CN-121563020-B - Team emergency management method and system for inspection robot of power plant

CN121563020BCN 121563020 BCN121563020 BCN 121563020BCN-121563020-B

Abstract

The application relates to the technical field of power plant inspection, in particular to a shift emergency management method for an inspection robot of a power plant. The method comprises the steps of collecting different kinds of data when each device of a power plant runs, obtaining adjacent data of each data, obtaining weight of each adjacent data and characteristic vectors of each data, further calculating characteristic distance between every two data, obtaining representative data in the data, obtaining average distance of each data, then calculating deviation degree of the data at the current moment, obtaining current abnormal coefficients of the device, further obtaining routing inspection priority coefficients of the device, obtaining adaptation degree of the routing inspection robot and the device based on the shortest route distance of the routing inspection robot from the device and the routing inspection priority coefficients of the device, and distributing the device, which needs routing inspection, of each routing inspection robot based on the adaptation degree. The application can improve the inspection efficiency of the inspection robot on the power plant.

Inventors

JI YUN
SUN XUEQIAN
ZHANG WENQI
HAN SHAOZU
RAO MUJIN
ZHANG HAICHUAN
DONG CHENGCHENG
CHEN JING
KONG HUAYONG
LI DONG
WU FEI

Assignees

国能信控技术股份有限公司

Dates

Publication Date: 20260508
Application Date: 20260123

Claims (10)

1. The team-grouping emergency management method for the inspection robot of the power plant is characterized by comprising the following steps of: acquiring different kinds of data of each device of the power plant when in operation, and acquiring adjacent data of one data according to the acquisition frequency of the data and the spectrogram of the data; Acquiring the weight of each adjacent data based on the time interval between one data and each adjacent data corresponding to the data, and acquiring the feature vector of one data based on each adjacent data of the one data, the first-order differential sequences corresponding to all adjacent data and the weight; Calculating the average value of the characteristic distance between one data in the data and each representative data in the data, and recording the average value as the average distance of the data; Calculating the deviation degree of the data at the current moment based on the average distance of each adjacent data of the data at the current moment of the data and the average distance of each adjacent data representing the data; Acquiring a patrol priority coefficient of a device based on the current anomaly coefficient and importance of the device; obtaining the adaptation degree of the inspection robot and the equipment based on the shortest route distance between the inspection robot and the equipment and the inspection priority coefficient of the equipment; the calculating the deviation degree of the data at the current moment based on the average distance of the adjacent data of the data at the current moment of one kind of data and the average distance of the adjacent data of the representative data comprises the following steps: The average distance of each adjacent data of the data at the current moment is recorded as an average distance sequence of the data at the current moment according to time sequence, the slope between every two average distances in the average distance sequence of the data at the current moment is weighted and averaged by utilizing the weight of other adjacent data except the first adjacent data in each adjacent data of the data at the current moment to obtain the corresponding weighted slope of the data at the current moment, the corresponding weighted slope of each representative data in the data is obtained by the same way, the average value of the difference value of the weighted slope of the data at the current moment and the corresponding weighted slope of each representative data in the data is mapped by utilizing an exponential function based on a natural constant to obtain a third mapping value, and the deviation degree of the data at the current moment of the data is obtained by multiplying the third mapping value and the average distance of the data at the current moment; The equipment for distributing each inspection robot to be inspected based on the adaptation degree comprises: And solving the adaptation degree matrix by using a Hungary algorithm to obtain an optimal matching relationship between the inspection robots and the equipment to obtain the equipment which needs inspection of each inspection robot.
2. The method for teamwork emergency management of a power plant inspection robot according to claim 1, wherein the step of obtaining the neighboring data of one of the data according to the collection frequency of the data and the spectrogram of the data comprises: the method comprises the steps of processing all data in one type of data by using fast Fourier transform to obtain a corresponding spectrogram, comparing the acquisition frequency of the data with the frequency corresponding to the maximum position in the spectrogram to obtain a first ratio value, carrying out negative correlation mapping on the difference value between the maximum frequency and the minimum frequency in the spectrogram by using an exponential function based on a natural constant to obtain a first mapping value, multiplying the first ratio value by the first mapping value and rounding up to obtain the data extraction quantity of the data, and obtaining the adjacent data of the data in the data according to the data extraction quantity of the data in the data.
3. The method for team emergency management of inspection robots in power plants according to claim 2, wherein the acquiring the neighboring data of one of the data from the data according to the data extraction amount of the data comprises: The data extraction quantity of one data of the data is used in the data, the data is sequentially collected from the data as a starting point, until the quantity of the collected data is equal to the data extraction quantity of the data, the collection is stopped, and the collected data is the adjacent data of the data.
4. The method for teamwork emergency management of a power plant inspection robot according to claim 1, wherein the step of obtaining the weight of each neighboring data based on the time interval between one data and each neighboring data corresponding thereto comprises: and comparing the second mapping value of one adjacent data of the data with the sum of the second mapping values of all adjacent data of the data to obtain the weight of the adjacent data.
5. The method for managing the teams and groups of the power plant inspection robots according to claim 1, wherein the step of obtaining the feature vector of one data based on each adjacent data of the one data and the first order differential sequences and weights corresponding to all the adjacent data comprises the steps of: The method comprises the steps of obtaining a first differential average value of all adjacent data by using the weight of other adjacent data except for a first adjacent data in each adjacent data of one data, obtaining a weighted data average value of each adjacent data by using the weight of each adjacent data of one data, obtaining a weighted standard deviation of each adjacent data by using the weight of each adjacent data of one data, obtaining a variation coefficient by using the weighted standard deviation and the weighted data average value, and recording the variation coefficient as the weighted variation coefficient of the data, wherein the first differential average value, the data average value and the weighted variation coefficient of the data form a characteristic vector of the data.
6. The method for team-based emergency management of a power plant inspection robot according to claim 1, wherein said obtaining representative data of the data comprises: And obtaining the data with the maximum representative index from the data as the representative data by carrying out negative correlation mapping on the average value of the characteristic distance between one data and other data in the data by using an exponential function based on a natural constant.
7. The method for team emergency management of inspection robots in power plants according to claim 1, wherein the step of obtaining the current anomaly coefficient of a device according to the deviation degree of the data of the current moment of the various data of the device comprises the following steps: The method comprises the steps of normalizing the deviation degree of data at the current moment of various data corresponding to the equipment to form a data set, recording the data set as a deviation degree data set, weighting and averaging all elements of the deviation degree data set by using the weight of various data to obtain the average deviation degree, and multiplying the average deviation degree, the maximum value in the deviation degree data set and the weight of one data corresponding to the maximum value to obtain the current anomaly coefficient of the equipment.
8. The method for teamwork emergency management of a power plant inspection robot according to claim 1, wherein the step of obtaining the inspection priority coefficient of a device based on the current abnormality coefficient and importance of the device comprises: multiplying the current abnormal coefficient of the equipment by the importance of the equipment to obtain the patrol priority coefficient of the equipment.
9. The method for teamwork emergency management of power plant inspection robots according to claim 1, wherein the obtaining the fitness of an inspection robot to a device based on the shortest route distance of the inspection robot from the device and the inspection priority coefficient of the device comprises: The method comprises the steps of respectively normalizing the shortest route distance between each inspection robot and each device and the inspection priority coefficient of each device to obtain a normalized shortest route distance and a normalized inspection priority coefficient, subtracting the normalized shortest route distance between one inspection robot and one device from a first preset value to obtain a distance characteristic coefficient, and carrying out weighted summation on the distance characteristic coefficient and the normalized inspection priority coefficient of the device to obtain the fitness of the inspection robot and the device.
10. A power plant inspection robot shift emergency management system comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the computer program when executed by the processor performs the steps of a power plant inspection robot shift emergency management method according to any one of claims 1 to 9.

Description

Team emergency management method and system for inspection robot of power plant Technical Field The invention relates to the technical field of power plant inspection, in particular to a shift emergency management method and system for an inspection robot of a power plant. Background With the ever-expanding scale of the power industry and the increasing complexity of the operating environment, power plants place higher demands on safety, reliability and response speed in daily operation and emergency handling. In recent years, the development of the Internet of things and the inspection robot technology promotes the transformation of the power plant inspection mode, and the inspection robot gradually replaces part of manual inspection tasks. Compared with the traditional manual mode, the inspection robot has the advantages that the problems of low efficiency, high labor intensity, slow emergency response and the like are solved to a certain extent. The existing inspection robot relies on a fixed route and preset rules to work, lacks a linkage mechanism with the real-time running state of equipment, and is difficult to dynamically adjust the inspection path and task allocation according to the equipment load condition. For example, when a certain generator set has abnormal temperature or vibration aggravated, the inspection robot still can inspect according to the original route, and the priority cannot be adjusted in time to perform important inspection, so that the potential abnormal discovery time is very easy to miss, and the problem is delayed to be exposed. Therefore, how to realize intelligent task scheduling based on equipment state and the like, improve inspection efficiency and the like becomes a key problem to be solved urgently for power plant inspection and emergency management. Disclosure of Invention In order to solve the technical problems, the invention aims to provide a team emergency management method and system for a power plant inspection robot, and the adopted technical scheme is as follows: In a first aspect, an embodiment of the present invention provides a method for teaming emergency management of a power plant inspection robot, including: acquiring different kinds of data of each device of the power plant when in operation, and acquiring adjacent data of one data according to the acquisition frequency of the data and the spectrogram of the data; Acquiring the weight of each adjacent data based on the time interval between one data and each adjacent data corresponding to the data, and acquiring the feature vector of one data based on each adjacent data of the one data, the first-order differential sequences corresponding to all adjacent data and the weight; Calculating the average value of the characteristic distance between one data in the data and each representative data in the data, and recording the average value as the average distance of the data; Calculating the deviation degree of the data at the current moment based on the average distance of each adjacent data of the data at the current moment of the data and the average distance of each adjacent data representing the data; The method comprises the steps of obtaining a current abnormal coefficient and importance of equipment, obtaining a routing inspection priority coefficient of the equipment based on the current abnormal coefficient and importance of the equipment, obtaining the adaptation degree of the routing inspection robot and the equipment based on the shortest route distance of the routing inspection robot from the equipment and the routing inspection priority coefficient of the equipment, and distributing the equipment which needs routing inspection of each routing inspection robot based on the adaptation degree. Preferably, acquiring the neighboring data of one of the data according to the acquisition frequency of the data and the spectrogram of the data includes: the method comprises the steps of processing all data in one type of data by using fast Fourier transform to obtain a corresponding spectrogram, comparing the acquisition frequency of the data with the frequency corresponding to the maximum position in the spectrogram to obtain a first ratio value, carrying out negative correlation mapping on the difference value between the maximum frequency and the minimum frequency in the spectrogram by using an exponential function based on a natural constant to obtain a first mapping value, multiplying the first ratio value by the first mapping value and rounding up to obtain the data extraction quantity of the data, and obtaining the adjacent data of the data in the data according to the data extraction quantity of the data in the data. Preferably, acquiring the neighboring data of one of the data in the data according to the data extraction amount of the data includes: The data extraction quantity of one data of the data is used in the data, the data is sequentially collected from the data as a starting point,