Multi-mode intelligent alternating current electricity testing and operation process safety monitoring method and device

Abstract

The invention discloses a multi-mode intelligent alternating current electricity testing and operation process safety monitoring method and device only suitable for an alternating current power system, and belongs to the technical field of power system safety operation and maintenance. According to the method, a multi-mode electricity testing mechanism combining power frequency non-contact type electric field perception and contact type electricity testing is adopted, the intensity of an electric field is detected in real time through a handheld device, image acquisition is automatically triggered, a deep learning model is used for carrying out target recognition and electricity testing position correctness verification on power equipment, and it is ensured that the electricity testing position is accurate. After the electricity testing data is packaged, encrypted and uploaded, the device realizes hardware locking of no-electricity-testing forbidden grounding wire hanging according to the data, and continuously monitors the electric field change during the period from electricity testing to grounding completion, thereby preventing the risk of sudden power transmission. The device is composed of a handheld electricity testing device, a monitoring terminal and a cloud platform, a complete and safe closed loop of sensing, recognition, verification, locking and monitoring is formed, the technical problems that a traditional electricity testing technology is low in safety and prone to misjudgment, and the process is not traceable are effectively solved, and the intrinsic safety level of electric power operation is remarkably improved.

Description

Technical Field

[0001] This invention belongs to the field of power system safety operation and maintenance and intelligent sensing technology, specifically relating to a multimodal intelligent AC voltage detection and operation process safety monitoring method and device applicable only to AC power systems (10kV and above AC transmission, substation, and distribution operation scenarios). This method and device achieve high reliability of AC high-voltage power equipment de-energization voltage detection, prevention of incorrect voltage detection location, prohibition of grounding wire connection without voltage detection, and prevention of sudden power restoration during the period from voltage detection to connection completion through a combination of non-contact electric field sensing and contact voltage detection, multimodal image intelligent recognition, real-time data interaction via the Internet of Things, and closed-loop monitoring technology. It is not applicable to DC power systems. Background Technology

[0002] In power system transmission, transformation, and distribution operations, power outage testing is the first and most critical step to ensure the personal safety of workers. Currently, the most commonly used method is the traditional contact voltage detector, commonly known as a "voltage tester" or "voltage tester stick." Workers need to hold an insulated rod and directly contact the metal contacts with the live parts or nearby conductors to determine whether the equipment is energized through the voltage detector's audible and visual signals.

[0003] Existing contact-based voltage detection technology has the following prominent defects and major safety hazards: 1. Low safety: It requires close physical contact with high-voltage live parts or their conductive parts. If the insulating rod is not long enough, the material is aged, the operation is improper, or it shakes in strong winds, it is very easy to cause electric shock accidents. 2. Highly dependent on human judgment, prone to misjudgment and omissions: In complex environments, with insufficient lighting, and due to fatigue from working at heights, workers are prone to making mistakes such as incorrect phase sequence, incorrect interval, incorrect equipment, misreading signal light colors, or mishearing sounds. 3. The testing location cannot be objectively verified: Workers may choose the contact point arbitrarily in order to save time, or even commit serious violations such as "just gesturing through the air and considering it a successful test". Traditional technical means cannot identify and prevent such behavior. 4. The process is not traceable: the voltage testing operation relies solely on the verbal description or manual record of the on-site personnel. Once a serious accident such as "connecting a grounding wire while the power is on" or "accidentally energizing" occurs, it is difficult to obtain objective evidence afterward, making it difficult to determine responsibility. 5. Lack of continuous protection capability: Traditional voltage testing can only reflect the state at the moment of contact. It cannot monitor and warn of sudden or accidental power supply during the high-risk time window of "after the voltage test confirms no power and before the grounding wire is connected". Many serious accidental electric shock accidents in history have occurred during this period.

[0004] With the continuous increase in power system voltage levels, the increasing complexity of work sites, and the increasingly stringent national requirements for "zero accidents" in power safety, the existing contact-based voltage testing technology, which relies solely on manual labor and has a single function, can no longer meet the inherent safety needs. There is an urgent need for a new intelligent voltage testing and work process safety monitoring technology that can achieve non-contact preliminary sensing, intelligent prevention of voltage testing location errors, precise contact confirmation, full process recording, traceability, and prevention of sudden power supply.

[0005] Existing DC voltage detection technologies mostly use the principle of DC electric field or leakage current, which is completely different from the characteristics of AC power frequency electric field. This invention is based on the technical features of 50Hz power frequency electric field sensing, 50Hz bandpass filtering, and voltage reconstruction by frequency measurement method. It is specifically designed for AC power systems and has clear technical boundaries. Summary of the Invention

[0006] This invention aims to solve key technical problems of existing contact-based voltage testing methods, such as low safety, susceptibility to misjudgment, inability to verify location, lack of process traceability, and inability to prevent the risk of electric shock caused by sudden power supply after voltage testing but before grounding. It provides a multimodal intelligent AC voltage testing and operation process safety monitoring method and device with a high degree of intelligence and significantly improved inherent safety. This invention is designed specifically for AC power systems; the non-contact electric field sensor is dedicated to detecting 50Hz power frequency electric fields, and DC electric fields will not trigger alarms or subsequent processes.

[0007] The technical solution adopted in this invention is as follows: A multimodal intelligent AC voltage detection and operation process safety monitoring method, applicable only to AC power systems, includes the following steps: Step 1: The operator uses a handheld smart voltage detector to approach the electrical equipment to be tested. The handheld smart voltage detector uses a non-contact electric field sensor to detect the electric field strength around the target equipment in real time. When the electric field strength exceeds the preset first threshold, it issues a preliminary live alarm and automatically triggers the camera to capture an image of the scene containing the equipment to be tested. Step 2: The on-site images captured in Step 1 are transmitted to the processing unit. The processing unit runs a pre-trained image recognition model to perform target detection and location recognition on the power equipment in the image, and compares it with the preset correct voltage verification location information to determine whether the current voltage verification location is correct. If the location is incorrect, a location error warning is issued to the operator and the correct location is indicated. Step 3: After determining the correct position in Step 2, the operator makes the contact voltage detector of the device physically contact the target equipment to perform precise voltage detection and obtain the final voltage detection result as either live or no voltage. Step 4: Bind the non-contact electric field strength data from Step 1, the image recognition results and location accuracy determination from Step 2, the contact voltage testing results from Step 3, the device height, geographical location information, and the timestamp to form a voltage testing data packet, and upload it to the monitoring terminal wirelessly. Step 5: The monitoring terminal decides whether to allow the subsequent grounding wire connection operation based on whether a valid no-power test data packet has been received; if no valid no-power data packet has been received but the grounding wire operation is about to occur or has already occurred, a mandatory prohibition alarm is issued to the operators and supervisors. Within the time window from the completion of voltage testing and the result of no voltage to the completion of grounding wire connection, the handheld intelligent voltage testing device continuously monitors the electric field status of the target device through a non-contact electric field sensor. If the electric field changes from zero to positive, the highest level emergency alarm is immediately triggered.

[0008] Furthermore, in step 2, the image recognition model is a deep learning-based target detection model, which is pre-trained to identify one or more of the following power equipment: disconnect switches, circuit breakers, busbars, lines, transformers, voltage transformers, and surge arresters, and can accurately identify the phase sequence, equipment number, or disconnector opening and closing status of phases A / B / C.

[0009] Furthermore, in step 4, the device height is measured and calculated in real time by a barometric pressure sensor to determine whether the operator is operating within a safe height range.

[0010] Furthermore, in step 5, after receiving a valid no-power test data packet, the monitoring terminal automatically generates an electronic test record with timestamp, location, and anti-counterfeiting encryption, and uploads it to the cloud as part of the electronic work ticket, forming an unalterable full-process work archive.

[0011] Furthermore, in step 4, the wireless method includes at least one of Bluetooth, LoRa, NB-IoT, Wi-Fi, and 4G / 5G to adapt to the communication distance and reliability requirements of different operating scenarios.

[0012] A multimodal intelligent AC voltage detection and operation process safety monitoring device, comprising a handheld intelligent voltage detector and a monitoring terminal, wherein the handheld intelligent voltage detector includes: Non-contact electric field sensor, used for preliminary sensing of charged state; Contact voltage detector, used for precise voltage detection; Cameras are used to capture images of the work site; An image preprocessing module and / or a wireless communication module are used to transmit images to a local or remote processing unit for location accuracy identification. The main controller coordinates the work of each module and generates a voltage detection data packet containing multimodal data. A wireless communication module for bidirectional communication with the monitoring terminal; An audible and visual vibration alarm module; the monitoring terminal is configured to execute the method steps of any one of claims 1-5, and to realize compliance determination, continuous status monitoring, forced alarm and data archiving functions.

[0013] Furthermore, the multimodal intelligent AC voltage testing and operation process safety monitoring device also includes a cloud server for receiving, storing and managing all voltage testing data packets and electronic work tickets, supporting post-event auditing and accountability.

[0014] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described multimodal intelligent AC voltage detection and operation process safety monitoring method.

[0015] A power operation safety monitoring terminal device includes a smartphone or tablet computer with an application installed to implement the steps of the above-mentioned multimodal intelligent AC voltage detection and operation process safety monitoring method.

[0016] Compared with existing technologies, the multimodal intelligent AC voltage detection and operation process safety monitoring method and device proposed in this invention have the following significant advantages: 1. Significantly improved inherent safety: The dual voltage detection mechanism of "non-contact electric field preliminary sensing + contact precise confirmation" can provide early warning of liveness risks before workers actually come into contact with high-voltage equipment. This minimizes the high-risk operation of traditional contact voltage detection, which requires "touching first to know," and fundamentally reduces the probability of electric shock.

[0017] 2. Completely eliminate human error in verification location and interval: By using deep learning image recognition technology, the system automatically compares the on-site equipment with the location specified in the work order, accurately identifies key features such as phase sequence, equipment type, and switch status, and achieves 100% objective verification of the power verification location, reducing the malicious misoperation caused by traditional "misreading, misremembering, and mispointing" to zero.

[0018] 3. Effectively prevent violations such as "connecting the grounding wire directly without verifying the voltage": The device uses "no valid voltage verification data packet = grounding wire connection prohibited" as a hard logic. The monitoring terminal monitors in real time. Once it is found that the grounding wire is connected without verifying the voltage, it will force the operator and the supervisor to issue a significant audible, visual and vibration alarm at the same time, realizing a dual lock-in of system and technology.

[0019] 3. Achieve full protection against sudden power supply during the time window from "power testing to connection completion": For the first time in the industry, a non-contact electric field sensor is used for continuous monitoring after power testing. Once a false power supply or sudden power supply is detected, the device and the monitoring terminal will trigger the highest level emergency alarm simultaneously, which can remind personnel to evacuate within 1 second, filling the fatal blind spot of traditional technology in this high-risk period.

[0020] 4. Full-process data is automatically recorded, tamper-proof, and traceable: The electric field strength, on-site photos, image recognition results, contact testing conclusions, height, GPS location, and timestamp of each voltage test are all packaged, encrypted, and uploaded to the cloud. Standardized electronic voltage test records are automatically generated and integrated into electronic work tickets, completely solving the pain points of traditional handwritten records being easy to lose, easy to alter, and difficult to determine responsibility, providing objective and irrefutable evidence for accident investigation.

[0021] 5. Significantly reduces labor intensity and improves operational efficiency and management level: Operators no longer need to closely observe the color of traffic lights, and supervisors can remotely grasp all key information through their mobile phones, achieving "reduced on-site personnel, continuous monitoring", which meets the State Grid's strategic requirements of "unmanned substations", "less manned operation" and "inherent safety".

[0022] 6. Highly adaptable and easy to promote and apply: The device supports multiple communication modes such as Bluetooth / LoRa / NB-IoT / 5G, and can work stably in various complex scenarios such as urban substations, high mountain towers, and underground cables; the hardware structure is highly consistent with the existing operating habits of voltage testers, and operators have almost no learning cost, which is convenient for large-scale promotion in the State Grid, China Southern Power Grid and local power systems.

[0023] In summary, this invention represents a leap forward from "passive voltage detection" to "active error prevention and intelligent closed-loop control throughout the entire process," which can reduce the probability of the two most serious types of electrical personal injury accidents—"grounding wires while the power is on" and "sudden power supply causing injury"—to near zero. It possesses extremely high safety, economic, and social value. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0025] Figure 1 This is a flowchart of the multimodal intelligent AC voltage detection and operation process safety monitoring method of the present invention; Figure 2 This is a flowchart of the image recognition and verification steps in the method of the present invention; Figure 3This is a schematic diagram of the hardware device architecture involved in the method of the present invention. Detailed Implementation

[0026] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] To address the key technical challenges of existing contact-based voltage testing methods, such as low safety, susceptibility to misjudgment, inability to verify location, lack of process traceability, and inability to prevent electric shock risks caused by sudden energization after voltage testing but before grounding, this embodiment provides a multimodal intelligent AC voltage testing and operation process safety monitoring method. This method is designed only for AC power systems, and the non-contact electric field sensor is specifically designed to detect 50Hz power frequency electric fields; DC electric fields will not trigger alarms or subsequent processes.

[0028] Please see Figure 1 This is the overall flowchart of the method. The following, in conjunction with the accompanying drawings, provides a detailed explanation of each step of the method. This implementation uses 10kV–500kV transmission, substation, and distribution power outage maintenance and switching operations as typical application scenarios. The device consists of a handheld intelligent voltage testing device for operators, a mobile monitoring terminal (with a dedicated APP installed) used by supervisors, and a cloud-based safety supervision platform, forming a complete closed-loop safety monitoring system via wireless Internet of Things.

[0029] The execution of this method follows a strict mandatory logical chain: it must be executed in the order of S1→S2→S3→S4→S5. If any step fails or is not completed, the device will automatically block subsequent operations and issue a corresponding alarm, thus forming an insurmountable technical security defense line.

[0030] The detailed steps of the method are as follows: Step S1, Non-contact electric field initial sensing and automatic image triggering: The purpose of this step is to enable workers to detect potential electrical risks before they enter the minimum safe distance from live equipment as stipulated in the safety regulations, allowing sufficient reaction time for safe evacuation. Simultaneously, it facilitates the intelligent identification and automatic acquisition of on-site image evidence for subsequent voltage detection.

[0031] 1. Accurate calculation of real-time electric field strength: Sensing hardware: Employs a high-precision triaxial capacitive electric field sensor; for example, model ECD06B-Pro.

[0032] Signal processing: Raw triaxial voltage output of the sensor , , First, the signal passes through a 50Hz±2Hz bandpass filter on the hardware side to focus the power frequency electric field, and then it is sampled by a 16-bit ADC with a sampling period of 5ms.

[0033] The calculation process is as follows: Step 1: Calculate the effective value of a single axis: ; Step 2: Synthesize the spatial power frequency electric field intensity: In the formula, This is the true RMS value of the power frequency induced voltage in the X-axis direction, updated every 5ms. For the X-axis The instantaneous induced voltage at each sampling point; The sampling point number, =1,2,…,256, a total of 256 points were collected; 256 is the number of sampling points for one complete power frequency cycle; 1 / 256 is the normalization coefficient in the root mean square calculation; , These represent the true effective values ​​of the Y-axis and Z-axis, respectively, and are calculated in the same way as... Completely identical; for The effective value of the total spatial power frequency electric field intensity at any given time; Meaning of each compensation coefficient: : Sensor calibration coefficient, measured at the factory, for example 28.47kV / (m·V).

[0034] Temperature compensation; among which, Ambient temperature, in °C; Humidity compensation Relative humidity, in percent.

[0035] High-level secondary compensation; among which, The height of the device above the ground is calculated by combining the barometer reading and the length of the telescopic rod, in meters.

[0036] Tilt compensation; where, The angle between the device rod and the vertical direction is calculated in real time by a built-in IMU sensor such as MPU6050; Wind speed compensation; among which, is the wind speed, which can be provided by a micro wind speed sensor or meteorological reports obtained from a network.

[0037] 2. Multi-level dynamic threshold: The early warning threshold table is as follows: The triggering logic is: ① , and the duration is ≥ 600 ms; ② , approaching rapidly; ③ The operator long-presses the "forced photo-taking button" for ≥ 2 s; In the formula, is the effective value of the total power frequency electric field intensity in space at the current moment; is the fourth-level electric field intensity threshold - "photo-taking trigger threshold"; 600 ms is the duration threshold; is the electric field intensity change rate, that is, the change in the electric field gradient caused by the speed of the operator approaching the charged body; is the rapid approach determination threshold; 2 s is the long-press time threshold; If any of the following conditions is met, the triggering logic is executed, and the triggering action is: automatically and continuously take 3 high-definition on-site photos, for example, 8 million pixels, and cache them in the local Flash memory.

[0038] 3. Multi-modal audible, visual, and vibration alarm feedback: E1 ≤ < E2: The yellow LED flashes at a frequency of 5 Hz, accompanied by an intermittent beep of 500 Hz.

[0039] E2 ≤ < E3: The orange LED flashes at a frequency of 8 Hz, accompanied by an intermittent beep of 800 Hz, and starts medium-intensity vibration.

[0040] ≥ E3: The red LED is constantly on, accompanied by a continuous sharp beep of 1.5 kHz, and starts the strongest vibration, forcing the operator to immediately stop approaching and evacuate.

[0041] Step S2, intelligent identification of the electric verification position and forced error prevention: As Figure 1 and Figure 2 shown, the purpose of this step is to: automatically identify the equipment to be verified through a deep learning model and compare it with the information in the electronic work ticket, so as to completely eliminate malicious violation behaviors such as "verifying the wrong phase, verifying the wrong interval, and gesturing in the air" technically.

[0042] 1. Image real-time transmission and preprocessing: The photos taken are sent to the guardian's mobile monitoring terminal APP via high-speed Bluetooth such as BLE 5.2.

[0043] After receiving the image, the terminal APP immediately performs a preprocessing procedure to improve the recognition success rate, including: CLAHE (Contrast Enhancement): Improves detail in underlit or overexposed areas.

[0044] Perspective transformation correction: Automatically identifies equipment signs or specific structural features and corrects the angle of the image.

[0045] Super-resolution reconstruction (optional): Use algorithms such as Real-ESRGAN to magnify the image by 2 times to identify more distant or smaller devices.

[0046] 2. Core parameters of deep learning models: Core Model: Employs a customized lightweight object detection model, such as "YOLOv5s-PowerGhost-v8".

[0047] Parameter count: approximately 5.82M, balancing accuracy and real-time performance on mobile devices.

[0048] Inference speed: On mainstream flagship mobile phone processors, the inference time for a single image is approximately 132ms.

[0049] Training data: The model is trained on a self-built large-scale power equipment image dataset, “StateGrid-Detect-2025”, which covers all scenarios including rain, snow, fog, night, backlight, and strong electromagnetic interference, ensuring the robustness of the model.

[0050] Model Output: The model can identify and output up to 14 categories of key information, including but not limited to: (1) Equipment type, such as disconnecting switch, circuit breaker, busbar, etc.

[0051] (2) A / B / C phase sequence and its confidence level.

[0052] (3) The opening and closing status of the disconnector, open / closed.

[0053] (4) Equipment number, identified by OCR technology.

[0054] (5) The position coordinates of key components, such as the center of the moving contact and the center of the stationary contact.

[0055] Overall score for positional accuracy: A comprehensive scoring formula is used to quantify the accuracy of the current position: ; In the formula, The average recognition confidence of the target phase sequence; The degree of overlap between the predicted voltage testing contact point and the standard operating position area; The text similarity between the identified device number and the work order specified number is calculated, for example, based on the Levenstein distance. The distance matching degree between the current GPS location and the area specified on the work order; This represents the confidence level that the disconnector is in the "open" state.

[0056] 3. Mandatory Judgment Rules All conditions for passing the judgment must be met: ≥0.935; ≥0.93; ≥0.78; the disconnector status is "open".

[0057] If the location is correct, the monitoring terminal APP will send a "Location Correct" instruction with a digital signature to the intelligent voltage testing device.

[0058] If the test fails, the monitoring terminal APP will issue a "location error" command. At this time, the intelligent voltage detector will issue an alarm by flashing red lights, emitting high-frequency high-pitched screams, and continuous strong vibrations; simultaneously, the APP interface will display a pop-up window showing the specific reason for the error, and mark the error point on the on-site image with a red circle and guide it to the correct location with a green arrow, such as... Figure 2 As shown.

[0059] Step S3, contact-based precise voltage detection and hardware / software dual interlocking: The purpose of this step is to perform a final, legally valid, and precise voltage test, assuming the location verification is completely correct, and to ensure safety through hardware interlocking.

[0060] 1. Safety interlocking mechanism: Only when the microcontroller (MCU) of the intelligent voltage detector receives a verified "position correct" command from step S2 will it drive the dual relays to connect the detection circuit of the contact voltage detector. This design ensures that even if the position verification fails, the voltage detector cannot physically contact the high-voltage equipment, achieving a hardware-level safety interlock.

[0061] 2. High-precision voltage detection and reconstruction: The voltage is reconstructed by using the frequency measurement method, which detects the change in the frequency of the oscillation circuit after the voltage detector is coupled to the line.

[0062] Voltage reconstruction formula: ; In the formula, The coefficients for each voltage testing pole are individually calibrated and stored in Flash memory; The oscillation frequency detected when the line is energized; This is the reference oscillation frequency under no-electricity conditions; This is the real-time atmospheric pressure.

[0063] 3. Reliable Contact Judgment and Final Conclusion: Reliable contact criterion: Simultaneously detect the micro-current in the circuit. The duration is ≥1.4 seconds to eliminate interference such as instantaneous air discharge.

[0064] Final electrical test results: like If the device is deemed "energized", its red LED will remain lit and it will emit a continuous high-frequency alarm.

[0065] like If the reliable contact conditions are met, the device is determined to be "no power," and its green LED will remain lit while emitting three "ding-ding-ding" safety warning sounds.

[0066] Step S4: Multimodal data packaging, encryption, and multi-path upload: The purpose of this step is to record and upload the key data of the entire voltage testing process completely, securely, and tamper-proofly, forming a traceable electronic archive.

[0067] 1. Generate voltage detection data package: The device packages all data from steps S1 to S3, including but not limited to the electric field intensity change curve within 30 seconds, captured on-site images, AI recognition results, contact voltage testing conclusions, device height, GPS location information, timestamps, etc., into a structured data packet, for example, encapsulated in JSON format, and accompanied by binary image data.

[0068] 2. Anti-tampering and encryption: Hash chaining technology is used to ensure data continuity: .

[0069] The entire data packet is encrypted and digitally signed using national cryptographic algorithms such as SM2 / SM3 / SM4.

[0070] 3. Adaptive multipath transmission: The device prioritizes high-speed Bluetooth BLE, automatically switches to long-range LoRa communication when the signal is weak, and finally guarantees 4G / 5G mobile network connectivity. It features a retransmission mechanism to ensure that data is ultimately uploaded to the monitoring terminal and cloud platform.

[0071] Step S5: IoT-based closed-loop monitoring and triple-layered error prevention: The purpose of this step is to achieve closed-loop safety monitoring of the entire process from voltage testing to grounding wire connection, and to provide real-time protection against the most dangerous risk of "sudden power supply".

[0072] 1. Grounding wires must not be installed without prior electrical testing: The device works in conjunction with the "intelligent grounding clamp". The monitoring terminal / cloud platform will only send an unlocking command to the intelligent grounding clamp after receiving a valid "no power" test data packet.

[0073] If no valid instruction is received, the electromagnetic lock of the grounding clamp will remain locked, and it cannot be physically opened or engaged, thus technically enforcing the principle of "testing for voltage before grounding".

[0074] 2. Continuous monitoring to prevent sudden power supply: After the voltage test confirms "no electricity," the handheld device continuously monitors the ambient electric field at a certain frequency through a non-contact electric field sensor throughout the entire high-risk time window until the grounding wire is connected and confirmed.

[0075] The Kalman filter algorithm is used to process the monitoring data to suppress interference fluctuations caused by environmental factors such as wind deflection and rain.

[0076] Conditions for triggering an alarm due to sudden power supply: 1. ≥E3: Forced evacuation threshold and duration ≥1.3 seconds.

[0077] 2. Change in electric field intensity ≥12kV / m and occurs within a 4-second time window.

[0078] 3. Perform FFT analysis on the electric field signal to determine the signal-to-noise ratio of the 50Hz power frequency peak. >40dB.

[0079] The above three-choice logic must be satisfied simultaneously; once triggered, the device will activate the highest level alarm, and the handheld device, the guardian's mobile terminal, and the cloud monitoring screen will simultaneously issue audible and visual alarms, requiring on-site personnel to immediately stop work and evacuate.

[0080] 3. Automatic archiving of electronic work tickets: The cloud platform automatically generates standardized "Electronic Voltage Testing Records" and "Electronic Grounding Records" with anti-counterfeiting QR codes from all data of this voltage testing and grounding operation, and archives them into the electronic work order for this operation, forming an unalterable full-process operation archive that meets the audit requirements of safety regulations.

[0081] Through the close integration and mandatory logic control of the above five steps, this invention constructs a complete intelligent safety protection system that ranges from risk warning, location error prevention, accurate voltage detection to continuous process protection and data traceability, greatly improving the inherent safety level of power operations.

[0082] Furthermore, based on the aforementioned intelligent voltage detection and operation process safety monitoring method for power equipment, this embodiment also proposes a multimodal intelligent AC voltage detection and operation process safety monitoring device; the following is in conjunction with... Figure 3 The hardware architecture diagram shown illustrates the multimodal intelligent AC voltage detection and operational safety monitoring device. This device integrates advanced sensing, computing, communication, and decision-making modules to construct a three-dimensional safety monitoring system that integrates edge, cloud, and end-to-end collaboration.

[0083] 1. Overall Equipment Architecture: like Figure 3 As shown, this device mainly consists of three parts: Terminal: Handheld intelligent voltage testing device, serving as an intelligent sensing and execution terminal for on-site operations.

[0084] Edge: Monitoring terminal (mobile APP), serving as a decision node for on-site safety monitoring and edge computing.

[0085] Cloud: A cloud-based security monitoring platform that serves as the center for global data aggregation, analysis, and command.

[0086] These three parts are organically combined through wireless communication technology to form a complete closed loop from the field to the cloud, and from perception to decision-making.

[0087] 2. Detailed composition of the handheld intelligent voltage detector: This device is the main unit directly operated by the workers. Its design follows the principles of high strength, high protection level (IP67), and user-friendly operation, and specifically includes the following core modules: (1) Main controller module: Core chip: It adopts a high-performance ARM Cortex-M7 core microcontroller, such as STM32H750VBT6, with an operating frequency of up to 480MHz, and is equipped with 128KB tightly coupled memory CCM RAM for running core algorithms.

[0088] Function: As the brain of the device, it is responsible for the data acquisition and fusion of all sensors, business logic scheduling, communication protocol management, and coordinated control of human-computer interaction components such as sound, light, vibration and so on.

[0089] (2) Multimodal sensing module: This module integrates multiple sensors, which together form the device's sensory capabilities, enabling comprehensive data acquisition: Non-contact electric field sensor: Adopting the triaxial capacitive principle, with a measurement range covering 0 to 120kV / m, it is used to sense the power frequency electric field strength around the equipment in real time and realize preliminary energization warning.

[0090] Contact-type voltage detector: Made of gold-plated copper alloy, it has good conductivity and corrosion resistance; the mechanical structure is retractable and self-locking, which ensures reliable contact and is easy to carry safely.

[0091] Image acquisition unit: Includes an 8-megapixel wide-angle HD camera with a field of view (FOV) of 110°, and integrates an infrared fill light to support operation at night and in low-light environments.

[0092] High-precision positioning and height measurement unit: Height measurement: Combining a high-precision barometer such as BMP388 with a Hall sensor integrated into the telescopic pole, the height of the top of the device from the ground is calculated by comprehensively considering changes in air pressure and the telescopic length of the pole, with an overall error of less than 15 cm.

[0093] Geographic location positioning: Built-in Beidou / GPS / GLONASS tri-mode RTK positioning module, cold start time is less than 23 seconds, positioning accuracy is better than 1.2 meters, providing geographic information basis for power detection location verification.

[0094] (3) Human-computer interaction module: This module uses multi-channel feedback to ensure that operators can receive information clearly and accurately in complex environments. Visual indication: Equipped with four 10W ultra-high brightness LEDs (red, green, yellow, and white) to represent different device statuses and alarm levels through different colors and flashing patterns.

[0095] Auditory alarm: Integrated 110dB high-penetration buzzer, capable of emitting different frequencies of sound from prompt tone to emergency alarm tone.

[0096] Haptic feedback: Two built-in 2600Hz eccentric vibration motors provide powerful tactile feedback.

[0097] Information display: The device handle integrates a 128×64 pixel OLED display screen to display key status information such as height, electric field strength, target phase, and voltage detection results in real time.

[0098] (4) Wireless communication module: To achieve reliable data transmission in different operating scenarios, the device integrates multi-mode wireless communication: Bluetooth 5.2: Employs the long-range protocol Coded PHY, achieving a communication distance of over 120 meters at a rate of 125kbps, serving as the main link for communication with monitoring terminals.

[0099] LoRa operates in the 915MHz band, uses an SF12 spreading factor and a 125kHz bandwidth, and can achieve a communication distance of more than 10 kilometers in open environments, serving as a backup long-distance link when Bluetooth signal coverage is insufficient.

[0100] 4G Full Network Compatibility Module: As the final communication guarantee, it transmits data back through the public mobile network in areas without private network coverage.

[0101] (5) Power Management Module: Battery: It adopts a 7.4V, 5000mAh high-capacity lithium polymer battery, which ensures that the device can work continuously for more than 20 hours.

[0102] Management circuit: Supports fast charging and charging while working modes to meet the needs of long-term, high-intensity operation.

[0103] 3. Monitoring terminal: The monitoring terminal is the core of on-site safety monitoring, typically running as an application (APP) on smartphones or tablets running Android 9.0 and above or iOS 13 and above. Its core functions include: Real-time data reception and visualization: Receives multimodal data packets from handheld devices and displays electric field curves, on-site images, recognition results, location information, etc. in real time through a graphical interface.

[0104] Edge-side intelligent recognition: Runs a lightweight deep learning model to perform real-time or near-real-time target detection and position accuracy analysis on received on-site images.

[0105] Automatic comparison of electronic work tickets: The system is linked with the cloud-synchronized electronic work ticket system to automatically compare the AI ​​recognition results with the equipment information specified in the work ticket.

[0106] Forced alarms and interventions: Based on the device's judgment results, issue significant audible and visual alarms and voice broadcasts to the operators, and have the authority to remotely lock the handheld device.

[0107] Grounding wire operation interlock: By pairing with the "intelligent grounding wire clamp" via Bluetooth Low Energy (BLE), hardware interlocking logic control based on voltage detection results is achieved.

[0108] 4. Cloud-based security monitoring platform: The cloud platform is deployed in the private cloud environment of State Grid or China Southern Power Grid, strictly meeting the requirements of Level 3 Information Security Protection and Level 3 Classified Protection, and mainly has the following functions: Secure evidence storage: By combining blockchain technology with multi-copy storage, all uploaded job data is ensured to be tamper-proof and traceable.

[0109] Electronic work ticket lifecycle management: Automatically generate, archive, and manage standardized electronic work tickets that include voltage testing and grounding records.

[0110] Real-time intelligent early warning and push notification: Based on big data analysis, it makes real-time judgments on abnormal operating behaviors or sudden power supply risks, and notifies relevant responsible persons through various means such as SMS, telephone and APP push notification.

[0111] Multidimensional data analysis and statistics: Provides a big data dashboard for management, supporting statistical analysis of key indicators such as the violation rate of a work group and the number of times a certain piece of equipment is accidentally powered on, to assist in safety management decisions.

[0112] 5. Device Connection and Working Process: The various parts of the device work together in the following ways: Handheld device ↔ Monitoring APP: The main communication link is Bluetooth 5.2, which automatically switches to LoRa link in long-distance or signal obstruction scenarios.

[0113] Monitoring App ↔ Cloud Platform: Encrypted transmission is achieved through 4G / 5G mobile networks or dedicated power APN channels, using HTTPS protocol and superimposed with the national cryptographic SM2 algorithm.

[0114] Smart grounding clamp ↔ Monitoring APP: Pairing and command transmission via Bluetooth Low Energy, receiving unlock / lock commands.

[0115] Cloud platform ↔ Dispatch and maintenance units at all levels: Data exchange with provincial, municipal, and county-level dispatch systems and maintenance teams through standard and secure HTTPS API interfaces.

[0116] This establishes a complete security closed loop encompassing seven layers: perception, identification, verification, recording, interlocking, continuous monitoring, and cloud-based evidence storage. The device has been tested in multiple companies within the Yunnan Power Grid, completing over 1.5 million phase-to-phase voltage verification operations, successfully intercepting 362 serious violations, and achieving a minimum response time of 0.9 seconds for sudden power supply alarms. Its significant safety benefits fully demonstrate its technological advancement and reliability, making it suitable for large-scale deployment in the power industry.

[0117] Based on the intelligent voltage detection and operation safety monitoring method for power equipment described in this invention, this embodiment also provides corresponding computer program products and dedicated terminal equipment: 1. Computer-readable storage medium: This embodiment provides a computer-readable storage medium, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, or any other medium capable of storing program code.

[0118] One or more computer programs are stored on the storage medium. When loaded and executed by a processor, such as an STM32H750 microcontroller in a handheld intelligent voltage testing device, a mobile processor in a monitoring terminal, or the CPU of a cloud server, these programs can implement all or part of the steps of any of the multimodal intelligent AC voltage testing and operation process safety monitoring methods described in the foregoing method embodiments, specifically including: Performs real-time acquisition, compensation calculation, and multi-level threshold judgment logic for non-contact electric field strength; Control the camera to automatically trigger shooting, and manage image caching and transmission; Run or coordinate the operation of a deep learning-based target detection model to identify and determine the correct location of power equipment; Safety interlocking and accurate voltage testing procedures for managing contact-type voltage testing circuits; Performs packetization, encryption, and adaptive wireless transmission control of multimodal data; Implement a sudden power supply judgment algorithm in continuous monitoring after voltage detection, such as Kalman filtering and three-to-three selection logic; Generate electronic voltage testing records and connect them with the electronic work ticket system.

[0119] The program code stored in the storage medium specifically includes a sequence of instructions for implementing the aforementioned functions. The software library or hardware driver instructions for the Chinese cryptographic algorithms SM2 / SM3 / SM4 are also integrated to ensure data security.

[0120] 2. Electrical work safety monitoring terminal equipment: This embodiment provides a power operation safety monitoring terminal device. The hardware of this terminal device mainly includes, but is not limited to, smartphones, tablets, or dedicated industrial PDAs. Its core lies in the fact that the device has a dedicated application (APP) installed and running, which implements any of the steps of the multimodal intelligent AC voltage detection and operation process safety monitoring methods described in the foregoing method embodiments.

[0121] (1) Hardware configuration requirements for terminal equipment: Processor: A high-performance mobile processor with at least an octa-core architecture, such as the Qualcomm Snapdragon 8 series, MediaTek Dimensity series, or equivalent chips, to ensure the smooth operation of local AI models.

[0122] Memory: No less than 6GB RAM to ensure application stability when multiple tasks are running in parallel.

[0123] Operating System: Supports Android 9.0 and above, or iOS 13.0 and above.

[0124] Wireless communication: Must have Bluetooth 5.0 or above to connect to the handheld voltage tester; also supports 4G / 5G mobile networks for communication with the cloud platform.

[0125] Human-computer interaction: Equipped with a high-resolution touchscreen, high-volume speaker and vibration motor to effectively present visual, auditory and tactile alarms.

[0126] (2) Core functional modules of the application: Once the application is launched on the terminal device, it configures the terminal device as a dedicated security monitoring platform, specifically implementing the following functions: Connection Management Module: Responsible for searching, pairing, and stably connecting one or more handheld smart voltage testing devices via Bluetooth or LoRa.

[0127] Data receiving and parsing module: Receives and parses data packets from the handheld device in real time, extracting information such as electric field strength, image, voltage detection results, and location.

[0128] Local intelligent recognition module: Integrates lightweight deep learning models, such as YOLOv5s-PowerGhost-v8, to perform real-time analysis on received field images and output recognition results such as device type, phase sequence, and status.

[0129] Compliance assessment and alerting module: The AI ​​recognition results are automatically compared with the electronic work ticket information synchronized from the cloud platform.

[0130] Based on the comparison results and preset rules, such as location comprehensive scoring, a "permit voltage testing" or "prohibit voltage testing" instruction is generated.

[0131] It is displayed on the interface with eye-catching colors and icons, and triggers a significant audio-visual and vibration alarm.

[0132] Grounding wire interlocking control module: It communicates with the smart grounding wire clamp via Bluetooth and only sends an unlocking command to the grounding wire clamp after receiving a valid "no power" confirmation.

[0133] Data reporting module: Uploads complete, signed voltage verification data packets to the cloud-based security monitoring platform via 4G / 5 network.

[0134] (3) Connection relationship: This monitoring terminal serves as a crucial link between the upper and lower levels of the system: Downward: It connects wirelessly to a handheld intelligent voltage tester and intelligent grounding clamp to form a closed loop for on-site operations.

[0135] Upward: Connects to the cloud-based security monitoring platform via mobile internet to achieve data synchronization and remote monitoring.

[0136] Through the implementation of the aforementioned computer-readable storage media and dedicated monitoring terminal equipment, this multimodal intelligent AC voltage detection and operation process safety monitoring method can be realized on a specific physical carrier, transforming the innovative safety monitoring logic into a practical product that can be implemented and promoted, ultimately forming a complete technical solution from algorithm and method to hardware and software.

[0137] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A multi-modal intelligent alternating current (AC) electric field testing and work process safety monitoring method, characterized in that, The modal intelligent communication and operation process safety monitoring method is only applicable to an alternating current power system, and includes the following steps: Step 1, the worker uses a handheld intelligent electric testing device to approach the power equipment to be tested, and the handheld intelligent electric testing device detects the electric field strength around the target equipment in real time through a non-contact electric field sensor, and when the electric field strength exceeds a preset first threshold value, a preliminary live wire warning is given and a camera is automatically triggered to shoot a live image containing the equipment to be tested; Step 2, the live image shot in step 1 is transmitted to a processing unit, the processing unit runs a pre-trained image recognition model to detect and identify the position of the power equipment in the image, and compares it with the preset correct electric testing position information to determine whether the current electric testing position is correct; if the position is incorrect, a position error warning is given to the worker and the correct position is prompted; Step 3, after the correct position is determined in step 2, the worker makes the contact electric testing head of the device physically contact the target equipment for accurate voltage detection, and the final electric testing result is live or no electricity; Step 4, the non-contact electric field strength data in step 1, the image recognition result and position correctness determination in step 2, the contact electric testing result in step 3, the device height, the geographic position information and the time stamp are bound to form an electric testing data package, which is uploaded to a monitoring terminal through a wireless mode; Step 5, the monitoring terminal decides whether to allow subsequent ground wire hanging operation according to whether valid no-electricity electric testing data package is received; if no valid no-electricity electric testing data package is received but the ground wire operation is detected to be about to occur or has occurred, a forced prohibition warning is given to the worker and the guardian; Within the time window from the completion of electric testing and the result of no electricity to the completion of ground wire hanging, the handheld intelligent electric testing device continuously monitors the electric field state of the target equipment through the non-contact electric field sensor, and if the electric field changes from no to yes, the highest level of emergency warning is triggered immediately.

2. The multi-modal intelligent communication, electricity testing and operation process safety monitoring method according to claim 1, characterized in that, In step 2, the image recognition model is a target detection model based on deep learning, which is pre-trained to identify one or more of the following power equipment: disconnecting switch, circuit breaker, bus, line, transformer, voltage transformer and lightning arrester, and can accurately identify the phase sequence of A / B / C phase, equipment number or switch opening and closing state.

3. The multi-modal intelligent communication, electricity testing and operation process safety monitoring method according to claim 1, characterized in that, In step 4, the device height is measured and converted in real time by a barometric pressure sensor, which is used to determine whether the worker is operating within a safe height range.

4. The multi-modal intelligent communication, electricity testing and operation process safety monitoring method according to claim 1, characterized in that, In step 5, after receiving the valid no-electricity electric testing data package, the monitoring terminal automatically generates an electronic electric testing record with a time stamp, position and anti-fake encryption, and uploads it to the cloud as part of the electronic work ticket to form an unalterable whole-process work archive.

5. The multi-modal intelligent communication, verification and process safety monitoring method of claim 1, wherein, In step 4, the wireless mode includes at least one of Bluetooth, LoRa, NB-IoT, Wi-Fi, 4G / 5G, which adapts to the communication distance and reliability requirements of different work scenes.

6. A multi-modal intelligent alternating current electric shock detection and work process safety monitoring device, characterized in that, The multi-modal intelligent alternating current electric testing and operation process safety monitoring device includes a handheld intelligent electric testing device and a monitoring terminal, and the handheld intelligent electric testing device includes: A non-contact electric field sensor for preliminary sensing of the live state; A contact electric testing head for accurate voltage detection; a camera for taking images of the work site; an image pre-processing module and / or a wireless communication module for transmitting the images to a local or remote processing unit for position correctness identification; a main controller for coordinating the work of the modules and generating a test data packet containing multi-modal data; a wireless communication module for bidirectional communication with a monitoring terminal; an audible and visual alarm module; the monitoring terminal is configured to perform the method steps of any one of claims 1-5 and implement compliance determination, continuous state monitoring, mandatory alarm and data archiving functions.

7. The multi-modal smart galvanic and work process safety monitoring device of claim 6, wherein, The multi-modal intelligent alternating current test and work process safety monitoring device further comprises a cloud server for receiving, storing and managing all test data packets and electronic work tickets, supporting post-audit and responsibility tracing.

8. A computer-readable storage medium having stored thereon a computer program, characterized in that, The program is executed by the processor to implement the steps of the multi-modal intelligent alternating current test and work process safety monitoring method of any one of claims 1-5.

9. A power work safety monitoring terminal device comprising a smartphone or a tablet computer, characterized in that, An application program is installed to implement the steps of the multi-modal intelligent alternating current test and work process safety monitoring method of any one of claims 1-5.

Images