Intelligent detection system and method for small animal invasion of transformer substation

Abstract

The invention discloses an intelligent detection system and method for small animal invasion of a transformer substation, and relates to the technical field of power system safety protection. The system comprises a multi-sensor detection module which at least comprises a motion detection sensor, a microwave radar, an infrared temperature sensor and an environment temperature and humidity detector; and the main controller is connected with the multi-sensor detection module and is used for receiving the multi-source sensing detection data output by the multi-sensor detection module and calculating the total confidence coefficient based on each sensing detection data in the multi-source sensing detection data and the respective weight so as to judge whether intrusion occurs or not based on the total confidence coefficient. According to the scheme, the false alarm rate is greatly reduced, the accuracy is remarkably improved, the usability is greatly improved, the traceability is enhanced, and the seasonal adaptability is good.

Description

Technical Field

[0001] This invention relates to the field of power system security protection technology, specifically to an intelligent detection system and method for small animal intrusion in substations. Background Technology

[0002] Currently, substation solutions for preventing small animal intrusion typically employ passive defense methods (e.g., barriers, blockages, sticky traps, rodenticides), which lack early warning capabilities; rely on regular inspections, resulting in low efficiency; lack of data records makes it difficult to trace the source, making it hard to determine whether the intrusion was caused by small animals or other factors after an incident, and impossible to analyze animal activity patterns; the methods are simplistic and unreliable, for example, using only blockages is easily breached because animals may intrude through other routes; and emergency response costs are high (for unmanned substations), requiring personnel to be dispatched urgently to remote substations to troubleshoot and restore power after a power outage, which is time-consuming and costly.

[0003] Based on the above analysis, existing solutions for preventing small animal intrusion into substations mainly have the following technical shortcomings:

[0004] (1. Low detection reliability: Single sensors are susceptible to environmental interference, and the false alarm rate is generally 25-35%;)

[0005] (2. Insufficient level of intelligence: lack of multi-sensor data fusion and intelligent decision-making mechanisms;)

[0006] (3. Poor system stability: lacks automatic fault diagnosis and recovery capabilities;)

[0007] (4. Limited early warning methods: unable to achieve tiered early warning and precise response;)

[0008] (5. High maintenance costs: frequent manual intervention and calibration are required.) Summary of the Invention

[0009] This invention addresses the problems of low detection reliability, insufficient intelligence, poor system stability, single early warning method, and high maintenance cost in existing substation small animal intrusion prevention schemes. It provides an intelligent detection system and method for small animal intrusion in substations, which significantly reduces the false alarm rate, significantly improves accuracy, greatly enhances usability, strengthens traceability, and improves seasonal adaptability.

[0010] The present invention is achieved through the following technical solution.

[0011] On one hand, the present invention provides an intelligent detection system for small animal intrusion in substations, the system comprising:

[0012] A multi-sensor detection module, wherein the multi-sensor detection module includes at least: a motion detection sensor, a microwave radar, an infrared temperature sensor, and an ambient temperature and humidity detector; and

[0013] The main controller is connected to the multi-sensor detection module and is used to receive multi-source sensor detection data output by the multi-sensor detection module, and calculate the total confidence score based on each sensor detection data and its respective weight in the multi-source sensor detection data, so as to determine whether it is a small animal invasion based on the total confidence score.

[0014] In some embodiments, the main controller is further configured to: calculate the total confidence score based on each sensor detection data and its respective weight in the multi-source sensor detection data, including:

[0015] Based on the sensor detection data, the data collected by the motion detection sensor, the microwave radar, the infrared temperature sensor, and the ambient temperature and humidity detector are acquired, and spatiotemporal alignment is performed on the data collected by the motion detection sensor, the microwave radar, the infrared temperature sensor, and the ambient temperature and humidity detector.

[0016] Based on the data collected by the environmental temperature and humidity detector, and using an environmental temperature and humidity compensation algorithm, the confidence levels of the data collected by the motion detection sensor, the microwave radar, and the infrared temperature sensor are calculated after spatiotemporal alignment; and

[0017] The total confidence level is calculated based on the confidence level and weight of the data collected by the motion detection sensor, the confidence level and weight of the data collected by the microwave radar, the confidence level and weight of the data collected by the infrared temperature sensor, and the temporal correlation and weight of the triggering of the motion detection sensor, the microwave radar and the infrared temperature sensor.

[0018] In some embodiments, the timing association is defined as the simultaneous triggering of all sensors in the multi-sensor detection module. Specifically, if at least two sensors in the multi-sensor detection module are triggered simultaneously within the same detection period, the timing association is 1. Alternatively, if fewer than two sensors in the multi-sensor detection module are triggered simultaneously within the same detection period, or if no sensor is triggered, the timing association is 0.

[0019] In some embodiments, the system further includes a watchdog monitoring subsystem for preventing the program from crashing or entering an infinite loop.

[0020] On the other hand, the present invention provides an intelligent detection method for small animal intrusion in substations, the method comprising:

[0021] Multi-source sensor data is acquired through a multi-sensor detection module, wherein the multi-sensor detection module includes at least: a motion detection sensor, a microwave radar, an infrared temperature sensor, and an environmental temperature and humidity detector; and

[0022] The main controller receives multi-source sensor detection data output by the multi-sensor detection module, and calculates the total confidence score based on each sensor detection data and its respective weight in the multi-source sensor detection data, so as to determine whether it is a small animal invasion based on the total confidence score.

[0023] In some embodiments, the total confidence score is calculated based on each sensor detection data and its respective weight in the multi-source sensor detection data, including:

[0024] Based on the sensor detection data, the data collected by the motion detection sensor, the microwave radar, the infrared temperature sensor, and the ambient temperature and humidity detector are acquired, and spatiotemporal alignment is performed on the data collected by the motion detection sensor, the microwave radar, the infrared temperature sensor, and the ambient temperature and humidity detector.

[0025] Based on the data collected by the environmental temperature and humidity detector, and using an environmental temperature and humidity compensation algorithm, the confidence levels of the data collected by the motion detection sensor, the microwave radar, and the infrared temperature sensor are calculated after spatiotemporal alignment; and

[0026] The total confidence level is calculated based on the confidence level and weight of the data collected by the motion detection sensor, the confidence level and weight of the data collected by the microwave radar, the confidence level and weight of the data collected by the infrared temperature sensor, and the temporal correlation and weight of the triggering of the motion detection sensor, the microwave radar and the infrared temperature sensor.

[0027] In some embodiments, the timing association is defined as the simultaneous triggering of all sensors in the multi-sensor detection module. Specifically, if at least two sensors in the multi-sensor detection module are triggered simultaneously within the same detection period, the timing association is 1. Alternatively, if fewer than two sensors in the multi-sensor detection module are triggered simultaneously within the same detection period, or if no sensor is triggered, the timing association is 0.

[0028] In some embodiments, the method further includes: using a watchdog monitoring subsystem to prevent the program from crashing or entering an infinite loop.

[0029] In some embodiments, the method further includes: performing fault diagnosis and self-recovery on the sensor, wherein the fault diagnosis and self-recovery includes:

[0030] Each sensor in the multi-sensor detection module is tested to determine whether any sensor is faulty.

[0031] If a sensor malfunction is detected, the malfunctioning sensor will be restored.

[0032] If the faulty sensor is successfully restored, a restoration log is recorded; if the faulty sensor is not successfully restored, the fault status is recorded.

[0033] In some embodiments, the method further includes: remotely configuring parameters for each sensor in the multi-sensor detection module, and updating the parameters of each sensor based on the remote parameter configuration.

[0034] Compared with existing technologies, this invention has the following advantages and beneficial effects: it adopts a four-level hierarchical verification fusion (three-layer sensor verification of motion detection sensor, microwave radar, and infrared temperature sensor, and a first-level fusion decision (calculation of confidence of each sensor weight based on environmental compensation and spatiotemporal relationship), upgrading from simple logical AND to confidence calculation, and realizing probabilistic risk assessment; deep software and hardware collaboration, realizing performance optimization; and three-level hierarchical early warning, realizing refined resource allocation. Attached Figure Description

[0035] To more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be considered as limiting the scope. For those skilled in the art, other related drawings can be obtained from these drawings without creative effort.

[0036] Figure 1 This is a structural block diagram of a substation small animal intrusion intelligent detection system according to an embodiment of the present invention.

[0037] Figure 2 This is a flowchart of a smart detection method for small animal intrusion in a substation according to an embodiment of the present invention.

[0038] Figure 3 This is a schematic diagram of the overall architecture of a substation small animal intrusion intelligent detection system according to an embodiment of the present invention.

[0039] Figure 4 This is a flowchart of a multi-sensor fusion algorithm according to an embodiment of the present invention.

[0040] Figure 5 This is a state diagram of a fault self-recovery mechanism according to an embodiment of the present invention. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0042] This invention addresses the problems of low detection reliability, insufficient intelligence, poor system stability, limited early warning methods, and high maintenance costs in existing substation small animal intrusion prevention solutions. It provides a smart small animal intrusion detection solution with high accuracy, low false alarm rate, and rapid response. This invention integrates PIR motion sensing, microwave radar detection, and infrared body temperature recognition, and incorporates an environmental temperature and humidity compensation algorithm to effectively detect small animal intrusions. Through hardware and software co-design and system-level design, this invention provides a highly reliable, self-managing, and traceable smart device, solving the pain points of high false alarm rate, poor reliability, and lack of data recording in traditional single-sensor solutions. This achieves an intelligent upgrade of substation protection from "passive response" to "active early warning."

[0043] On the one hand, the present invention provides an intelligent detection system for small animal intrusion in substations. Figure 1 This is a structural block diagram of a substation small animal intrusion intelligent detection system according to an embodiment of the present invention. (Reference) Figure 1 The substation's intelligent animal intrusion detection system includes at least a multi-sensor detection module and a main controller. Additionally, the system may include peripheral devices: a storage module, a display module, and an alarm module.

[0044] The multi-sensor detection module includes at least: a motion detection sensor, a microwave radar, an infrared temperature sensor, and an ambient temperature and humidity detector.

[0045] A motion sensor (Passive Infrared Sensor, PIR) is an electronic device that senses movement by detecting infrared radiation emitted by the human body or animal. It is widely used in smart lighting, security systems, and other fields. As a motion detection layer, its core principle is to use pyroelectric elements to sense infrared light with wavelengths of 8-14 micrometers, combined with a Fresnel lens to enhance detection sensitivity. However, it cannot detect stationary targets. PIRs can quickly capture motion signals. This invention employs a dual PIR redundancy design (arranged left and right), which can improve detection coverage and sensitivity, but also results in a higher false alarm rate.

[0046] Microwave radar (e.g., the AT5820 from Airsoft) is an electronic device that uses electromagnetic waves in the microwave frequency band (5.8GHz±75MHz) for detection, ranging, and velocity measurement. Its core principle is to transmit microwave signals and receive the echoes reflected from targets, analyzing information such as the time delay and frequency changes of the echoes to obtain parameters such as the target's position and velocity. Microwave radar acts as a presence confirmation layer, continuously tracking targets, possesses strong penetration capabilities, is unaffected by environmental conditions, and has moderate sensitivity and a moderate false alarm rate.

[0047] Infrared temperature sensors (e.g., MLX90614) are devices that use the infrared radiation naturally emitted by objects to perform non-contact temperature measurement. As a bio-verification layer, its core principle is to calculate the temperature by detecting the infrared energy emitted by the object, enabling biometric identification; body temperature range detection (set according to the animal's body temperature characteristics), accurate identification of small animals, with low sensitivity and low false alarm rate.

[0048] An environmental temperature and humidity detector is a device used to monitor changes in ambient temperature and humidity, and is widely used in industrial, residential, and laboratory settings.

[0049] The main controller (e.g., ESP32-S3) is connected to the multi-sensor detection module to receive multi-source sensor detection data output by the multi-sensor detection module, and calculates the total confidence score based on each sensor detection data and its respective weight in the multi-source sensor detection data, so as to determine whether it is a small animal invasion based on the total confidence score.

[0050] In this invention, by using a multi-sensor fusion algorithm, combined with (dual-channel) PIR motion sensing, microwave radar ranging, and infrared temperature measurement (MLX90614) for collaborative judgment, the detection accuracy can be greatly improved, effectively distinguishing small animals from other disturbances (such as wind blowing grass, light changes, and temperature fluctuations), fundamentally reducing false alarms and missed alarms.

[0051] In some embodiments, the main controller is further configured to: when calculating the total confidence score based on each sensor detection data and its respective weight in the multi-source sensor detection data, include: acquiring the data collected by the motion detection sensor, the microwave radar, the infrared temperature sensor, and the environmental temperature and humidity detector based on each sensor detection data, and performing spatiotemporal alignment on the data collected by the motion detection sensor, the microwave radar, the infrared temperature sensor, and the environmental temperature and humidity detector; calculating the confidence scores of the spatiotemporally aligned data collected by the motion detection sensor, the microwave radar, and the infrared temperature sensor based on the environmental temperature and humidity detector using an environmental temperature and humidity compensation algorithm; and calculating the total confidence score based on the confidence scores and their weights of the data collected by the motion detection sensor, the microwave radar, and the infrared temperature sensor, as well as the temporal correlation and weights of the triggering of the motion detection sensor, the microwave radar, and the infrared temperature sensor.

[0052] Using ambient temperature and humidity data for algorithmic compensation of motion sensors is a key method to improve system stability and accuracy, as changes in ambient temperature and humidity can cause sensor signal drift or nonlinear errors. Ambient temperature and humidity data can be used to correct parameters of various sensors, perform data fusion and filtering, etc., to reduce the impact of environmental factors on the data collected by each sensor.

[0053] Calculating the confidence level of sensor data is a crucial step in evaluating the reliability of measurement results, especially in multi-sensor fusion systems, where it directly impacts the accuracy of decision-making. Confidence level represents the probability that a rule or measurement result will be satisfied under given conditions. In sensor data fusion, it quantifies how closely a measured value approximates the true value, and is typically evaluated using statistical methods (such as confidence intervals) or support functions. Common methods for calculating sensor data confidence levels include statistical distance methods, weighted fusion methods, and confidence interval methods.

[0054] In some embodiments, timing association is defined as the simultaneous triggering of all sensors in the multi-sensor detection module. Specifically, if at least two sensors in the multi-sensor detection module are triggered simultaneously within the same detection period, the timing association is 1 (indicating multi-sensor collaborative triggering). Alternatively, if fewer than two sensors in the multi-sensor detection module are triggered simultaneously within the same detection period or no sensor is triggered, the timing association is 0 (indicating that the collaborative triggering condition has not been met).

[0055] The total confidence calculation module can be represented as:

[0056] Total confidence level = Weight 1 × PIR + Weight 2 × Radar + Weight 3 × Infrared temperature + Weight 4 × Time series

[0057] Among them, weights 1 to 4 can be set empirically or obtained by building a model and solving it, and satisfy: weight 1 + weight 2 + weight 3 + weight 4 = 1.

[0058] In some embodiments, the system further includes a watchdog monitoring subsystem for preventing the program from crashing or entering an infinite loop.

[0059] In some embodiments, the system can also implement a three-level early warning strategy based on the total confidence level: low confidence level, only log recording; medium confidence level, local audible and visual alarm; high confidence level, remote alarm + on-site handling.

[0060] In this invention, a four-level hierarchical verification fusion (three-layer sensor verification of motion detection sensor, microwave radar, and infrared temperature sensor, and a first-level fusion decision (calculation of confidence of each sensor weight based on environmental compensation and spatiotemporal relationship) is adopted, which upgrades from simple logical AND to confidence calculation and realizes probabilistic risk assessment; deep software and hardware collaboration realizes performance optimization; and three-level hierarchical early warning realizes refined resource allocation.

[0061] On the other hand, the present invention provides an intelligent detection method for small animal intrusion in substations. Figure 2 This is a flowchart of a smart detection method for small animal intrusion in a substation according to an embodiment of the present invention. Figure 3 This is a schematic diagram of the overall architecture of a substation small animal intrusion intelligent detection system according to an embodiment of the present invention. Figure 3 As shown, the overall architecture of this system involves a sensor layer, a data processing layer, and an execution output layer. The sensor layer involves four layers of verification, the data processing layer involves multi-sensor fusion algorithms and fault diagnosis and self-recovery, and the execution output layer involves WeChat robot, OLED (organic light-emitting diode) display, and data storage.

[0062] refer to Figure 2 and Figure 3 The intelligent detection method for small animal intrusion in this substation includes: S10 and S20.

[0063] In S10, multi-source sensor detection data is collected through a multi-sensor detection module, which includes at least: a motion detection sensor, a microwave radar, an infrared temperature sensor, and an environmental temperature and humidity detector.

[0064] In S20, the main controller receives multi-source sensor detection data output by the multi-sensor detection module, and calculates the total confidence score based on each sensor detection data and its respective weight in the multi-source sensor detection data, so as to determine whether it is a small animal invasion based on the total confidence score.

[0065] In some embodiments, calculating the total confidence score based on the multi-source sensor detection data and their respective weights includes: acquiring data from a motion detection sensor, a microwave radar, an infrared temperature sensor, and an environmental temperature and humidity detector based on the sensor detection data, and performing spatiotemporal alignment on the data from the motion detection sensor, microwave radar, infrared temperature sensor, and environmental temperature and humidity detector; calculating the confidence scores of the spatiotemporally aligned data from the motion detection sensor, microwave radar, infrared temperature sensor, and environmental temperature and humidity detector using an environmental temperature and humidity compensation algorithm based on the data from the environmental temperature and humidity detector; and calculating the total confidence score based on the confidence scores and their weights for the data from the motion detection sensor, microwave radar, and infrared temperature sensor, as well as the temporal correlation and weights of the triggers for the motion detection sensor, microwave radar, and infrared temperature sensor. This can greatly improve the detection accuracy, effectively distinguish small animals from other disturbances (such as wind blowing grass, light changes, and temperature fluctuations), and fundamentally reduce false alarms and missed alarms.

[0066] Figure 4 This is a flowchart of a multi-sensor fusion algorithm according to an embodiment of the present invention. (Refer to...) Figure 4 The multi-sensor fusion algorithm involves the aforementioned four-level verification.

[0067] In some embodiments, timing association is defined as the simultaneous triggering of all sensors in the multi-sensor detection module. Specifically, if at least two sensors in the multi-sensor detection module are triggered simultaneously within the same detection period, the timing association is 1. Alternatively, if fewer than two sensors in the multi-sensor detection module are triggered simultaneously within the same detection period, or if no sensor is triggered, the timing association is 0.

[0068] In some embodiments, the method further includes: using a watchdog monitoring subsystem to prevent the program from crashing or entering an infinite loop. By using a watchdog timer and periodic sensor reconnection detection, highly reliable "unattended" operation and automatic recovery are achieved, adapting to the harsh environment of substations, improving system stability, significantly reducing the number of on-site maintenance visits by maintenance personnel, and lowering lifecycle costs.

[0069] In some embodiments, the method further includes: fault diagnosis and self-recovery of the sensor, wherein fault diagnosis and self-recovery includes: detecting each sensor in the multi-sensor detection module to determine whether each sensor is faulty; if a sensor fault is detected, recovering the faulty sensor; if the recovery of the faulty sensor is successful, recording a recovery log, or if the recovery of the faulty sensor is unsuccessful, recording a fault status. Figure 5 This is a state diagram of the fault self-recovery mechanism according to an embodiment of the present invention. Additionally, it can also achieve automatic sensor recalibration: periodic health checks and parameter calibration; communication link self-repair: automatic reconnection after WiFi (Wireless Fidelity) disconnection, and dual-key backup for the WeChat robot; seasonal adaptability: optimization of detection parameters according to different seasons.

[0070] In some embodiments, the method further includes: remotely configuring parameters for each sensor in the multi-sensor detection module, and updating parameters for each sensor based on the remote parameter configuration.

[0071] In some embodiments, intelligent log management can be achieved through data storage: logs are stored on SD (Secure Digital Card) cards, organized by date, with automatic cleanup of old logs and support for log viewing modes. Even during network outages, all events are fully recorded, providing valuable data for incident tracing, fault analysis, and optimization of protection strategies, thus compensating for the shortcomings of purely cloud-based solutions.

[0072] In some embodiments, enterprise WeChat notifications and local audio-visual alarms are used, and a backup enterprise WeChat key is considered to ensure that alarm information is effectively delivered, improve the reliability of notifications, and utilize modern communication methods, which is more in line with current technological trends.

[0073] In some embodiments, network time synchronization is employed: time is synchronized via an NTP (Network Time Protocol) server to ensure the accuracy of log time.

[0074] In some embodiments, sensor de-jitter processing is performed on PIR and radar sensors to avoid false alarms. This makes the output signal more stable and reliable, avoids false triggering caused by sensor characteristics or environmental electromagnetic interference, and improves the system's anti-interference capability.

[0075] In some embodiments, cost-effective sensors and controllers are used to achieve a low-cost system, while performance is improved through algorithms.

[0076] In some embodiments, machine interaction and log display are achieved through an OLED interface, which integrates a large amount of status information, improves compatibility and interaction, and allows maintenance personnel to quickly diagnose the system status on-site even without connecting to a mobile phone or computer.

[0077] The technical solution of the present invention will be described below through specific application examples.

[0078] System hardware example: Hardware configuration:

[0079] Main controller: ESP32-S3 chip (dual-core processor, low power consumption).

[0080] Sensor array:

[0081] Dual-channel PIR motion sensor (detection distance 6m, response time 0.5s), microwave radar module (detection range 11m, supports micro-motion detection), MLX90614 infrared temperature sensor (accuracy ±0.5℃, non-contact measurement), and temperature and humidity sensor (environmental parameter monitoring).

[0082] Communication module: WiFi communication, supports offline operation.

[0083] Storage unit: MicroSD card (maximum support 32GB, log rotation management).

[0084] Alarm unit: audible and visual alarm + status indicator light.

[0085] Power management: Wide voltage input (5-24V), overvoltage and overcurrent protection.

[0086] Software Algorithm Example: Core Code Example of a Hierarchical Verification Algorithm

[0087] class HierarchicalFusionSystem:

[0088] def __init__(self):

[0089] self.weights = [X1,X2,X3,X4] # PIR, radar, temperature, time-series weights

[0090] def process_intrusion_detection(self, sensor_data):

[0091] # First layer: PIR motion detection

[0092] pir_score = self.pir_motion_analysis(sensor_data['pir'])

[0093] # Second layer: Radar presence confirmed

[0094] radar_score = self.radar_presence_analysis(sensor_data['radar'])

[0095] # Third Layer: Temperature Bio-verification

[0096] temp_score=self.temperature_biological_analysis(sensor_data['temperature'])

[0097] # Fourth layer: Spatiotemporal correlation analysis

[0098] temporal_score = self.temporal_correlation_analysis(sensor_data)

[0099] # Integrated Decision

[0100] total_confidence = (self.weights[0] * pir_score +

[0101] self.weights[1] * radar_score +

[0102] self.weights[2] * temp_score +

[0103] self.weights[3] * temporal_score)

[0104] return self.execute_alert_strategy(total_confidence)

[0105] Fault recovery examples

[0106] Automatic recovery process implementation:

[0107] 1. Health monitoring: Periodic sensor status checks.

[0108] 2. Anomaly Diagnosis: Intelligent diagnosis based on historical data and thresholds.

[0109] 3. Recovery strategies: Soft reset attempt (communication module, sensor), parameter recalibration (temperature compensation, sensitivity adjustment), software reset (watchdog trigger).

[0110] 4. Status Verification: Functional testing and status confirmation after recovery.

[0111] 5. Log recording: Detailed records of fault information and recovery process.

[0112] 6. The technical solution of the present invention will be verified and explained through specific experimental examples below.

[0113] Experimental environment

[0114] Test sites: substations in different geographical locations

[0115] Testing period: 6 months (covering different seasons)

[0116] Test subjects: common small animals such as mice, snakes, and birds.

[0117] Table 1 Test Results

[0118]

[0119] Actual testing has verified that this invention has the following significant advantages over existing technologies.

[0120] 1. Significantly improved detection accuracy: Detection accuracy rate: estimated to increase from 70% to over 95%; False alarm rate: estimated to decrease from 25-35% to below 5%; achieving reliable monitoring around the clock.

[0121] 2. Significantly enhanced system reliability: Mean time between failures (MTBF): estimated to exceed 2000 hours; Automatic fault recovery success rate: over 90%; Continuous operation capability in offline mode.

[0122] 3. Deep optimization of intelligence: multi-level intelligent early warning, accurate response; adaptive to environmental changes; remote diagnosis and parameter optimization.

[0123] 4. Significantly reduced maintenance costs: It is estimated that the frequency of manual inspections will be reduced by more than 50%; automatic calibration reduces the workload of maintenance; remote upgrades and diagnostic support are available.

[0124] 5. Significant economic benefits: Reduces equipment failure rate; reduces power outage losses; improves power supply reliability.

[0125] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A smart detection system for small animal intrusion in a substation, characterized in that, The system includes: A multi-sensor detection module, wherein the multi-sensor detection module includes at least: a motion detection sensor, a microwave radar, an infrared temperature sensor, and an ambient temperature and humidity detector; and The main controller is connected to the multi-sensor detection module and is used to receive multi-source sensor detection data output by the multi-sensor detection module, and calculate the total confidence score based on each sensor detection data and its respective weight in the multi-source sensor detection data, so as to determine whether it is a small animal invasion based on the total confidence score.

2. The system according to claim 1, characterized in that, The main controller is further configured to: calculate the total confidence score based on each sensor detection data and its respective weight in the multi-source sensor detection data, including: Based on the sensor detection data, the data collected by the motion detection sensor, the microwave radar, the infrared temperature sensor, and the ambient temperature and humidity detector are acquired, and spatiotemporal alignment is performed on the data collected by the motion detection sensor, the microwave radar, the infrared temperature sensor, and the ambient temperature and humidity detector. Based on the data collected by the environmental temperature and humidity detector, and using an environmental temperature and humidity compensation algorithm, the confidence levels of the data collected by the motion detection sensor, the microwave radar, and the infrared temperature sensor are calculated after spatiotemporal alignment; and The total confidence level is calculated based on the confidence level and weight of the data collected by the motion detection sensor, the confidence level and weight of the data collected by the microwave radar, the confidence level and weight of the data collected by the infrared temperature sensor, and the temporal correlation and weight of the triggering of the motion detection sensor, the microwave radar and the infrared temperature sensor.

3. The system according to claim 2, characterized in that, The timing association is defined as the simultaneous triggering of all sensors in the multi-sensor detection module. Specifically, if at least two sensors in the multi-sensor detection module are triggered simultaneously within the same detection period, the timing association is 1; or if fewer than two sensors in the multi-sensor detection module are triggered simultaneously within the same detection period, or if no sensor is triggered, the timing association is 0.

4. The system according to claim 1, characterized in that, The system also includes a watchdog monitoring subsystem to prevent the program from crashing or entering an infinite loop.

5. A smart detection method for small animal intrusion in substations, characterized in that, The method includes: Multi-source sensor data is acquired through a multi-sensor detection module, wherein the multi-sensor detection module includes at least: a motion detection sensor, a microwave radar, an infrared temperature sensor, and an environmental temperature and humidity detector; and The main controller receives multi-source sensor detection data output by the multi-sensor detection module, and calculates the total confidence score based on each sensor detection data and its respective weight in the multi-source sensor detection data, so as to determine whether it is a small animal invasion based on the total confidence score.

6. The method according to claim 5, characterized in that, The overall confidence score is calculated based on each sensor detection data and its respective weight in the multi-source sensor detection data, including: Based on the sensor detection data, the data collected by the motion detection sensor, the microwave radar, the infrared temperature sensor, and the ambient temperature and humidity detector are acquired, and spatiotemporal alignment is performed on the data collected by the motion detection sensor, the microwave radar, the infrared temperature sensor, and the ambient temperature and humidity detector. Based on the data collected by the environmental temperature and humidity detector, and using an environmental temperature and humidity compensation algorithm, the confidence levels of the data collected by the motion detection sensor, the microwave radar, and the infrared temperature sensor are calculated after spatiotemporal alignment; and The total confidence level is calculated based on the confidence level and weight of the data collected by the motion detection sensor, the confidence level and weight of the data collected by the microwave radar, the confidence level and weight of the data collected by the infrared temperature sensor, and the temporal correlation and weight of the triggering of the motion detection sensor, the microwave radar and the infrared temperature sensor.

7. The method according to claim 6, characterized in that, The timing association is defined as the simultaneous triggering of all sensors in the multi-sensor detection module. Specifically, if at least two sensors in the multi-sensor detection module are triggered simultaneously within the same detection period, the timing association is 1; or if fewer than two sensors in the multi-sensor detection module are triggered simultaneously within the same detection period, or if no sensor is triggered, the timing association is 0.

8. The method according to claim 5, characterized in that, The method also includes: using a watchdog monitoring subsystem to prevent the program from crashing or entering an infinite loop.

9. The method according to claim 5, characterized in that, The method further includes: performing fault diagnosis and self-recovery on the sensor, wherein the fault diagnosis and self-recovery includes: Each sensor in the multi-sensor detection module is tested to determine whether any sensor is faulty. If a sensor malfunction is detected, the malfunctioning sensor will be restored. If the faulty sensor is successfully restored, a restoration log is recorded; if the faulty sensor is not successfully restored, the fault status is recorded.

10. The method according to claim 9, characterized in that, The method further includes: remotely configuring the parameters of each sensor in the multi-sensor detection module, and updating the parameters of each sensor based on the remote parameter configuration.

Images