CN-121435081-B - Intelligent fusion terminal detection method and system based on AI identification

CN121435081BCN 121435081 BCN121435081 BCN 121435081BCN-121435081-B

Abstract

The application relates to an intelligent fusion terminal detection method and system based on AI identification, and relates to the technical field of power system detection, comprising the steps of collecting the sensor state of a preset sensor array; the method comprises the steps of detecting a preset intelligent fusion terminal according to a sensor state control sensor array to generate terminal detection parameters, inputting the terminal detection parameters into a preset state identification model to identify the terminal detection parameters to generate a terminal state identification result, judging whether the terminal state identification result is consistent with a preset abnormal state-free result, if so, continuing to control the sensor array to detect the preset intelligent fusion terminal to generate the terminal detection parameters, and if not, giving an abnormal state alarm. The method and the device have the effect of improving the detection accuracy of the intelligent fusion terminal.

Inventors

BAO YIFENG
SHEN JINGYE
YU ZHONGJIAN
Rao Huipan

Assignees

杭州华罡智能科技有限公司

Dates

Publication Date: 20260508
Application Date: 20251229

Claims (5)

1. The intelligent fusion terminal detection method based on AI identification is characterized by comprising the following steps: collecting a sensor state of a preset sensor array; detecting a preset intelligent fusion terminal according to a sensor state control sensor array to generate terminal detection parameters, wherein the terminal detection parameters comprise terminal self detection parameters and terminal environment detection parameters; Inputting the terminal detection parameters into a preset state identification model for identification so as to generate a terminal state identification result; Judging whether the terminal state identification result is consistent with a preset abnormal state-free result; If yes, continuing to control the sensor array to detect a preset intelligent fusion terminal so as to generate terminal detection parameters; if not, alarming in abnormal state; The step of detecting the preset intelligent fusion terminal according to the sensor state control sensor array to generate terminal detection parameters comprises the following steps: Judging whether the sensor state is consistent with a preset sensor fault state or not; If not, the sensor array is controlled to detect the intelligent fusion terminal so as to generate terminal detection parameters; If yes, controlling the sub-sensors in the sensor array to detect the intelligent fusion terminal so as to generate terminal detection sub-data; collecting a sensor fault type of a sensor array and preset adjacent detection parameters of an adjacent sensor array; analyzing the terminal detection sub-data, the sensor fault type and the adjacent detection parameters to determine terminal detection parameters; The step of analyzing the terminal detection sub-data, the sensor fault type and the adjacent detection parameters to determine the terminal detection parameters includes: Judging whether the sensor fault type is consistent with a preset single environment sensor fault type or not; if the sensor faults are inconsistent, performing sensor fault alarm; If the detection parameters are consistent with the terminal detection sub-data, analyzing the adjacent detection parameters and the terminal detection sub-data to determine fault sensor data; Associating the fault sensor data with the terminal detection sub-data to generate a terminal detection parameter; The step of analyzing the adjacent detection parameters and the terminal detection sub-data to determine faulty sensor data includes: calling a terminal to detect environment sub-data in the sub-data; Invoking adjacent environment data which is consistent with the category of the environment sub-data in the adjacent detection parameters and replacement environment data which is inconsistent with the category of the environment sub-data in the adjacent detection parameters; calculating the similarity of the environment sub-data and the adjacent environment data to generate an adjacent environment similarity score; Judging whether the similarity scores of adjacent environments are consistent with the preset environment consistency scores or not; If yes, defining the replacement environment data as fault sensor data; if not, analyzing the environment sub-data and the replacement environment data to generate fault sensor data.
2. The AI-recognition-based intelligent fusion terminal detection method of claim 1, wherein the step of calculating the similarity of the environment sub-data and the neighboring environment data to generate the neighboring environment similarity score comprises: correcting the environment sub-data and the adjacent environment data according to a preset threshold correction rule to generate environment correction sub-data and adjacent environment correction data; judging whether the environment correction sub-data is consistent with the adjacent environment correction data or not; If the difference similarity scores are inconsistent, defining the preset difference similarity scores as adjacent environment similarity scores; If so, defining the preset consistent similarity score as the adjacent environment similarity score.
3. The AI-recognition-based intelligent fusion terminal detection method of claim 1, wherein analyzing the environment sub-data and the replacement environment data to generate the faulty sensor data includes: analyzing the sensitivity of the environmental sub-data to generate sub-data range constraint values; respectively acquiring a first reference distance and a second reference distance between the sensor array and a preset first reference point and between the adjacent sensor array and a preset first reference point, wherein the first reference distance is smaller than the second reference distance; Calculating a first reference distance and a second reference distance according to a preset distance weight formula to generate local data weight and influence data weight; the sub-data range constraint values and the replacement context data are weighted summed according to the local data weights and the impact data weights to generate faulty sensor data.
4. The AI-recognition-based intelligent fusion terminal detection method of claim 3, wherein analyzing the sensitivity of the environmental sub-data to generate the sub-data range constraint value comprises: according to the environment sub-data, a corresponding data sensitivity weight is found out from a preset environment data sensitivity relation; weighting and summing the environment sub-data according to the data sensitivity weight to generate a basic range constraint value; calculating the change rate of the environment sub-data to generate an environment data change rate; searching a corresponding data change constraint coefficient in a preset environment data change relation according to the environment data change rate; The product of the data change constraint coefficients and the base range constraint values is calculated to generate sub-data range constraint values.
5. An AI-recognition-based detection system, comprising: the acquisition module is used for acquiring the state of the sensor; A memory for storing a program of the AI-recognition-based intelligent fusion terminal detection method according to any one of claims 1 to 4; A processor, a program in a memory can be loaded and executed by the processor and implement the AI-recognition-based intelligent fusion terminal detection method according to any one of claims 1 to 4.

Description

Intelligent fusion terminal detection method and system based on AI identification Technical Field The application relates to the technical field of power system detection, in particular to an intelligent fusion terminal detection method and system based on AI identification. Background The inspection of the power system is a very important operation and maintenance link, and aims to ensure the stability of the last kilometer of the power distribution network. In the related art, in the inspection of a power system, generally, the inspection personnel detect parameters of the intelligent fusion terminal of the power system, such as voltage and current, and parameters of the environment where the intelligent fusion terminal is located, such as temperature, humidity, smoke and the like, then compare the detected parameters with a set threshold value, and after exceeding the threshold value, the inspection personnel mark the abnormal intelligent fusion terminal. Aiming at the related technology, the intelligent fusion terminal relies on manpower inspection, a large amount of manpower resources are needed, the manual inspection is time-consuming and labor-consuming, missed inspection and misjudgment are easy to be caused by fatigue, distraction and the like, long-time correct detection is difficult to ensure, the detection accuracy of the intelligent fusion terminal is low, and the improvement is still provided. Disclosure of Invention In order to improve the detection accuracy of the intelligent fusion terminal, the application provides an intelligent fusion terminal detection method and system based on AI identification. In a first aspect, the present application provides an intelligent fusion terminal detection method based on AI identification, which adopts the following technical scheme: the intelligent fusion terminal detection method based on AI identification comprises collecting sensor state of a preset sensor array; detecting a preset intelligent fusion terminal according to a sensor state control sensor array to generate terminal detection parameters, wherein the terminal detection parameters comprise terminal self detection parameters and terminal environment detection parameters; Inputting the terminal detection parameters into a preset state identification model for identification so as to generate a terminal state identification result; Judging whether the terminal state identification result is consistent with a preset abnormal state-free result; If yes, continuing to control the sensor array to detect a preset intelligent fusion terminal so as to generate terminal detection parameters; If not, the abnormal state alarm is carried out. Optionally, the step of detecting the preset intelligent fusion terminal by controlling the sensor array according to the sensor state to generate the terminal detection parameter includes: Judging whether the sensor state is consistent with a preset sensor fault state or not; If not, the sensor array is controlled to detect the intelligent fusion terminal so as to generate terminal detection parameters; If yes, controlling the sub-sensors in the sensor array to detect the intelligent fusion terminal so as to generate terminal detection sub-data; collecting a sensor fault type of a sensor array and preset adjacent detection parameters of an adjacent sensor array; and analyzing the terminal detection sub-data, the sensor fault type and the adjacent detection parameters to determine the terminal detection parameters. Optionally, the step of analyzing the terminal detection sub-data, the sensor fault type and the adjacent detection parameter to determine the terminal detection parameter includes: Judging whether the sensor fault type is consistent with a preset single environment sensor fault type or not; if the sensor faults are inconsistent, performing sensor fault alarm; If the detection parameters are consistent with the terminal detection sub-data, analyzing the adjacent detection parameters and the terminal detection sub-data to determine fault sensor data; the fault sensor data and the terminal detection sub-data are correlated to generate a terminal detection parameter. Optionally, the step of analyzing the adjacent detection parameter and the terminal detection sub-data to determine the fault sensor data includes: calling a terminal to detect environment sub-data in the sub-data; Invoking adjacent environment data which is consistent with the category of the environment sub-data in the adjacent detection parameters and replacement environment data which is inconsistent with the category of the environment sub-data in the adjacent detection parameters; calculating the similarity of the environment sub-data and the adjacent environment data to generate an adjacent environment similarity score; Judging whether the similarity scores of adjacent environments are consistent with the preset environment consistency scores or not; If yes, defining the replacement envi