High-voltage test field personnel intrusion real-time monitoring and warning system and working process

Abstract

The invention discloses a high-voltage test field personnel intrusion real-time monitoring and warning system and a working process, and relates to the technical field of high-voltage safety protection. Comprising an intrusion detection module, a pulse warning module, a monitoring feedback module, a power supply control module and a central control unit. Progressive response logic is adopted: the system firstly confirms intrusion through double-sensor cross validation, then triggers a low-voltage electric field pulse sequence to carry out noninvasive sensory expelling, and synchronously starts multi-mode interaction auxiliary warning; and then the system verifies the personnel response effect, and only when the expelling is invalid, the system is upgraded to cut off the power supply and the physical isolation. In addition, the system is integrated with an environment self-adaptive module, and pulse parameters are dynamically adjusted according to temperature and humidity to improve reliability. And parameter iterative optimization is carried out based on historical data in an event recovery stage to form a self-learning closed loop. The optimal balance between safety protection and test continuity is realized, the warning mode is universal and effective, and the system operation is intelligent and reliable.

Description

Technical Field

[0001] This invention relates to the field of high-voltage safety protection technology, and in particular to a real-time monitoring and alarm system for personnel intrusion in high-voltage test sites and its workflow. Background Technology

[0002] High-voltage testing environments (such as substations and power laboratories) pose a deadly risk of electric shock, making the reliability of personnel intrusion monitoring extremely important. Existing technologies primarily rely on a core architecture of "basic monitoring + audible and visual alarms + rigid protection," resulting in limited functionality and poor adaptability. Alarm systems generally depend on fixed audible and visual modules, using sound signals and flashing lights to issue warnings. The core technology relies on static intrusion detection techniques, such as infrared beams and microwave detectors, to identify risks. In field applications, these systems suffer from the following main drawbacks: First, mainstream audible and visual alarm systems have security blind spots. For example, existing static intrusion detection technology is ineffective in noisy or brightly lit environments and is completely ineffective for hearing or visually impaired individuals, posing a significant security risk.

[0003] Secondly, some systems suffer from "over-protection." For example, they tend to directly cut off the test power supply after an alarm. While this "one-size-fits-all" strategy ensures safety, it frequently interrupts important high-voltage tests, causing huge economic losses and research delays.

[0004] Furthermore, existing systems generally lack sufficient environmental adaptability. Most systems use fixed sensing thresholds, making them unable to adapt to complex environmental changes such as rain, fog, and high humidity. This easily leads to false alarms (false positives) or missed alarms, and the system reliability is insufficient to meet the stringent requirements of industrial sites. Therefore, how to achieve an effective tiered warning system that intelligently adapts to the environment and maximizes the continuity of testing while ensuring safety is the urgent technical problem that needs to be solved in this case. Summary of the Invention

[0005] To address the above problems, this invention provides a real-time monitoring and alarm system and workflow for personnel intrusion in high-voltage test sites, which features effective graded warnings, intelligent environmental adaptation, and maximizes test continuity while ensuring safety.

[0006] The technical solution of this invention is: A real-time monitoring and alarm system for personnel intrusion in a high-voltage test field includes: An intrusion detection module, deployed at the site boundary, is used to monitor the physical boundary in real time and generate intrusion warning signals; The pulse warning module is connected to the central control unit and the field boundary conductor. After receiving the intrusion warning signal, it generates a low-voltage electric field pulse sequence and outputs it through the field boundary conductor. The monitoring and feedback module is connected to the central control unit and is used to collect response data after pulse output and determine whether the response is valid based on a preset confidence threshold of cross-validation. The power control module is connected to the central control unit. When the response is invalid, it cuts off the field power and activates physical isolation. Together with the central control unit, it coordinates the work of each module and performs system self-tests and event recovery.

[0007] Specifically, the intrusion detection module includes: Infrared sensors are used to detect and monitor changes in infrared radiation in physical boundary areas; The ultrasonic sensor array, according to its layout, periodically scans the physical boundary area at fixed time intervals of 0.3-0.5 seconds, and obtains target spatial information by emitting ultrasonic signals and receiving reflected echoes. Specifically, the pulse warning module includes: The IGBT driving circuit is connected to the central control unit and is used to receive the pulse width modulation signal from the central control unit, generate a low-voltage electric field pulse sequence, and the generated pulse is coupled to the boundary conductor of the field through a safety isolation transformer. The Hall current sensor communicates with the central control unit to monitor the pulse current in real time, ensuring that it is always below the safety threshold of 1 mA.

[0008] Specifically, the monitoring feedback module includes: The current detection unit communicates with the central control unit. The current detection unit can be implemented using common Hall current sensors such as Allegro's ACS712.

[0009] Specifically, the power control module includes a solid-state relay array connected to the central control unit, used to quickly cut off the field power supply after receiving an "isolation upgrade signal".

[0010] Specifically, it also includes an environment adaptive threshold module connected to the central control unit; the environment adaptive threshold module includes: The temperature and humidity sensor is connected to the central control unit to collect the ambient relative humidity (RH) and temperature in real time, providing a basis for the central control unit to adjust the pulse parameters; A photosensitive sensor, connected to the central control unit, is used to collect ambient illuminance in real time, providing a basis for the central control unit to adjust pulse parameters; Specifically, it also includes a multimodal interaction layer connected to the central control unit; the multimodal interaction layer includes: Directional speakers, placed at the physical boundaries of the area and connected to the central control unit, are used to play pre-recorded voice warnings in a specific direction; Vibration sensors, embedded in the boundary ground in the form of vibration pads, are connected to the central control unit to generate tactile feedback with adjustable intensity. The Bluetooth communication module connects to the central control unit and is used for short-range wireless interaction with the safety helmets of workers that are integrated with RFID tags, sending coded commands. A workflow for a real-time monitoring and alarm system for personnel intrusion in a high-voltage test field includes the following steps: Step S1: Environmental perception and intrusion pre-detection; After the system starts up, the intrusion detection module continuously scans the boundary of the field. When a potential target is initially detected, the central control unit does not immediately issue an alarm, but instead activates a dual-sensor cross-validation algorithm to effectively distinguish whether it is a real person breaking in and reduce the false positive rate. Specifically, the dual-sensor cross-validation algorithm is built into the central control unit, and its execution steps are as follows: S1.1: Data Acquisition and Signal Preprocessing The central control unit reads the raw signals from the infrared sensor and the ultrasonic sensor in parallel through a serial communication interface; The infrared sensor signal is converted into an infrared confidence value ranging from 0% to 100%, and the calculation formula is as follows: Infrared confidence level = [(Detection temperature - Ambient baseline temperature) / (Human body temperature threshold - Ambient baseline temperature)] * 100% The ultrasonic sensor signal is converted into an ultrasonic confidence value ranging from 0% to 100%, which is calculated as follows: Ultrasonic confidence score = min[100%, (threshold distance - actual distance) / threshold distance * 100%] * motion weighting factor in: The threshold distance is preferably 1m; The motion weighting factor is defined in segments based on the real-time motion velocity v (in m / s) of the monitored object: If the real-time motion speed v > 0.2 m / s, then the motion weighting factor is set to 1.2 (to strengthen the confidence weight of dynamic intrusion). If the real-time motion speed v≤0.2m / s, then the motion weighting factor is set to 1.0 (to avoid excessive amplification of static interference). If the confidence value of any of the above sensors is lower than 20%, it will be directly determined as invalid interference, and the verification will end. S1.2 Cross-validation and fusion computation: The central control unit fuses the two confidence levels mentioned above to calculate the overall confidence level, using the following formula: Overall confidence level = (Infrared confidence level * Ultrasonic confidence level) / 100; S1.3 Threshold Judgment and Decision Output: The central control unit compares the calculated overall confidence level with a preset 80% confidence threshold: If the overall confidence level is >80%, it is determined as "verification passed", and an "intrusion warning signal" containing a timestamp is generated and the process is advanced to the pulse warning step; If the overall confidence level is ≤80%, it is judged as "verification failed", and the system can trigger a resampling for a second verification. If the verification fails multiple times in a row, the system can enter a low-alert mode to shorten the scanning cycle and enhance monitoring.

[0011] Step S2: Pulse Warning and Multimodal Output: Upon receiving an intrusion warning signal, the central control unit immediately triggers the pulse warning module. The pulse warning module generates a sequence of three pulses with progressively increasing voltages, the highest of which does not exceed the internationally recognized low-voltage safety threshold of 50V. Each pulse is 30 milliseconds wide and spaced 100 milliseconds apart. This sequence is applied to the boundary conductor through an isolation transformer, producing a tactile sensation sufficient to attract attention while ensuring safety.

[0012] Step S3: Response Verification and Threshold Determination After the pulse sequence ends, the system opens a response verification window to observe changes in personnel behavior and ensure the effectiveness of the expulsion. The monitoring feedback module collects two types of data in this window: One is the rate of change of boundary motion obtained by measuring it again using an ultrasonic sensor; Second, multimodal confirmation signals, such as confirmation receipts received by the Bluetooth module; The central control unit performs a weighted evaluation based on the boundary motion change rate and multimodal confirmation signals to obtain the overall response rate; the calculation formula is as follows: Response rate = (Boundary motion change rate + Multimodal confirmation signal) / 2 * 100%; If the response rate is greater than 85%, the response is deemed valid, the system generates a "safety recovery signal" and returns to the initial monitoring state; If the response rate does not reach this threshold (for example, the measured value is 70%), the response is deemed invalid and an "isolation upgrade signal" is generated. Among them, the multimodal confirmation signal comes from the response feedback data after the interaction layer is activated, and is obtained by independently scoring and normalizing each modal feedback; The calculation formula for the adaptive dynamic pre-adjustment of the pulse warning parameter in step S1.3 is as follows: V adapt =V base * (1 - 0.2 * RH / 100), Among them, V adapt The adjusted voltage value; V base This is a reference voltage value (e.g., 40V, which is an optimized value under standard ambient RH < 60%). RH represents the percentage of relative humidity collected in real time. A coefficient of 0.2 indicates that for every 10% increase in humidity, the voltage is reduced by 20%, which is the approximate voltage drop required to maintain a safe current.

[0013] Step S4: Power off and field isolation: Once the "isolation upgrade signal" is received, the power control module sequentially cuts off the auxiliary power supply and the main high-voltage power supply of the field; at the same time, it activates physical isolation devices such as electromagnetic door locks, sets the field status to "safe mode", and broadcasts a pause signal.

[0014] Step S5: Event Recovery and System Self-Check After maintaining the isolation of the field for a configurable period of time, the system confirms that personnel have retreated to a safe distance and that the power supply is stable. Subsequently, the central control unit performs a system self-test, which includes sensor calibration, leakage current testing of the pulse module, and verification of multimodal connectivity. After the self-test passes, the system restores power in sequence.

[0015] This invention includes intrusion detection, pulse warning, monitoring feedback, power control, and a central control unit. It employs a progressive response logic: the system first confirms intrusion through cross-verification using dual sensors, then triggers a low-voltage electric field pulse sequence for harmless sensory deterrence, simultaneously initiating a multimodal interactive auxiliary warning; subsequently, the system verifies the effectiveness of the personnel's response, escalating to power cut-off and physical isolation only if deterrence proves ineffective. Furthermore, the system integrates an environmental adaptive module, dynamically adjusting pulse parameters based on temperature and humidity to improve reliability; and iteratively optimizing parameters based on historical data during the event recovery phase, forming a self-learning closed loop. This invention achieves an optimal balance between security protection and test continuity, with universally effective warning methods and intelligent and reliable system operation. Attached Figure Description

[0016] Figure 1 This is a diagram illustrating the architecture of the real-time monitoring and alarm system for personnel intrusion in the high-voltage test field according to the present invention. Figure 2 This is a flowchart of the real-time monitoring and alarm system for personnel intrusion in high-voltage test fields according to the present invention. Detailed Implementation

[0017] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0018] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0019] Referring to Figures 1 and 2 below, according to an embodiment of the present invention, the system is applicable to high-voltage environments such as 500kV substations and can achieve safe and reliable personnel intrusion management under complex weather conditions.

[0020] The real-time monitoring and alarm system for personnel intrusion in high-voltage test sites mainly includes: Central control unit: The central control unit connects to all modules, coordinates their operation, and performs system self-tests and event recovery. The central control unit includes a microprocessor, such as an ARM Cortex-M series chip. It is responsible for running state machine logic, processing sensor data, and coordinating the actions of each module; this unit is connected to non-volatile memory for storing system logs and historical adaptive parameter data.

[0021] The intrusion detection module includes: Infrared sensors are used to detect changes in infrared radiation in physical boundary areas. When a heat-generating target such as a human or animal enters or moves, an initial sensing signal is triggered. The ultrasonic sensor array, arranged in a linear, circular, or matrix pattern (adjusted and supplemented according to the actual scenario), periodically scans the physical boundary area at fixed time intervals of 0.3-0.5 seconds. By emitting ultrasonic signals and receiving reflected echoes, it obtains spatial information such as the target's distance and movement trajectory. The initial sensing signal triggered by the infrared sensor is correlated and verified in real time with the spatial information obtained by the ultrasonic sensor array. When the detection results of the two types of sensors meet the preset matching conditions (including synchronous target appearance time, overlapping spatial positions, and consistent movement trends) and the confidence threshold of the verification result is greater than 80%, an effective intrusion warning signal is generated. If a single sensor triggers a signal or the verification confidence is ≤80%, it is judged as invalid interference (such as changes in ambient light, airflow disturbances, non-intrusive target obstruction, etc.), and no warning is generated. After the ultrasonic sensor array outputs a pulse signal, the system must complete the above cross-verification and determination of the effectiveness of the intrusion warning signal within 2 seconds to ensure the real-time nature of intrusion detection and avoid missed detections due to delayed response.

[0022] The intrusion detection module, deployed at the physical boundary of the area (e.g., a metal fence), consists of multiple sensor nodes. Each node integrates a pyroelectric infrared sensor (with a detection range of at least 8 meters and a temperature accuracy of ±0.5℃) and an ultrasonic ranging sensor (with a range of 0.3 to 4 meters and a ranging accuracy of ±3 millimeters). These sensors scan at 0.5-second intervals, and their output analog signals are digitized by a 16-bit analog-to-digital converter (ADC) and transmitted to the central control unit. The central control unit executes a dual-sensor cross-validation algorithm, confirming an intrusion only when the detection results from both sensors jointly support an intrusion event and the overall confidence level exceeds 80%.

[0023] The pulse warning module includes: The IGBT driver circuit is connected to the central control unit and is used to receive the pulse width modulation signal from the central control unit, generate a low-voltage electric field pulse sequence, and the generated pulse is coupled to the boundary conductor of the field through a safety isolation transformer. The IGBT driver circuit belongs to the existing circuit structure, such as the dedicated driver chip based on optocoupler isolation, such as CONCEPT's 2SC0108T chip. The Hall current sensor, communicating with the central control unit, is used to monitor the pulse current in real time, ensuring it remains below the safe threshold of 1 mA. The low-voltage electric field pulse sequence generated by the pulse warning module consists of 3-5 millisecond-level pulses, with gradually increasing voltage not exceeding 50V, current not exceeding 1mA, pulse width of 10-50ms, and pulse interval of 100-150ms.

[0024] Specifically, the pulse warning module is connected to the central control unit and the field boundary conductor. After receiving the intrusion warning signal, it generates a low-voltage electric field pulse sequence and outputs it through the field boundary conductor. In this case, the pulse warning module couples the low-voltage electric field pulse to the field boundary conductor through a safety isolation transformer. Its core is an insulated gate bipolar transistor (IGBT) drive circuit. This circuit receives a pulse width modulation (PWM) signal from the central control unit, generates a low-voltage electric field pulse sequence, and the generated pulse is coupled to the field boundary conductor through the safety isolation transformer. The module integrates a Hall current sensor to monitor the pulse current in real time, ensuring that it is always below the safety threshold of 1 mA.

[0025] Environment-adaptive threshold module: The environmental adaptive threshold module integrates temperature and humidity sensors and a photosensor to collect parameters such as relative humidity (RH) and ambient illuminance in real time, providing a basis for the central control unit to adjust pulse parameters. Specifically: Connected to the central control unit, including: The temperature and humidity sensor is connected to the central control unit to collect the ambient relative humidity (RH) and temperature in real time, providing a basis for the central control unit to adjust the pulse parameters; A photosensitive sensor, connected to the central control unit, is used to collect ambient illuminance in real time, providing a basis for the central control unit to adjust pulse parameters; The environmental adaptive threshold module is used to adjust the pulse parameters of the XX module based on environmental data; in this case, the environmental adaptive threshold module integrates temperature, humidity, and photosensitivity sensors, and adjusts the parameters using formula V. adapt =V base * (1 - 0.2 * RH / 100), calculate the adjusted voltage; Where V adapt For the adjusted voltage value, V base RH represents the reference voltage value and the relative humidity percentage of the current environment, collected in real time.

[0026] To address the poor environmental adaptability of fixed threshold systems, an environmentally adaptive threshold module and specific parameter adjustment mechanisms (such as humidity-based linear voltage adjustment) are introduced. This allows for dynamic optimization of system parameters, effectively suppressing leakage risks and false alarms in high-humidity environments. Simultaneously, the intrusion detection module employs cross-validation using both infrared and ultrasonic sensors (confidence threshold > 80%). Through data fusion, it effectively filters transient interference, raising the system's false alarm rate to an industry-leading level and significantly improving the system's reliability and adaptability in complex environments.

[0027] The multimodal interaction layer includes: Directional speakers, placed at the physical boundaries of the area and connected to the central control unit, are used to play pre-recorded voice warnings in a specific direction; Vibration sensors, embedded in the boundary ground in the form of vibration pads, are connected to the central control unit to generate tactile feedback with adjustable intensity. The Bluetooth communication module connects to the central control unit and is used for short-range wireless interaction with the safety helmets of workers that are integrated with RFID tags, sending coded commands. A multimodal interaction layer is used to simultaneously activate at least one of a directional speaker, vibration sensor, or directional wireless signal (tactile feedback followed by auditory / vibration) upon pulse warning, and the multimodal interaction is completed within 3 seconds; the central control unit implements progressive upgrades of multimodal interaction based on state machine IF-THEN logic. The directional wireless signal interacts with the wearable device via Bluetooth, sending short pulse sequences encoded to indicate the exit path.

[0028] This invention addresses the ineffectiveness of sound and light alarms for specific groups by using a universal tactile warning—a low-voltage electric field pulse—as its core, combined with a multimodal interaction layer (directional speaker, vibration pad, Bluetooth signal) to form a three-dimensional warning network. This design ensures that all individuals, including the blind and deaf, can effectively receive the danger signal, overcoming the limitations of traditional warning methods and achieving comprehensive and effective warnings for all types of people.

[0029] Power control module: The power control module includes a solid-state relay array connected to the central control unit. It is used to quickly disconnect the field power supply upon receiving an "isolation upgrade signal." If the response is invalid, it disconnects the field power supply and activates physical isolation. Composed of a solid-state relay array, its response time is less than 50 milliseconds, used to quickly disconnect the auxiliary power supply and main high-voltage power supply within the field.

[0030] The monitoring and feedback module includes: The current detection unit communicates with the central control unit. It can be implemented using common Hall effect current sensors such as Allegro's ACS712, to collect response data after pulse output and determine the validity of the response based on a pre-set confidence threshold using cross-validation. It reuses the intrusion detection module's sensor and includes an additional independent current monitoring unit to collect personnel response data after the warning phase.

[0031] The monitoring feedback module collects response data after the pulse output and determines the validity of the response based on a pre-set confidence threshold using cross-validation. Validity assessment includes calculating the response rate, which is obtained by weighting the boundary motion change rate and the multimodal confirmation signal. A response rate greater than the preset response rate threshold (85%) is considered valid. Assume that after the pulse warning sequence is output, the monitoring feedback module collects the following data within a 2-second response verification window: Boundary motion change rate: The ultrasonic sensor measures that the intruding target moved 1.5 meters away from the boundary within 2 seconds. Change rate = 1.5 meters / 2 seconds = 0.75 meters / second. Normalizing this rate to the 0-1 range, assuming a maximum expected rate of 1 meter / second, the score for this item is 0.75.

[0032] Multimodal Acknowledgment Signal: The system activated the directional speaker and Bluetooth module. Speaker Feedback: The microphone detected a clear voice reflection, score 1.0. Bluetooth Feedback: Successfully received the ACK confirmation signal from the worker's safety helmet, score 1.0. Total Multimodal Acknowledgment Signal = (1.0 + 1.0) / 2 = 1.0. Calculate Response Rate: Response Rate = (Boundary Motion Change Rate Score + Multimodal Acknowledgment Signal) / 2 * 100% = (0.75 + 1.0) / 2 * 100% = 87.5%. Judgment: Since 87.5% > the preset threshold of 85%, the system determines the response is valid, generates a "Safety Recovery Signal," does not trigger power cut-off, and the system returns to the initial monitoring state.

[0033] The system in this case implements a progressive response from intrusion detection, pulse warning, response verification to power cut-off. The pulse warning module is activated before the power control module before the response is invalid, thus achieving gradual and harmless isolation by simulating the perception of high voltage with low voltage.

[0034] The central control unit system's self-test sequence includes, in order: intrusion detection module (infrared and ultrasonic sensors) calibration, pulse warning module leakage current testing, and multimodal connectivity verification. The leakage current threshold for the leakage current test is less than 0.1mA, and the multimodal connectivity verification threshold is greater than 95%. The central control unit's event recovery function executes 5-10 minutes after isolation and includes power sequence restoration and V-shaped recovery based on historical response rates less than 85%. adapt Adaptive parameter pre-tuning.

[0035] A workflow for a real-time monitoring and alarm system for personnel intrusion in a high-voltage test field includes the following steps: Step S1: Environmental perception and intrusion pre-detection; After the system starts up, the intrusion detection module continuously scans the boundary of the field. When a potential target is initially detected (for example, the target is less than 1 meter away and moves at a speed greater than 0.2 meters per second), the central control unit does not immediately issue an alarm, but instead starts a dual-sensor cross-validation algorithm to effectively distinguish whether it is a real person intruding, avoid interference from wind, grass, small animals, etc., and significantly reduce the false positive rate. Specifically, the dual-sensor cross-validation algorithm is built into the central control unit, and its execution steps are as follows: S1.1: Data Acquisition and Signal Preprocessing The central control unit reads the raw signals (converted by ADC) from the infrared sensor and the ultrasonic sensor in parallel through a serial communication interface. The infrared sensor signal is converted into an infrared confidence value ranging from 0% to 100%, and the calculation formula is as follows: Infrared confidence level = [(Detection temperature - Ambient baseline temperature) / (Human body temperature threshold - Ambient baseline temperature)] * 100% The ultrasonic sensor signal is converted into an ultrasonic confidence value ranging from 0% to 100%, which is calculated as follows: Ultrasonic confidence score = min[100%, (threshold distance - actual distance) / threshold distance * 100%] * motion weighting factor in: The threshold distance is preferably 1m; The motion weighting factor is defined in segments based on the real-time motion velocity v (in m / s) of the monitored object: If the real-time motion speed v > 0.2 m / s, then the motion weighting factor is set to 1.2 (to strengthen the confidence weight of dynamic intrusion). If the real-time motion speed v≤0.2m / s, then the motion weighting factor is set to 1.0 (to avoid excessive amplification of static interference). If the confidence value of any of the above sensors is lower than 20%, it will be directly determined as invalid interference, and the verification will end. S1.2 Cross-validation and fusion computation: The central control unit fuses the two confidence levels mentioned above to calculate the overall confidence level, using the following formula: Overall confidence level = (Infrared confidence level * Ultrasonic confidence level) / 100; This step emphasizes the "joint confirmation" of the two sensors. A lower confidence level for either sensor will significantly lower the overall value, thus forming a conservative and reliable judgment strategy. For example, when the confidence level of infrared is 85% and the confidence level of ultrasound is 90%, the calculated overall confidence level is 76.5%.

[0036] S1.3 Threshold Judgment and Decision Output: The central control unit compares the calculated overall confidence level with a preset 80% confidence threshold: If the overall confidence level is >80%, it is determined as "verification passed", and an "intrusion warning signal" containing a timestamp is generated and the process is advanced to the pulse warning step; If the overall confidence level is ≤80%, it is judged as "verification failed", and the system can trigger a resampling for a second verification. If the verification fails multiple times in a row, the system can enter a low-alert mode to shorten the scanning cycle and enhance monitoring.

[0037] At the same time, the environment adaptive threshold module performs environment adaptive pre-adjustment: Simultaneously with intrusion detection, the central control unit reads real-time environmental data from the environmental adaptive threshold module and, based on the pre-stored V... adapt =V base The algorithm (1-0.2*RH / 100) dynamically pre-adjusts the pulse warning parameters to address the increased leakage risk caused by decreased conductor surface resistance in high humidity environments, ensuring the system always operates within safe thresholds. Specifically, the formula for calculating the adaptive dynamic pre-adjustment of the pulse warning parameters is as follows: V adapt =V base * (1 - 0.2 * RH / 100), Among them, V adapt The adjusted voltage value; V base This is a reference voltage value (e.g., 40V, which is an optimized value under standard ambient RH < 60%). RH represents the percentage of relative humidity collected in real time. A coefficient of 0.2 indicates that for every 10% increase in humidity, the voltage is reduced by 20%, which is the approximate voltage drop required to maintain a safe current.

[0038] For example, in high humidity scenarios (e.g., RH=90%), a fixed 40V pulse may cause condensation discharge, posing a risk of current exceeding the 1mA safety threshold; at this time, the reference voltage V base =40V, then use the following formula to calculate the pulse warning parameter that should be adjusted in real time: Calculate the humidity normalization ratio: RH / 100 = 90 / 100 = 0.9; Calculate the humidity impact factor: 0.2*RH / 100 = 0.2*0.9 = 0.18; Calculate the scaling factor: 1 - 0.18 = 0.82; Apply scaling to obtain the adjusted voltage: V adapt =V base*(1-0.2*RH / 100)=40*0.82=32.8V, Therefore, the planned output pulse voltage reference value was lowered from 40V to 32.8V to enhance safety; It should be noted that the above process is a real-time parameter adjustment of the system for a single intrusion event. At a more macro level, the system also has the ability to iteratively optimize based on historical data to continuously improve its long-term operating efficiency. This process will be executed in the event recovery phase of step S5.

[0039] Step S2: Pulse Warning and Multimodal Output: Upon receiving an intrusion warning signal, the central control unit immediately triggers the pulse warning module. The pulse warning module generates a sequence of three pulses with progressively increasing voltages (e.g., 20V → 25V → 32.8V), with the highest voltage not exceeding the internationally recognized low-voltage safety threshold of 50V. Each pulse is 30 milliseconds wide and spaced 100 milliseconds apart. This sequence is applied to the boundary conductor via an isolation transformer, generating a tactile sensation sufficient to attract attention while ensuring safety.

[0040] Almost simultaneously, the multimodal interaction layer is activated; the central control unit, based on the IF-THEN logic of the state machine, simultaneously activates the directional speakers to play voice warnings and activates the ground vibration pads. Specifically, auditory stimulation of the monitored object is achieved by playing voice warnings through a loudspeaker, such as "High voltage danger, please retreat immediately," which quickly conveys clear instructions. This is suitable for people with normal hearing, enhances immediate alertness, and compensates for the limitations of visual dependence, such as in foggy weather. By activating the ground vibration pad to provide tactile feedback with low-frequency vibration intensity of 3g, it simulates a ground alarm and is suitable for visually impaired or disabled people in noisy environments, ensuring universality. It also works in conjunction with pulse tactile sensation to form a multi-layered sensory experience. Meanwhile, if the Bluetooth module detects an authorized safety helmet nearby, it sends a specific short pulse code to it. This code can be interpreted as an "exit path" indication, such as sending path guidance (e.g., a virtual safety line), to avoid generalized alarm interference.

[0041] Step S3: Response Verification and Threshold Determination After the pulse sequence ends, the system opens a 2-second response verification window (based on an optimal value for physiological recovery time and engineering response window) to observe changes in personnel behavior and ensure the effectiveness of the deportation. The monitoring feedback module collects two types of data in this window: The first is the boundary motion change rate (change distance / change time) obtained by measuring it again through an ultrasonic sensor. Second, multimodal confirmation signals, such as confirmation receipts received by the Bluetooth module.

[0042] The central control unit performs a weighted evaluation based on the boundary motion change rate and multimodal confirmation signals to obtain the overall response rate; the calculation formula is as follows: Response rate = (Boundary motion change rate + Multimodal confirmation signal) / 2 * 100%; If the response rate is greater than 85%, the response is deemed valid, the system generates a "safety recovery signal" and returns to the initial monitoring state; If the response rate does not reach this threshold (for example, the measured value is 70%), the response is deemed invalid and an "isolation upgrade signal" is generated. The multimodal confirmation signal originates from the response feedback data after the interaction layer is activated, and is obtained by independently scoring and normalizing each modal feedback. Specifically, the central control unit collects the following feedback in real time within the response verification window: Speaker feedback: Monitors ambient volume reflections after voice playback via a built-in microphone.

[0043] Vibration pad feedback: Monitors changes in ground vibration attenuation caused by personnel movement.

[0044] Bluetooth module feedback: Receives acknowledgment (ACK) signal returned by the safety helmet.

[0045] The feedback intensity of each modality is mapped to a score between 0 and 1.

[0046] For example, in one specific implementation: when the volume reflection is greater than 50 dB, the speaker score can be 1.0; when the vibration attenuation is greater than 80%, the vibration score can be 1.0; when a complete Bluetooth ACK signal is received, the Bluetooth score can be 1.0; then, these scores are summed and divided by the number of currently active modes, and the result is finally limited to the range of 0 to 1, serving as a comprehensive multimodal acknowledgment signal; an example is shown below: In high-response scenarios, if all modal feedbacks reach the highest threshold, the confirmation signal is 1.0; In partial response scenarios, if only some modes reach the medium threshold, the confirmation signal is 0.5; In a no-response scenario, the acknowledgment signal is 0.0.

[0047] Step S4: Power off and field isolation: Once the "isolation upgrade signal" is received, the power control module sequentially cuts off the auxiliary power supply and the main high-voltage power supply of the field; at the same time, it activates physical isolation devices such as electromagnetic door locks, sets the field status to "safe mode", and broadcasts a pause signal.

[0048] Step S5: Event Recovery and System Self-Check After the field isolation is maintained for a configurable period of time (e.g., 8 minutes), the system confirms that personnel have retreated to a safe distance (e.g., 2 meters away) and that the power supply is stable. Subsequently, the central control unit performs a system self-test, which includes sensor calibration, leakage current test of the pulse module (ensuring leakage current is less than 0.1 mA), and multimodal connectivity verification (requiring a success rate of over 95%).

[0049] After the self-test passes, the system restores power sequentially. It should be noted that, in addition to the real-time adjustment for a single event, the system has iterative optimization capabilities. Its core lies in the fact that the central control unit can learn and adjust its strategies based on historical event data, thereby achieving continuous improvement in operational efficiency.

[0050] During the event recovery phase, the system executes the following optimization process: Scene analysis and statistics: The central control unit filters historical records from the stored historical logs that are similar to the current event environment characteristics (e.g., relative humidity deviation within ±10%), and calculates the average response rate of these records.

[0051] Optimize trigger judgment: If the calculated average response rate does not reach the preset effective threshold (e.g., ≤85%), it is determined that the response performance in this type of scenario needs to be improved, and the parameter pre-adjustment mechanism is triggered.

[0052] Parameter adaptive pre-tuning: The system fine-tunes key parameters in the environmental adaptive formula. For example, the scaling factor in the voltage adjustment formula can be appropriately increased (calculated as: new scaling factor = original scaling factor + 0.05 * (1 - average response rate / 0.85)). Alternatively, the reference voltage value can be appropriately lowered. This aims to adopt a more conservative and safer strategy when facing similar environments in the future, further improving the success rate of decoupling and system security.

[0053] Closed-loop application: The optimized parameters will be updated to the system's default configuration to guide subsequent response decisions. In this way, the system can learn from actual operating experience, continuously improve itself, and form a negative feedback loop for continuous improvement.

[0054] For example, in a continuously high-humidity environment, if the historical average response rate is low, the system can automatically increase the voltage reduction ratio according to the above process. This optimizes the psychological deterrent effect of the pulse warning while ensuring safety, ultimately achieving an effective improvement in the response rate. This optimization process is fully automated and requires no manual intervention, fully demonstrating the system's level of intelligence.

[0055] This invention addresses the shortcomings of existing technologies that use a "one-size-fits-all" approach to power disconnection by proposing a progressive response core logic of "pulse warning → response verification → power disconnection." The system prioritizes using low-voltage, harmless electrical pulses for sensory disengagement, only escalating to power isolation when verification confirms the disengagement is ineffective. This "warning priority, disconnection backup" mechanism, while ensuring absolute personal safety, reduces the probability of test interruption to an extremely low level, significantly improving the continuity and economy of high-voltage testing, and achieving a precise balance between safety and efficiency.

[0056] Meanwhile, during the event recovery phase, this system iteratively pre-adjusts adaptive parameters based on historical response data, enabling the system to learn from actual operation and continuously optimize response strategies. This achieves an intelligent upgrade from "passive response" to "active optimization," and its long-term operational efficiency and reliability are far superior to existing static systems.

[0057] Regarding the information disclosed in this case, the following points need to be clarified: (1) The accompanying drawings of the embodiments disclosed in this case only involve the structures involved in the embodiments disclosed in this case. Other structures can refer to the general design. (2) Where there is no conflict, the embodiments and features disclosed in this case can be combined with each other to obtain new embodiments; The above are merely specific embodiments disclosed in this case, but the scope of protection of this disclosure is not limited thereto. The scope of protection disclosed in this case shall be determined by the scope of protection of the claims.

Claims

1. A real-time monitoring and alarm system for personnel intrusion in a high-voltage test field, characterized in that, include: An intrusion detection module, deployed at the site boundary, is used to monitor the physical boundary in real time and generate intrusion warning signals; The pulse warning module is connected to the central control unit and the field boundary conductor. After receiving the intrusion warning signal, it generates a low-voltage electric field pulse sequence and outputs it through the field boundary conductor. The monitoring and feedback module is connected to the central control unit and is used to collect response data after pulse output and determine whether the response is valid based on a preset confidence threshold of cross-validation. The power control module is connected to the central control unit. When the response is invalid, it cuts off the field power and activates physical isolation. Together with the central control unit, it coordinates the work of each module and performs system self-tests and event recovery.

2. The real-time monitoring and alarm system for personnel intrusion in a high-voltage test field according to claim 1, characterized in that, The intrusion detection module includes: Infrared sensors are used to detect and monitor changes in infrared radiation in physical boundary areas; The ultrasonic sensor array, according to its layout, periodically scans the physical boundary area at fixed time intervals of 0.3-0.5 seconds, and obtains target spatial information by emitting ultrasonic signals and receiving reflected echoes.

3. The real-time monitoring and alarm system for personnel intrusion in a high-voltage test field according to claim 1, characterized in that, The pulse warning module includes: The IGBT driving circuit is connected to the central control unit and is used to receive the pulse width modulation signal from the central control unit, generate a low-voltage electric field pulse sequence, and the generated pulse is coupled to the boundary conductor of the field through a safety isolation transformer. The Hall current sensor communicates with the central control unit to monitor the pulse current in real time, ensuring that it is always below the safety threshold of 1 mA.

4. The real-time monitoring and alarm system for personnel intrusion in a high-voltage test field according to claim 1, characterized in that, The monitoring feedback module includes: The current detection unit communicates with the central control unit; the current detection unit can be implemented using common Hall current sensors such as Allegro's ACS712.

5. The real-time monitoring and alarm system for personnel intrusion in a high-voltage test field according to claim 1, characterized in that, The power control module includes a solid-state relay array connected to the central control unit, used to quickly cut off the field power supply after receiving an "isolation upgrade signal".

6. The real-time monitoring and alarm system for personnel intrusion in a high-voltage test field according to claim 1, characterized in that, It also includes an environment adaptive threshold module connected to the central control unit; The environment adaptive threshold module includes: The temperature and humidity sensor is connected to the central control unit to collect the ambient relative humidity (RH) and temperature in real time, providing a basis for the central control unit to adjust the pulse parameters; A photosensitive sensor, connected to the central control unit, is used to collect ambient illuminance in real time, providing a basis for the central control unit to adjust pulse parameters.

7. The real-time monitoring and alarm system for personnel intrusion in a high-voltage test field according to claim 1, characterized in that, It also includes a multimodal interaction layer connected to the central control unit; the multimodal interaction layer includes: Directional speakers, placed at the physical boundaries of the area and connected to the central control unit, are used to play pre-recorded voice warnings in a specific direction; Vibration sensors, embedded in the boundary ground in the form of vibration pads, are connected to the central control unit to generate tactile feedback with adjustable intensity. The Bluetooth communication module connects to the central control unit and is used for short-range wireless interaction with the safety helmets of workers that are equipped with RFID tags, sending coded commands.

8. A workflow for a real-time monitoring and alarm system for personnel intrusion in a high-voltage test field, comprising the real-time monitoring and alarm system for personnel intrusion in a high-voltage test field as described in claim 1, characterized in that, Includes the following steps: Step S1: Environmental perception and intrusion pre-detection; After the system starts up, the intrusion detection module continuously scans the boundary of the field. When a potential target is initially detected, the central control unit does not immediately issue an alarm, but instead activates a dual-sensor cross-validation algorithm to effectively distinguish whether it is a real person breaking in and reduce the false positive rate. Specifically, the dual-sensor cross-validation algorithm is built into the central control unit, and its execution steps are as follows: The central control unit reads the raw signals from the infrared sensor and the ultrasonic sensor in parallel through a serial communication interface; The infrared sensor signal is converted into an infrared confidence value ranging from 0% to 100%, and the calculation formula is as follows: Infrared confidence level = [(Detection temperature - Ambient baseline temperature) / (Human body temperature threshold - Ambient baseline temperature)] * 100% The ultrasonic sensor signal is converted into an ultrasonic confidence value ranging from 0% to 100%, which is calculated as follows: Ultrasonic confidence score = min[100%, (threshold distance - actual distance) / threshold distance * 100%] * motion weighting factor in: The threshold distance is preferably 1m; The motion weighting factor is defined in segments based on the real-time motion velocity v (in m / s) of the monitored object: If the real-time motion speed v > 0.2 m / s, then the motion weighting factor is set to 1.2; If the real-time motion velocity v ≤ 0.2 m / s, then the motion weighting factor is set to 1.0; If the confidence value of any of the above sensors is lower than 20%, it will be directly determined as invalid interference, and the verification will end. Cross-validation and fusion computing: The central control unit fuses the two confidence levels mentioned above to calculate the overall confidence level, using the following formula: Overall confidence level = (Infrared confidence level * Ultrasonic confidence level) / 100; Threshold judgment and decision output: The central control unit compares the calculated overall confidence level with a preset 80% confidence threshold: If the overall confidence level is >80%, it is determined as "verification passed", and an "intrusion warning signal" containing a timestamp is generated and the process is advanced to the pulse warning step; If the overall confidence level is ≤80%, it is judged as "verification failed", and the system can trigger a resampling for a second verification. If the verification fails multiple times in a row, the system can enter a low-alert mode to shorten the scanning cycle and enhance monitoring. Step S2: Pulse Warning and Multimodal Output: Upon receiving an intrusion warning signal, the central control unit immediately triggers the pulse warning module. The pulse warning module generates a sequence of three pulses with progressively increasing voltages, the highest of which does not exceed the internationally recognized low-voltage safety threshold of 50V. Each pulse is 30 milliseconds wide and spaced 100 milliseconds apart. This sequence is applied to the boundary conductor through an isolation transformer, producing a tactile sensation sufficient to attract attention while ensuring safety. Step S3: Response Verification and Threshold Determination After the pulse sequence ends, the system opens a response verification window to observe changes in personnel behavior and ensure the effectiveness of the expulsion. The monitoring feedback module collects two types of data in this window: One is the rate of change of boundary motion obtained by measuring it again using an ultrasonic sensor; Second, multimodal confirmation signals, such as confirmation receipts received by the Bluetooth module; The central control unit performs a weighted evaluation based on the boundary motion change rate and multimodal confirmation signals to obtain the overall response rate; the calculation formula is as follows: Response rate = (Boundary motion change rate + Multimodal confirmation signal) / 2 * 100%; If the response rate is greater than 85%, the response is considered valid, the system generates a "safety recovery signal" and returns to the initial monitoring state; If the response rate does not reach this threshold, the response is deemed invalid, and an "isolation escalation signal" is generated. Among them, the multimodal confirmation signal comes from the response feedback data after the interaction layer is activated, and is obtained by independently scoring and normalizing each modal feedback; Step S4: Power off and field isolation: Once the "isolation upgrade signal" is received, the power control module sequentially cuts off the auxiliary power supply and the main high-voltage power supply of the field; at the same time, it activates physical isolation devices such as electromagnetic door locks, sets the field status to "safe mode", and broadcasts a pause signal.

9. The workflow of a real-time monitoring and alarm system for personnel intrusion in a high-voltage test field according to claim 8, characterized in that, The calculation formula for the adaptive dynamic pre-adjustment of the pulse warning parameter in step S1.3 is as follows: In adapt =V base *(1-0.2*RH / 100), Among them, V adapt The adjusted voltage value; V base The reference voltage value; RH represents the percentage of relative humidity collected in real time. A coefficient of 0.2 indicates that for every 10% increase in humidity, the voltage is reduced by 20%, which is the approximate voltage drop required to maintain a safe current.

10. The workflow of a real-time monitoring and alarm system for personnel intrusion in a high-voltage test field according to claim 8, characterized in that, Also includes: Step S5: Event Recovery and System Self-Check After maintaining the isolation of the field for a configurable period of time, the system confirms that personnel have retreated to a safe distance and that the power supply is stable. Subsequently, the central control unit performs a system self-test, which includes sensor calibration, leakage current testing of the pulse module, and verification of multimodal connectivity. After the self-test passes, the system restores power in sequence.

Images