting the operator's behavioral characteristics is then extracted, such as the frequency of the electrocardiogram signal, the gaze trajectory of eye tracking, and the speed and amplitude of the movement trajectory. Machine learning and deep learning algorithms are then used to detect anomalies in the extracted features, identifying abnormal data that deviates significantly from normal patterns. Based on the output of the anomaly detection algorithm, it is determined whether the operator is exhibiting abnormal behavior or improper operation.

[0096] In this embodiment, another implementation method for determining whether abnormal operation exists by using electrical parameter feature sets, device data feature sets, and biometric feature data sets includes the following steps:

[0097] Construct a multi-source feature set from electrical parameter feature sets, equipment data feature sets, and biometric feature sets;

[0098] Extracting multidimensional features from a multi-source feature set;

[0099] Multidimensional features are mapped to third-order tensors, where the third-order tensors include time dimension, spatial dimension and channel dimension;

[0100] Based on the third-order tensor, a three-dimensional security situation matrix is ​​constructed, where the time dimension corresponds to the rows of the three-dimensional security situation matrix, the spatial dimension corresponds to the columns of the three-dimensional security situation matrix, and the channel dimension corresponds to the third dimension of the three-dimensional security situation matrix.

[0101] First, a multi-source feature set is constructed by combining the electrical parameter feature set, the equipment data feature set, and the biometric feature set. In other words, the three feature sets are merged into a single multi-source feature set by concatenating the feature vectors. Since features from different data sources may have different dimensions and distributions, the multi-source feature set needs to be standardized to ensure that all features are on the same scale. This can be achieved using methods such as Z-score standardization and Min-Max standardization.

[0102] Secondly, multidimensional features are extracted from the multi-source feature set. This can be done automatically using deep neural networks (such as CNNs). After obtaining the multidimensional features, they are mapped to a third-order tensor. In this embodiment, the third-order tensor includes a time dimension, a spatial dimension, and a channel dimension. A third-order tensor is a multidimensional array with three dimensions, defined as the time dimension, spatial dimension, and channel dimension. The time dimension represents the change of data over time; the spatial dimension represents the distribution of data in space; and the channel dimension represents different attributes or channels of the data. Specifically, the multidimensional features are divided into time, spatial, and channel dimensions. This needs to be determined based on the specific application scenario and data characteristics. Based on the result of the dimension division, the multi-source feature set is reshaped into a third-order tensor. This can be achieved using data processing tools or libraries such as NumPy and Pandas.

[0103] Subsequently, a three-dimensional security posture matrix is ​​constructed based on a third-order tensor. The time dimension corresponds to the rows of the matrix, the spatial dimension to the columns, and the channel dimension to the third dimension. In this embodiment, the time dimension represents the change of security events or states over time. In the three-dimensional security posture matrix, the time dimension corresponds to the rows, with each row representing the security posture at a specific point in time or time period. The spatial dimension represents the spatial distribution of security events or states, corresponding to the columns. Each column represents the security posture of a specific spatial location or region. The channel dimension represents different attributes or types of security data, corresponding to the third dimension (depth or layer) of the matrix. Each layer represents a specific security attribute or data type, such as intrusion detection system alarms, network traffic statistics, or user behavior patterns. Next, the data is aggregated or interpolated as needed to match the dimensional requirements of the three-dimensional matrix. Specifically, a three-dimensional matrix is ​​populated using electrical parameter feature sets, device data feature sets, and biometric feature sets. Each data point is then placed in its corresponding position within the matrix based on the mapping relationships between time, space, and channels. If data is missing at certain locations (e.g., no safety event occurred at a certain spatial location at a certain time point), it can be marked with zero or a specific missing value.

[0104] Determining the presence of abnormal operations by using electrical parameter feature sets, equipment data feature sets, and biometric data sets also includes the following steps:

[0105] Based on the time dimension, the matrix elements in the three-dimensional security situation matrix are updated to provide real-time feedback on the matrix elements in the three-dimensional security situation matrix.

[0106] Using a pre-defined tensor decomposition method, the updated three-dimensional security situation matrix is ​​decomposed into a tensor matrix to obtain the decomposed three-dimensional security situation matrix.

[0107] Using a preset filter, abnormal feature data is detected in the decomposed three-dimensional security situation matrix, and the detected abnormal feature data is weighted and highlighted.

[0108] Determine the dynamic risk index by analyzing abnormal feature data;

[0109] Dynamic risk indices are used to determine whether any abnormal operations exist.

[0110] First, based on the time dimension, the matrix elements in the three-dimensional security situation matrix are updated to provide real-time feedback. In this embodiment, the three-dimensional security situation matrix is ​​a high-order data structure used to represent network security situation. Specifically, the latest security data is collected from various security monitoring devices, systems, and logs, ensuring that the collected data covers the three dimensions of time, space, and channel. A fixed time interval (such as per second, per minute, per hour, etc.) is set, and the matrix elements are updated periodically according to this time interval. If data before a certain point in time is outdated or invalid, it can be marked as expired or deleted.

[0111] Secondly, the updated 3D security situation matrix is ​​decomposed using a preset tensor decomposition method to obtain the decomposed 3D security situation matrix. Tensor decomposition is a high-order data processing technique. In this embodiment, the preset tensor decomposition method can be CP decomposition, Tucker decomposition, etc. Subsequently, the updated 3D security situation matrix is ​​used as input, and the selected tensor decomposition method is applied. Through the decomposition process, the high-order security situation matrix is ​​decomposed into multiple low-order tensors or matrices to obtain the decomposed 3D security situation matrix. The decomposed result is usually a set of low-order tensors or matrices, representing the information of the original high-order security situation matrix.

[0112] Using pre-defined filters, anomaly detection is performed on the decomposed 3D security situation matrix data. The detected anomalies are then weighted and highlighted. Specifically, before anomaly detection, a suitable filter needs to be selected. The filter selection should be based on the data characteristics and anomaly detection requirements; for example, a median-based Hampel filter, a median filter, or other filters suitable for time series or multidimensional data can be used. The decomposed 3D security situation matrix is ​​then used as input data for the filter. This data contains multi-dimensional security information, such as time, space (e.g., network nodes), and security event types. The filter processes the input data to detect anomalies. The filter identifies outliers by comparing the differences or statistical characteristics of data points with surrounding data points. Next, the detected anomalies are weighted to highlight their importance in the overall dataset. This weighting can be visualized by increasing the weight of outliers, changing their color, or adjusting their size.

[0113] A dynamic risk index is determined by analyzing anomalous feature data. Specifically, in this embodiment, anomalous feature data refers to data points discovered during security monitoring or data analysis that significantly differ from normal patterns or expected behaviors. The dynamic risk index is an indicator used to quantitatively assess the current security situation. It is calculated based on the analysis results of anomalous feature data, combined with specific algorithms or models. The value of the dynamic risk index fluctuates with changes in the security situation, thus reflecting the current level of security risk. Machine learning algorithms, such as cluster analysis and classification algorithms, can be used to process the anomalous feature data, and a risk index model can be constructed based on the processing results. The anomalous feature data is then input into the risk index model to obtain the dynamic risk index.

[0114] The dynamic risk index is used to determine whether any abnormal operations are occurring. Specifically, a higher dynamic risk index value indicates a greater current security risk and a potential for abnormal operations or threats. Conversely, a lower dynamic risk index value indicates a relatively stable current security situation and a lower likelihood of abnormal operations.

[0115] By using cameras and wearable devices, combined with biometric data sets, to monitor abnormal human operations, we can not only improve the comprehensiveness and accuracy of monitoring, but also achieve real-time early warning and intervention, thereby enhancing the security and fairness of examinations and promoting the development of intelligence and automation.

[0116] In one embodiment of this invention, electrical parameter anomalies include grounding monitoring anomalies, three-phase phase loss anomalies, and safety voltage anomalies. Features of the electrical parameter set are extracted to obtain an electrical parameter feature set, including:

[0117] S410. The three-phase voltage, grounding current and grounding resistance values ​​of the equipment are obtained in real time through electrical parameter sensors.

[0118] S420. Based on the three-phase voltage, determine the three-phase phase loss anomaly;

[0119] S430: Calculate the average resistance and standard deviation of resistance over a preset time period using the grounding resistance value;

[0120] S440: Calculate the average current and standard deviation of the current within a preset time period using the grounding current value;

[0121] S450. Extract the characteristics of the average resistance and standard deviation of resistance to determine the abnormal state characteristics and stable state characteristics of the grounding resistance of the equipment.

[0122] S460. Determine grounding monitoring anomalies by using the characteristics of abnormal grounding resistance and the characteristics of stable grounding resistance.

[0123] S470. Extract the characteristics of the average current value and standard deviation of the current to determine the abnormal state characteristics and stable state characteristics of the grounding current of the equipment.

[0124] S480. Determine the safety voltage anomaly by using the characteristics of abnormal ground current state and stable ground current state.

[0125] In this embodiment, abnormal electrical parameters include grounding monitoring anomalies, three-phase phase loss anomalies, and safety voltage anomalies. Grounding monitoring anomalies refer to abnormal changes in the grounding status of the electrical system, which may involve excessive grounding resistance, broken grounding wires, or poor grounding. Three-phase phase loss anomalies refer to the phenomenon that one or more phases in a three-phase AC power system cannot supply power normally due to a fault. Safety voltage anomalies refer to the phenomenon that the voltage value in the electrical system exceeds the safe range. Safe voltage refers to a voltage that will not directly cause death or disability; under normal environmental conditions, the "extra-low safety voltage" that allows continuous contact is 36V.

[0126] The electrical parameter set is extracted to obtain an electrical parameter feature set. Specifically, firstly, the three-phase voltage, grounding current, and grounding resistance values ​​of the equipment are acquired in real time using electrical parameter sensors. In this embodiment, the three-phase voltage refers to the voltage values ​​of phases A, B, and C in a three-phase AC power system; the grounding current refers to the current flowing into the ground through the grounding wire when the electrical equipment is operating normally or experiencing a fault; and the grounding resistance refers to the resistance value between the grounding electrode of the electrical equipment and the ground. Specifically, a voltage sensor is connected to the three-phase power lines of the equipment. The sensor can capture and convert voltage signals in real time, outputting readable values ​​or signals, thereby acquiring the three-phase voltage values ​​in real time. A current sensor is connected in series with the grounding wire of the equipment. The sensor can measure the current in the grounding wire in real time and output the corresponding current value or signal. The grounding resistance value can be obtained by measuring the grounding resistance value in real time or periodically.

[0127] After obtaining the three-phase voltage, grounding current, and grounding resistance values, a three-phase phase loss anomaly is determined based on the three-phase voltage. This is achieved by monitoring the voltage values ​​of phases A, B, and C in the three-phase AC system to determine if one or more phases are experiencing a voltage loss. Specifically, a three-phase phase loss anomaly refers to a situation in a three-phase AC system where, due to some reason (such as line breakage, poor contact, power supply failure, etc.), one or more phases lose voltage, resulting in a three-phase voltage imbalance. This can be determined by using voltage sensors or voltage transmitters to monitor the voltage values ​​of phases A, B, and C in the three-phase AC system in real time, comparing the monitored voltage values ​​with the normal voltage values, and observing whether there are significant differences, thereby confirming the presence of a three-phase phase loss anomaly.

[0128] Next, the average resistance and standard deviation of the resistance over a preset time period are calculated using the grounding resistance values. In other words, this is the arithmetic mean of all grounding resistance measurements within the preset time period, and the square root of the average of the squares of the differences between each grounding resistance measurement and the average resistance. The average resistance can be calculated by assuming n grounding resistance measurements were performed within the preset time period. The formula for calculating the average resistance is:

[0129]

[0130] in, This represents the average resistance value, where R1, R2, ..., Rn are the resistance values ​​measured each time, and n is the number of measurements.

[0131] The formula for calculating the standard deviation of resistance is as follows:

[0132]

[0133] Among them, R i This indicates the resistance value measured each time. σ represents the average resistance. R Indicates the standard deviation of resistance;

[0134] Subsequently, the average current and standard deviation of the current within a preset time period are calculated using the grounding current values. The average current is the arithmetic mean of all grounding current measurements within the preset time period, reflecting the overall level or trend of the grounding current during that time period. The standard deviation of the current measures the degree of dispersion between the grounding current value and the average value, reflecting the magnitude of the fluctuation of the grounding current within the preset time period.

[0135] Assuming there are n grounding current measurements within a preset time period, denoted as I1, I2, ..., In, the formula for calculating the average current is:

[0136]

[0137] in, I represents the average current value, n represents the number of current measurements, and I i This represents the sum of n measured grounding current values, denoted as I1, I2, ..., In;

[0138] Next, the characteristics of the average resistance and standard deviation of resistance are extracted to determine the abnormal and stable characteristics of the grounding resistance of the equipment. Specifically, under normal circumstances, the average grounding resistance should remain within a relatively stable range. This range is usually determined by factors such as equipment design requirements, installation environment, and surface conditions. If the average grounding resistance remains within this range for a long time, the grounding system can be considered to be in a stable state. If the average grounding resistance suddenly deviates from the normal range, it may indicate a problem with the grounding system. For example, a significant increase in the average grounding resistance may mean an increase in the contact resistance between the grounding electrode and the ground surface, or poor contact in the grounding wire. Similarly, under stable conditions, the standard deviation of grounding resistance should be small, indicating that the measured values ​​of grounding resistance are relatively concentrated with small fluctuations, and the grounding system performance is stable. If the standard deviation of grounding resistance increases significantly, it indicates that the measured values ​​of grounding resistance are more dispersed with larger fluctuations, suggesting that the grounding system is affected by external factors, such as changes in surface humidity or corrosion of the grounding electrode, leading to unstable grounding resistance. In other words, if the average resistance deviates from the normal range and the standard deviation increases significantly, the grounding system can be considered to be in an abnormal state, requiring further inspection and repair. If the average resistance is within the normal range and the standard deviation is small, the grounding system can be considered to be in a stable state. Feature extraction can be performed using algorithms such as machine learning and deep learning to extract abnormal and stable grounding resistance characteristics of various equipment.

[0139] By analyzing the abnormal and stable characteristics of grounding resistance, grounding monitoring anomalies can be identified. Specifically, when the grounding resistance exceeds the normal range, abnormal characteristics will appear, such as a significant increase in grounding resistance value or abnormal electrical equipment. A significant increase in grounding resistance may be due to damage to the grounding device, insufficient surface humidity, or a fault in the grounding line. Abnormal electrical equipment may be due to equipment overload or abnormal switching, which may be related to the increased grounding resistance. In a stable state, the grounding resistance value is within the normal range, indicating that the current equipment meets the power system's requirements for grounding resistance and can ensure the effective conduction of fault current. Simultaneously, the electrical equipment operates normally without overload or abnormal switching. In other words, comparing and analyzing the abnormal and stable characteristics of grounding resistance can determine whether grounding monitoring is abnormal.

[0140] After identifying grounding monitoring anomalies, the characteristics of the average current and current standard deviation are extracted to determine the abnormal and stable grounding current characteristics of the equipment. In this embodiment, the average current refers to the sum of current measurements over a period of time divided by the number of measurements, reflecting the average level or central trend of the current. The current standard deviation is the square root of the sum of the squares of the deviations of the current measurements from the average value, reflecting the dispersion or fluctuation range of the current measurements. When the average grounding current deviates significantly from the normal range, it may indicate an anomaly in the grounding system. For example, a sudden increase in the average grounding current may mean a decrease in grounding resistance, a short circuit in the grounding line, or equipment leakage. If the standard deviation of the grounding current increases, it indicates that the fluctuation range of the current value has increased, which may indicate the presence of unstable factors in the grounding system. For example, poor grounding line contact, aging grounding equipment, or fluctuations in grounding resistance caused by environmental factors (such as changes in surface humidity). When the average current is within the normal range, it indicates that the average grounding current remains within the normal range, indicating that the grounding system is in a stable state; when the current standard deviation is small, it indicates that the standard deviation of the grounding current is small, indicating that the fluctuation range of the current value is small, and the grounding system is relatively stable. The features of the average current and the standard deviation of the current can be extracted using algorithms such as machine learning and deep learning, thereby determining the abnormal state features and stable state features of the grounding current of the equipment.

[0141] Subsequently, by analyzing the abnormal and stable characteristics of the grounding current, an abnormal safety voltage is determined. In this embodiment, the abnormal grounding current characteristic refers to the grounding current of the electrical equipment deviating from the normal range, which may manifest as excessively high or low current values, large current fluctuations, etc. The stable grounding current characteristic refers to the grounding current of the electrical equipment fluctuating within the normal range, with a small fluctuation range and relative stability. An abnormal safety voltage means that the operating voltage of the electrical equipment exceeds the safe voltage range. Specifically, when a grounding fault occurs in electrical equipment, the grounding current will increase significantly, which may also cause changes in the operating voltage of the equipment. If this change exceeds the safe voltage range, it may pose a danger to human health. For example, in a low-voltage power distribution system, if a single-phase grounding fault occurs, the voltage of the non-faulty phase may rise, exceeding the safe voltage limit. In other words, by comparing and analyzing the abnormal grounding current characteristics with the stable grounding current characteristics, it can be determined whether the voltage is within the safe range.

[0142] By acquiring and analyzing electrical parameters in real time, abnormal states in electrical systems can be detected and addressed promptly, ensuring the safe, stable, and efficient operation of these systems. This is of great significance for improving equipment reliability, reducing maintenance costs, and protecting personnel safety.

[0143] In one embodiment of this invention, the three-phase voltage includes a first-phase voltage, a second-phase voltage, and a third-phase voltage. Based on the three-phase voltage, determining a three-phase phase loss anomaly includes:

[0144] S510. Using a preset phase-shifting circuit, the first phase voltage, the second phase voltage, and the third phase voltage are delayed by a preset phase angle to obtain the first delayed phase voltage, the second delayed phase voltage, and the third delayed phase voltage.

[0145] S520. Add the first phase voltage to the second delayed phase voltage, add the second phase voltage to the third delayed phase voltage, and add the third phase voltage to the first delayed phase voltage to obtain the first voltage difference, the second voltage difference, and the third voltage difference, respectively.

[0146] S530, Rectify the first voltage difference, the second voltage difference and the third voltage difference to obtain the first rectified voltage, the second rectified voltage and the third rectified voltage;

[0147] S540. Determine whether the voltages of the first rectified voltage, the second rectified voltage, and the third rectified voltage are zero;

[0148] S550. If at least one of the first rectified voltage, the second rectified voltage, and the third rectified voltage is zero, then a three-phase phase loss anomaly is determined.

[0149] Figure 2 A waveform diagram of the first phase voltage and the second phase voltage before and after phase shift is provided for an embodiment of this application, such as... Figure 2 As shown, the first phase voltage and the second phase voltage are divided by a phase shifting circuit with a phase angle delay of 60°. Figure 3 A waveform diagram of the first phase voltage and the third phase voltage before and after phase shift is provided for an embodiment of this application, as shown below. Figure 3 As shown, the first phase voltage and the third phase voltage are divided by a phase shift circuit with a phase angle delay of 60°.

[0150] In this embodiment, the three-phase voltage includes a first-phase voltage, a second-phase voltage, and a third-phase voltage. The first-phase voltage corresponds to phase A in the three-phase electrical system; the second-phase voltage corresponds to phase B in the three-phase electrical system; and the third-phase voltage corresponds to phase C in the three-phase electrical system.

[0151] Using a preset phase-shifting circuit, the first, second, and third phase voltages are delayed by a preset phase angle to obtain a first delayed phase voltage, a second delayed phase voltage, and a third delayed phase voltage. In this embodiment, the preset phase-shifting circuit is used to change the phase angle of the alternating current. Specifically, the first phase voltage is input into the preset phase-shifting circuit, which configures the first phase voltage according to the required phase delay angle. The phase delay angle can be determined according to the actual situation; in this embodiment, the phase delay angle can be 60°. That is, by adjusting the inductance or capacitance value in the circuit, the output voltage (i.e., the first delayed phase voltage) lags behind the input voltage by 60°. Similarly, the second phase voltage is input into another preset phase-shifting circuit, and according to the same principle, the circuit parameters are adjusted to obtain the required phase delay, outputting the second delayed phase voltage. The same process is performed on the third phase voltage, using a third preset phase-shifting circuit to output the third delayed phase voltage, which lags behind the original third phase voltage by a specified phase angle. In this embodiment, the phase angles delayed by the first, second, and third phase voltages are 60°.

[0152] Next, the first phase voltage is added to the second delayed phase voltage, the second phase voltage is added to the third delayed phase voltage, and the third phase voltage is added to the first delayed phase voltage to obtain the first voltage difference, the second voltage difference, and the third voltage difference, respectively. Specifically, adding the first phase voltage to the second delayed phase voltage can be implemented using an adder circuit. An adder circuit refers to an electronic circuit that performs the addition operation of two or more input signals. That is, the adder circuit is used to superimpose the delayed voltage with the previous phase voltage. For example, superimposing Ua and Ub' yields Uab, where Ua represents the first phase voltage; Ub' represents the voltage of the second phase voltage after processing by the delay circuit; and Uab represents the first voltage difference. The sum obtained by adding Ua and Ub' using the adder circuit is denoted as Uab. In this embodiment, the first voltage difference refers to the voltage sum obtained by adding the first phase voltage Ua and the second phase voltage Ub' after processing by the delay circuit using the adder circuit. Similarly, an adder circuit is used to add the second-phase voltage to the third-delayed-phase voltage to obtain the second voltage difference. In this embodiment, the second voltage difference refers to the sum of voltages obtained by adding the second-phase voltage to the third-delayed-phase voltage processed by the delay circuit using the adder circuit. For example, superimposing Ub and Uc' yields Ubc, where Ub represents the second-phase voltage; Uc' represents the voltage of the third-phase voltage after processing by the delay circuit, and the sum is denoted as Ubc. In this embodiment, the second voltage difference refers to the sum of voltages obtained by adding the second-phase voltage Ub to the third-delayed-phase voltage Uc' using the adder circuit. Likewise, adding the third-phase voltage to the first-delayed-phase voltage can be achieved using an adder circuit; that is, the adder circuit is used to superimpose the delayed voltage with the previous phase voltage. For example, superimposing Uc and Ua' yields Uca, where Uc represents the third-phase voltage; Ua' represents the voltage of the first-phase voltage after processing by the delay circuit; the sum of Uc and Ua' is denoted as Uca. In this embodiment, the third phase voltage is obtained by adding the third phase voltage Uc to the first delayed phase voltage Ua' through an adder circuit.

[0153] Subsequently, the first voltage difference, the second voltage difference, and the third voltage difference are rectified to obtain a first rectified voltage, a second rectified voltage, and a third rectified voltage. In this embodiment, the first rectified voltage refers to the DC voltage obtained by rectifying the first voltage difference; the second voltage difference refers to the DC voltage obtained by rectifying the second voltage difference; and the third rectified voltage refers to the DC voltage obtained by rectifying the third voltage difference. Specifically, a preset rectifier circuit is used to rectify the first voltage difference, the second voltage difference, and the third voltage difference. A rectifier circuit is a circuit that converts alternating current (AC) into direct current (DC). It uses nonlinear components such as diodes to achieve unidirectional conductivity, thereby converting AC voltage into unidirectional pulsating DC voltage. In this embodiment, the rectifier circuit can be a full-wave rectifier bridge. A full-wave rectifier bridge can convert AC voltage into full-wave DC voltage. It consists of four diodes and can convert AC voltage into DC voltage with higher efficiency and a smoother output waveform. Using the first voltage difference Uab, the second voltage difference Ubc, and the third voltage difference Uca as input voltages, specifically, Uab is input to the input terminal of the full-wave rectifier bridge, and the output terminal of the rectifier bridge will output the first rectified voltage Vab; similarly, Ubc and Uca are input to the input terminals of the other two full-wave rectifier bridges to obtain the second rectified voltage Vbc and the third rectified voltage Vca.

[0154] After obtaining the first, second, and third rectified voltages, the system determines whether these voltages are zero. A comparator circuit can be used to determine if the rectified voltages are zero. A comparator circuit is an electronic circuit whose core function is to compare the magnitudes of two input voltages and output a logic level signal representing "high" or "low" based on the comparison result. Specifically, a comparator circuit is set up for each rectified voltage, Vab, Vbc, and Vca. The reference voltage of the comparator is set to a threshold voltage very close to zero. Considering the non-ideal factors of the rectifier circuit and power supply, the threshold voltage can be slightly adjusted as needed to ensure accuracy. If the rectified voltage is less than the threshold voltage, the comparator outputs a high level; if the rectified voltage is greater than the threshold voltage, the comparator outputs a low level. Subsequently, the outputs of the three comparators are combined using logic gates. If the outputs of all comparators indicate that the rectified voltage is close to 0 (i.e., all output high level or all output low level, depending on the comparator configuration), the logic gates output a signal indicating that the phase sequence is correct. If the output of any comparator indicates that the rectified voltage is not 0, the logic gates output a signal indicating that the phase sequence is incorrect or that a phase is missing. This allows the system to determine whether the voltages of the first rectified voltage, the second rectified voltage, and the third rectified voltage are zero.

[0155] If at least one of the first, second, and third rectified voltages is zero, a three-phase phase loss is confirmed. In other words, under normal circumstances, if the three-phase system is balanced and the phase sequence is correct, each rectified voltage should have a non-zero voltage value. If one of the rectified voltages is detected to be zero or close to zero, it may indicate a problem with the corresponding power supply or line.

[0156] By utilizing a pre-set phase-shifting circuit, calculating the voltage difference, rectifying, and determining whether the rectified voltage is zero, a three-phase phase loss anomaly can be identified. This not only improves the accuracy of phase loss detection and the robustness of the system, but also achieves rapid response and reduces maintenance costs.

[0157] In one embodiment of this invention, an alarm message is issued in the event of an abnormal operation, including:

[0158] S610. If the abnormal operation is any one of the following: abnormal personnel operation, abnormal equipment status, or abnormal electrical parameters, issue a Level 1 alarm message.

[0159] S620. In the case of any two of the abnormal operation conditions, namely abnormal personnel operation, abnormal equipment status, and abnormal electrical parameters, issue a level 2 alarm message.

[0160] S630. When abnormal operation occurs due to a combination of abnormal personnel operation, abnormal equipment status, and abnormal electrical parameters, a level three alarm message is issued.

[0161] In the event of abnormal operation, an alarm message is issued. Specifically, a Level 1 alarm message is issued if the abnormal operation is any one of the following: abnormal personnel operation, abnormal equipment status, or abnormal electrical parameters. In this embodiment, abnormal personnel operation refers to abnormal conditions caused by operator errors or improper operation, such as misoperation of switches or failure to follow operating procedures. Abnormal equipment status refers to abnormal conditions caused by equipment malfunctions or performance degradation, such as motor overheating, transformer oil leakage, or circuit breaker vibration. Abnormal electrical parameters refer to abnormal conditions where the electrical parameters of the power system deviate from the normal range, such as voltage fluctuations, current imbalances, or frequency deviations. In other words, when any one of the abnormal conditions of abnormal personnel operation, abnormal equipment status, or abnormal electrical parameters occurs, a Level 1 alarm message will be issued. A Level 1 alarm message indicates an abnormal situation that requires immediate attention, which can be marked by displaying a semi-transparent color block in the visualization layer of the electronic device.

[0162] A Level 2 alarm will be issued if any two of the following abnormal operation conditions are detected: abnormal personnel operation, abnormal equipment status, or abnormal electrical parameters. In other words, a Level 2 alarm will be issued when any two of these three abnormal conditions are detected simultaneously, indicating a more serious abnormal situation. When a Level 2 alarm occurs, the device with the abnormal condition will be displayed in a flashing red box in the electronic device's visualization hierarchy, along with voice prompts such as "Stop operation!" or "Left cable overload."

[0163] When three abnormal operations occur simultaneously—human error, equipment status abnormality, and electrical parameter abnormality—a Level 3 alarm message is issued. In other words, a Level 3 alarm message indicates an extremely serious abnormality that may cause significant damage to the entire system or equipment. When a Level 3 alarm message appears, a full-screen mask can be applied to the electronic device's visualization layer. A full-screen mask refers to covering the user interface with a semi-transparent or opaque background, obscuring other content and focusing the user's attention on the current task or prompt. The abnormal device is then highlighted with a striking color (such as red or orange). Semi-transparent masks (such as black or gray) retain some background visibility, while opaque masks completely obscure the background. A pulsed navigation arrow can also be added to draw the user's attention. A pulsed navigation arrow is a dynamic visual element that attracts the user's attention through periodic changes (such as size, color, or transparency) and guides the user to complete specific operations. For example, it can clearly indicate the action the staff needs to perform (such as clicking a button to indicate that the staff has confirmed the abnormal device status).

[0164] By implementing a tiered alarm mechanism, abnormal operations can be handled more efficiently, response speed and accuracy can be improved, resource allocation can be optimized, system stability can be enhanced, security management can be improved, and continuous improvement can be promoted.

[0165] In one embodiment of this invention, locating the work equipment corresponding to the abnormal operation includes:

[0166] S710, Obtain the time when the abnormal operation occurred;

[0167] S720. Perform time matching on the time of the abnormal operation, equipment data, and operator biometric data to obtain the time matching result;

[0168] S730. Spatial matching is performed between equipment data and operator biometric data to obtain spatial matching results;

[0169] S740. Combining the time matching results and the space matching results, determine the type of equipment corresponding to the abnormal operation.

[0170] First, the time of the abnormal operation is obtained by checking the logs of the low-voltage electrical work safety examination monitoring system. These logs include timestamps. Second, the time of the abnormal operation, equipment data, and operator biometric data are time-matched to obtain the matching results. Specifically, it is necessary to ensure that the abnormal operation records, equipment data records, and biometric data records all contain accurate timestamps. Then, by comparing these timestamps, the equipment data and biometric data closest to the time of the abnormal operation are found. Due to various factors (such as data transmission delays and equipment response times), there may be slight differences between timestamps. Therefore, a reasonable time window can be set, considering equipment data and biometric data occurring within this time window as related to the abnormal operation. Using this time window, the time of the abnormal operation, equipment data, and operator biometric data within the time window are time-matched. These data contain timestamp information for subsequent time matching to obtain the matching results.

[0171] After obtaining the time matching results, the equipment data and the operator's biometric data are spatially matched to obtain spatial matching results. Spatial matching refers to aligning and associating data with spatial attributes (such as equipment location data, location information in the operator's biometric data, etc.) to reveal the spatial relationship between them. Specifically, equipment data and operator biometric data are collected, ensuring that these data contain spatial attributes (such as coordinate information, address descriptions, etc.). Based on the spatial attributes of the data, the equipment data and operator biometric data are aligned spatially. Appropriate spatial matching algorithms can be selected, such as matching based on coordinate information, matching based on description information, etc., and the selected matching algorithm is applied to spatially match the equipment data and operator biometric data. In this embodiment, a data association algorithm can also be used to further determine which equipment data and biometric data are most relevant to abnormal operations, thereby associating the data. Subsequently, the correlation between equipment data and biometric data and abnormal operations is analyzed. For example, it can be checked whether the equipment data shows abnormal states (such as excessive temperature, excessive current, etc.) and whether the biometric data shows abnormal behavior of the operator (such as abnormal operating patterns, abnormal physiological reactions, etc.). Based on the results of time matching and data correlation analysis, possible causes of abnormal operations can be inferred. For example, if the equipment data shows an abnormal state and the biometric data shows that the operator is fatigued or distracted, it may be inferred that the abnormal operation is caused by equipment failure or operator inattention.

[0172] By combining temporal and spatial matching results, the type of equipment corresponding to the abnormal operation is determined. Specifically, the results of temporal and spatial matching are comprehensively analyzed to identify equipment data and biometric data that simultaneously meet the temporal and spatial conditions. This data serves as the key basis for determining the type of equipment corresponding to the abnormal operation. Based on the comprehensive data, the type, model, and location of the equipment related to the abnormal operation are identified, which can be achieved by consulting equipment databases and comparing equipment identification information. Subsequently, based on the type, model, and purpose of the equipment, the type of work related to the abnormal operation is determined. Finally, the identified type of equipment is confirmed and verified to ensure the accuracy and reliability of the results. This can be achieved through communication with on-site operators, equipment managers, or production process experts, and by comparing with the actual on-site conditions.

[0173] By locating the work equipment corresponding to abnormal operations, the accuracy of location can be significantly improved, the ability to detect anomalies can be enhanced, safety management and decision-making can be optimized, and data analysis and processing capabilities can be improved, which helps to improve the efficiency and level of safety management.

[0174] In one embodiment of this example, the time of the abnormal operation, equipment data, and the operator's biometric data are time-matched to obtain a time-matching result, including:

[0175] S810, Acquire the timestamps of the biometric data set and device data;

[0176] S820: Use the timestamp of the abnormal operation as the master clock, and use the timestamps of the biometric data set and device data as slave clocks;

[0177] S830 uses a preset network time synchronization protocol to send the master clock's timestamp to the slave clock;

[0178] S840, Get the timestamp of network latency;

[0179] S850 adjusts the slave clock's timestamp using the master clock's timestamp and the network delay timestamp to obtain a time matching result.

[0180] First, the timestamps for the biometric data set and equipment data are obtained. This can be done by searching for records containing biometric and equipment data in the low-voltage electrical work safety examination monitoring system. These records should contain a timestamp associated with each data point, indicating the time the data was recorded. The biometric data set, as the name suggests, is a collection of biometric data. In this embodiment, biometric data can be electrocardiogram (ECG) signal data, eye-tracking data, and motion trajectory data. An ECG signal refers to the electrical signal generated by the heart during beating; eye-tracking data refers to the tracking and recording of eye movements using specific devices (such as an eye tracker); and motion trajectory data refers to the path formed when a human body or object moves in space.

[0181] Secondly, the timestamp of the abnormal operation is used as the master clock, and the timestamps of the biometric data set and device data are used as slave clocks. In other words, the timestamp of the abnormal operation is set as the master clock, indicating that the time of the abnormal operation is considered the reference time, and the timestamps of other data will be synchronized or compared with it. The role of the master clock is to provide a unified time reference point for the entire system or dataset, facilitating subsequent data analysis and troubleshooting. Simultaneously, the timestamps of the biometric data set and device data are set as slave clocks. These timestamps are used to synchronize or compare with the master clock (the timestamp of the abnormal operation) to ensure data consistency and accuracy. The role of the slave clocks is to record the generation time of the biometric data and device data, facilitating the correlation of this data with the abnormal operation in subsequent analysis.

[0182] Subsequently, using a preset network time synchronization protocol, the master clock's timestamp is sent to the slave clock. A network time synchronization protocol is a protocol used to synchronize computer system clocks over a network, enabling computers on the network to calibrate their system clocks through network communication, ensuring that all system clocks remain highly consistent. In this embodiment, the preset network time synchronization protocol can be the NTP protocol, which is used to synchronize computer clocks. Specifically, first, the relevant parameters of the network time synchronization protocol are configured on the master clock, including the server address, port number, synchronization interval, etc., to ensure that the master clock can send timestamp information. Second, the client parameters of the network time synchronization protocol are configured on the slave clock, enabling it to receive and apply the timestamp information from the master clock, including setting the server address, port number, and synchronization strategy. Once configured, the slave clock will begin attempting to communicate with the master clock and synchronize its time. This typically involves sending a message requesting a timestamp to the master clock and receiving the timestamp information returned by the master clock.

[0183] To obtain network latency timestamps, firstly, select an appropriate measurement method based on the application scenario and requirements, such as the Ping command. Record the sending timestamp when sending data packets and the receiving timestamp when receiving data packets. Ensure the accuracy and synchronization of the timestamps. Calculate the network latency based on the recorded timestamps. For round-trip time (RTT), the sending timestamp can be subtracted from the receiving timestamp. For one-way time, more complex measurement and calculation methods may be required. Analyze the calculated network latency data to evaluate network performance and stability. Reports or charts can be generated to visualize the latency data as needed.

[0184] After obtaining the network latency timestamp, the slave clock's timestamp is adjusted using the master clock's timestamp and the network latency timestamp to achieve time matching. Specifically, the network latency period is measured by sending and receiving data packets, and the send and receive timestamps are recorded, with the difference between them calculated. Then, the calculated adjustment amount is added to the slave clock's timestamp to obtain the adjusted timestamp. This adjusted timestamp should be closer to the master clock's timestamp. The slave clock's timestamp is then adjusted based on this adjusted timestamp, thus achieving time synchronization. The time matching result refers to the degree of matching between the slave clock's timestamp and the master clock's timestamp after the adjustment process.

[0185] By performing time matching between the time of abnormal operation, equipment data, and operator biometric data, a high-precision correlation was achieved between these three data points. This not only improved the accuracy and relevance of data analysis but also provided stronger security compliance and decision support capabilities.

[0186] In one embodiment of this invention, spatial matching is performed between device data and operator biometric data to obtain spatial matching results, including:

[0187] S910. Construct a coordinate system using any piece of equipment as the origin;

[0188] S920 maps various types of sensors, cameras, wearable devices, and operators to a coordinate system;

[0189] S930. Obtain the preset coordinates of the QR code in the preset coordinate system;

[0190] S940. Measure the coordinates of a preset QR code in the coordinate system using a camera, and correct the preset coordinates using the measured coordinates to obtain the corrected coordinate system.

[0191] S950, Convert equipment data and operator biometric data into the calibrated coordinate system;

[0192] S960. In the calibrated coordinate system, the equipment data and the operator's biometric data are spatially synchronized using a preset spatial registration algorithm to obtain spatial matching results.

[0193] Spatial matching is performed between equipment data and operator biometric data to obtain spatial matching results. Specifically, firstly, a coordinate system is constructed using any type of equipment as the origin. This involves selecting one type of equipment from among many options as the origin and then determining a specific point on that selected equipment as the origin. Next, based on the equipment's geometry and operational requirements, a mathematical model is established to describe the coordinate system. This model may include information such as the equipment's dimensions, shape, and positional relationships. Finally, using the mathematical model and measuring tools (such as a coordinate measuring machine), the coordinate values ​​of each point within the working area are determined within the coordinate system.

[0194] Subsequently, various types of sensors, cameras, wearable devices, and operators are mapped onto a coordinate system. For each type of sensor, their data is transformed into a unified coordinate system based on their installation location and orientation. The spatial positions of cameras in the 3D layout are mapped to the coordinate system, as are the wearable devices, which are then mapped to the unified coordinate system.

[0195] The preset coordinates of the QR code in the coordinate system are obtained. In this embodiment, the preset QR code usually determines its position in a specific coordinate system. If the coordinates of the QR code have been explicitly marked in the coordinate system when it is set, then this information can be read directly.

[0196] The process involves using a camera to measure the coordinates of a pre-defined QR code within a coordinate system, and then correcting these pre-defined coordinates to obtain a corrected coordinate system. Specifically, the camera captures an image of the scene containing the QR code, and image processing algorithms (such as edge detection and contour extraction) are used to identify the QR code's position. The unique features of the QR code (such as its position detection pattern and correction pattern) are then used for precise positioning. Based on the parameters obtained from camera calibration and the QR code's position in the image, a 3D reconstruction algorithm is used to convert the QR code's coordinates from image coordinates to measured coordinates in 3D space. Subsequently, the measured QR code coordinates are compared with the pre-defined coordinates. These pre-defined coordinates are typically determined during QR code deployment based on actual needs, and the error between the measured and pre-defined coordinates is calculated. Based on the error analysis results, an appropriate correction algorithm (such as least squares or iterative nearest point algorithm) is used to correct the pre-defined coordinates. The goal of this correction is to make the pre-defined coordinates as consistent as possible with the measured coordinates, thus obtaining the corrected coordinate system.

[0197] Subsequently, the equipment data and operator biometric data are transformed into a calibrated coordinate system. Specifically, spatial information transformation is performed on the equipment data and operator biometric data. First, the spatial location information (such as coordinates) of the equipment corresponding to the electrical parameter data is determined. Then, based on the spatial location information of the equipment, the electrical parameter data is transformed to a unified spatial reference, which can be achieved using a Geographic Information System (GIS) or related spatial data processing software. For operator biometric data, such as facial recognition, 3D biometric data can be transformed from different perspectives to the same perspective or coordinate system using 3D reconstruction technology. For 2D biometric data such as fingerprints and iris scans, spatial information transformation can be performed using methods such as image registration and affine transformation. Next, the spatial location information (such as coordinates) of the equipment corresponding to the equipment status data is determined. Then, based on the spatial location information of the equipment, the spatial information in the equipment status data is transformed to a unified spatial reference, which can be achieved using a Geographic Information System (GIS) or related spatial data processing software.

[0198] Finally, in the corrected coordinate system, a preset spatial registration algorithm is used to perform spatial synchronization processing on the device data and the operator's biometric data to obtain a spatial matching result. In this embodiment, the preset spatial registration algorithm can be selected according to the specific application scenario and data characteristics. The preset spatial registration algorithm can be SIFT (Scale Invariant Feature Transform), which is a classic image feature extraction algorithm. Specifically, the preset spatial registration algorithm is used to transform the device data and biometric data into the corrected coordinate system. Based on the spatial registration, the device data and biometric data are matched, including calculating the spatial relationships such as distance and angle between them to determine whether they are in the same spatial location or nearby.

[0199] By real-time monitoring and matching of equipment data and operator biometric data, and performing spatial matching, equipment can be identified and operated more accurately, improving work efficiency and response speed. It can also promptly detect potential safety hazards and abnormal behaviors, allowing for appropriate preventative and handling measures.

[0200] This application also provides an electronic device, including:

[0201] The memory is configured to store instructions; and

[0202] The processor is configured to retrieve instructions from memory and, when executing instructions, to implement the aforementioned intelligent early warning method for low-voltage electrical work safety based on multi-dimensional data.

[0203] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0204] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1A device that provides the functions specified in one or more boxes.

[0205] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0206] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0207] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0208] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0209] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0210] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0211] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for intelligent early warning of low-voltage electrical work safety based on multi-dimensional data, characterized in that, This method is applied to a low-voltage electrical work safety examination and monitoring system, which includes various types of sensors, cameras, and wearable devices. In response to the start command, obtain a three-dimensional layout map of the low-voltage electrical work area; The location of each type of equipment in the low-voltage electrical work area is determined based on the 3D layout map, and the equipment data of each type of equipment is acquired in real time through multiple types of sensors. Determine if any abnormal operation exists by analyzing device data; In the event of abnormal operation, an alarm message is issued and the corresponding work equipment is located; Send alarm information to the specific equipment so that the equipment can visualize the alarm information. Abnormal operations include abnormal equipment status and abnormal electrical parameters of the equipment. The location and abnormal operation of the corresponding work equipment include: Get the time when the abnormal operation occurred; Acquire timestamps for biometric data sets and device data; Use the timestamp of the abnormal operation as the master clock, and the timestamps of the biometric data set and device data as slave clocks; Using a preset network time synchronization protocol, the master clock's timestamp is sent to the slave clock; Get the timestamp of network latency; The timestamp of the slave clock is adjusted using the timestamp of the master clock and the timestamp of the network delay to obtain the time matching result; Spatial matching is performed between equipment data and operator biometric data to obtain spatial matching results; By combining time matching results and spatial matching results, the type of equipment corresponding to the abnormal operation is determined; By using sets of electrical parameter characteristics, equipment data characteristics, and biometric data, it is determined whether any abnormal operations exist, including: Construct a multi-source feature set from electrical parameter feature sets, equipment data feature sets, and biometric feature sets; Extracting multidimensional features from a multi-source feature set; Multidimensional features are mapped to third-order tensors, where the third-order tensors include time dimension, spatial dimension and channel dimension; Based on the third-order tensor, a three-dimensional security situation matrix is ​​constructed, where the time dimension corresponds to the row of the three-dimensional security situation matrix, the spatial dimension corresponds to the column of the three-dimensional security situation matrix, and the channel dimension corresponds to the third dimension of the three-dimensional security situation matrix. Determining the presence of abnormal operations by using electrical parameter feature sets, equipment data feature sets, and biometric data sets also includes: Based on the time dimension, the matrix elements in the three-dimensional security situation matrix are updated to provide real-time feedback on the matrix elements in the three-dimensional security situation matrix. Using a pre-defined tensor decomposition method, the updated three-dimensional security situation matrix is ​​decomposed into a tensor matrix to obtain the decomposed three-dimensional security situation matrix. Using a preset filter, abnormal feature data is detected in the decomposed three-dimensional security situation matrix, and the detected abnormal feature data is weighted and highlighted. Determine the dynamic risk index by analyzing abnormal feature data; Dynamic risk indices are used to determine whether any abnormal operations exist.

2. The method according to claim 1, characterized in that, The multiple types of sensors include electrical parameter sensors and equipment status sensors, and the equipment data includes electrical parameter data and equipment data. The method for determining whether abnormal operation exists through equipment data includes: By using multiple types of sensors, the electrical parameter sets and equipment data sets of each type of equipment are obtained respectively; Features are extracted from the electrical parameter set and the equipment data set respectively to obtain the electrical parameter feature set and the equipment data feature set; Based on the electrical parameter feature set and the equipment data feature set, determine whether there are any abnormal equipment status or electrical parameter abnormalities.

3. The method according to claim 2, characterized in that, The low-voltage electrical work safety examination and monitoring system also includes a camera and wearable devices; the abnormal operation also includes abnormal operation by the operator; and the method further includes: Acquire a set of biometric data of operators through cameras and wearable devices; By using biometric data sets, it can be determined whether there are any abnormalities in personnel operations.

4. The method according to claim 2, characterized in that, The electrical parameter anomalies include grounding monitoring anomalies, three-phase phase loss anomalies, and safety voltage anomalies. The extraction of features from the electrical parameter set yields an electrical parameter feature set, including: The three-phase voltage, grounding current and grounding resistance values ​​of the equipment are obtained in real time through electrical parameter sensors. Based on the three-phase voltage, a three-phase phase loss anomaly was determined; The average resistance and standard deviation of the resistance over a preset time period are calculated using the grounding resistance value. The average current and standard deviation of the current over a preset time period are calculated using the grounding current value. Extract the characteristics of the average resistance and standard deviation of the resistance to determine the abnormal state characteristics and stable state characteristics of the grounding resistance of the equipment for a particular type of work. Anomalies in grounding monitoring can be identified by analyzing the characteristics of abnormal and stable grounding resistance. Extract the characteristics of the average current value and the standard deviation of the current to determine the abnormal state characteristics and stable state characteristics of the grounding current of the equipment. Anomalies in the safety voltage are determined by the characteristics of abnormal ground current and stable ground current.

5. The method according to claim 4, characterized in that, The three-phase voltage includes the first-phase voltage, the second-phase voltage, and the third-phase voltage. The determination of a three-phase loss anomaly based on the three-phase voltage includes: Using a preset phase-shifting circuit, the first phase voltage, the second phase voltage, and the third phase voltage are delayed by a preset phase angle to obtain the first delayed phase voltage, the second delayed phase voltage, and the third delayed phase voltage; Add the first phase voltage to the second delayed phase voltage, add the second phase voltage to the third delayed phase voltage, and add the third phase voltage to the first delayed phase voltage to obtain the first voltage difference, the second voltage difference, and the third voltage difference, respectively. The first voltage difference, the second voltage difference, and the third voltage difference are rectified to obtain the first rectified voltage, the second rectified voltage, and the third rectified voltage; Determine whether the voltages of the first rectified voltage, the second rectified voltage, and the third rectified voltage are zero; If at least one of the first, second, and third rectified voltages is zero, then a three-phase phase loss anomaly is determined.

6. The method according to claim 1, characterized in that, The alarm message issued in the event of abnormal operation includes: A Level 1 alarm message will be issued if the abnormal operation is caused by any one of the following: abnormal personnel operation, abnormal equipment status, or abnormal electrical parameters. A level 2 alarm message will be issued if the abnormal operation is caused by any two of the following: abnormal personnel operation, abnormal equipment status, and abnormal electrical parameters. When abnormal operation occurs due to a combination of abnormal personnel operation, abnormal equipment status, and abnormal electrical parameters, a level three alarm message will be issued.

7. The method according to claim 1, characterized in that, The step of spatially matching equipment data and operator biometric data to obtain spatial matching results includes: Construct a coordinate system using any one type of equipment as the origin; Map multiple types of sensors, cameras, wearable devices, and operators to a coordinate system; Obtain the preset coordinates of the QR code in the preset coordinate system; The coordinates of a preset QR code in the coordinate system are measured using a camera, and the preset coordinates are corrected using the measured coordinates to obtain the corrected coordinate system. Convert equipment data and operator biometric data into a calibrated coordinate system; In the calibrated coordinate system, a preset spatial registration algorithm is used to perform spatial synchronization processing on the equipment data and the operator's biometric data to obtain spatial matching results.

8. An electronic device, characterized in that, include: The memory is configured to store instructions; as well as A processor is configured to retrieve the instructions from the memory and, when executing the instructions, to implement the intelligent early warning method for low-voltage electrical work safety based on multidimensional data according to any one of claims 1 to 7.

Images