CN112766251A - Infrared detection method and system for power transformation equipment, storage medium and computer equipment - Google Patents
Infrared detection method and system for power transformation equipment, storage medium and computer equipment Download PDFInfo
- Publication number
- CN112766251A CN112766251A CN202011630610.1A CN202011630610A CN112766251A CN 112766251 A CN112766251 A CN 112766251A CN 202011630610 A CN202011630610 A CN 202011630610A CN 112766251 A CN112766251 A CN 112766251A
- Authority
- CN
- China
- Prior art keywords
- equipment
- infrared
- image
- power transformation
- infrared image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/10—Image acquisition
- G06V10/12—Details of acquisition arrangements; Constructional details thereof
- G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
- G06V10/143—Sensing or illuminating at different wavelengths
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
- Radiation Pyrometers (AREA)
Abstract
The invention provides a transformer equipment infrared detection method, a transformer equipment infrared detection system, a storage medium and computer equipment aiming at the technical problem that misjudgment is easily caused by transformer equipment infrared detection in the prior art, wherein the real infrared image characteristics of the transformer equipment are accurately extracted through an image characteristic extraction model based on a convolutional neural network, whether the equipment is abnormal or not is accurately judged, the abnormal infrared image can be diagnosed through matching of an expert library, the reason of the abnormality is analyzed, and the load current of the equipment can be predicted according to the infrared image characteristics and the temperature and the humidity of the environment; and finally, an infrared temperature measurement report is automatically generated, so that the workload of testing personnel is reduced, the working efficiency is improved, and an accurate and reliable detection result is obtained.
Description
Technical Field
The invention relates to the technical field of power grid equipment detection, in particular to detection of primary power transformation equipment, and more particularly relates to an infrared detection method and system of power transformation equipment, a storage medium and computer equipment.
Background
The primary power transformation equipment, such as a transformer, a high-voltage circuit breaker, a disconnecting link and the like, is used as an important component of a power system, and the safe and reliable operation of the primary power transformation equipment is of great significance for guaranteeing the stable power supply of a power grid. While periodic preventive tests are performed on primary equipment, infrared temperature measurement is required to monitor the state of the equipment. However, in the process of shooting the infrared imaging spectrogram of the device, the infrared imaging spectrogram is easily affected by a complex environment, shooting angles, shooting distances and light intensity, and if the problems are not well processed in the shooting process, the subsequent detection result can be misjudged.
The Chinese patent application with the publication number of CN106597185B is published on the publication date 2019-03-19: a transformer substation primary equipment infrared analysis system and a method thereof disclose a scheme for integrating the existing thermal infrared imagers, intensively and uniformly managing all the thermal infrared imagers in a transformer substation by using one system and unifying infrared thermal image formats. However, this solution and other prior arts still fail to solve the above technical problems.
Disclosure of Invention
Aiming at the limitation of the prior art, the invention provides a transformer equipment infrared detection method, a transformer equipment infrared detection system, a storage medium and computer equipment, and the technical scheme adopted by the invention is as follows:
a transformer equipment infrared detection method comprises the following steps:
acquiring an infrared image of a to-be-detected power transformation device, and environmental temperature data and environmental humidity data when the infrared image of the to-be-detected power transformation device is shot, and extracting image characteristics, a highest temperature value and a lowest temperature value of the infrared image of the to-be-detected power transformation device by using an image characteristic extraction model based on a convolutional neural network;
classifying the infrared image of the to-be-detected power transformation equipment by using an image classification model based on an extreme learning machine according to the image characteristics of the infrared image of the to-be-detected power transformation equipment to obtain the equipment category corresponding to the infrared image of the to-be-detected power transformation equipment;
matching the infrared images of the to-be-detected power transformation equipment by using normal cases and fault cases in the equipment category in a preset expert database, and checking whether the infrared images of the to-be-detected power transformation equipment are normal or not;
according to the maximum temperature value, the minimum temperature value, the environmental temperature data and the environmental humidity data, acquiring a load current value of equipment in the infrared image of the to-be-detected power transformation equipment by using an equipment load current prediction model based on an extreme learning machine;
and generating a corresponding equipment infrared detection report according to the maximum temperature value, the minimum temperature value, the environment temperature data, the environment humidity data, the load current value and the detection result of the infrared image of the to-be-detected power transformation equipment.
Compared with the prior art, the method has the advantages that the real infrared image characteristics of the power transformation equipment are accurately extracted through the image characteristic extraction model based on the convolutional neural network, whether the equipment is abnormal or not is accurately judged, the abnormal infrared image can be diagnosed through matching of the expert database, the reason of the abnormality can be analyzed, and the load current of the equipment can be predicted according to the infrared image characteristics and the temperature and the humidity of the environment; and finally, an infrared temperature measurement report is automatically generated, so that the workload of testing personnel is reduced, the working efficiency is improved, and an accurate and reliable detection result is obtained.
As a preferable scheme, the infrared detection method for the power transformation equipment further comprises the following steps:
and if the detection result of the to-be-detected substation equipment infrared image is normal, adding the to-be-detected substation equipment infrared image into the expert database as a normal case under the equipment category.
Further, the infrared detection method for the power transformation equipment further comprises the following steps:
if the detection result of the to-be-detected substation equipment infrared image is abnormal, acquiring a fault reason according to a matching result of matching the to-be-detected substation equipment infrared image;
and adding the infrared image of the to-be-detected power transformation equipment into the expert database as a fault case under the equipment category.
As a preferred scheme, the convolutional neural network adopts a super-resolution test sequence network.
As a preferred scheme, the image feature extraction model is obtained by performing image feature extraction training on a preset sample data set by a convolutional neural network; the sample data set comprises a front side, a back side, a left side, a right side, a height-changing sleeve and a sample infrared image of the height-changing sleeve, environmental temperature data and environmental humidity data when the sample infrared image is shot and load current of a power transformation device; the sample infrared image is marked with the highest temperature value and the lowest temperature value of equipment in the sample infrared image;
and the convolutional neural network outputs the highest temperature value and the lowest temperature value of the equipment in the sample infrared image after the sample infrared image is trained.
Further, the image classification model is obtained by an extreme learning machine after image classification training is carried out on the image features of the sample infrared images and the expert database.
Further, the equipment load current prediction model is obtained by performing equipment load current prediction training on the maximum temperature value and the minimum temperature value of the equipment in the sample infrared image, the environment temperature data and the environment humidity data when the sample infrared image is shot and the load current of the power transformation equipment by an extreme learning machine.
The present invention also provides the following:
a transformer equipment infrared detection system comprises a data acquisition and processing module, an image classification module, an image detection module, an equipment load current value prediction module and an equipment infrared detection report generation module; the image classification module is connected with the data acquisition processing module; the image detection module is connected with the data acquisition processing module and the image classification module; the equipment load current value prediction module is connected with the data acquisition and processing module; the device infrared detection report generation module is connected with the data acquisition and processing module, the image detection module and the device load current value prediction module; wherein:
the data acquisition and processing module is used for acquiring an infrared image of the to-be-detected power transformation equipment, and environmental temperature data and environmental humidity data when the infrared image of the to-be-detected power transformation equipment is shot, and extracting image characteristics, a highest temperature value and a lowest temperature value of the infrared image of the to-be-detected power transformation equipment by using an image characteristic extraction model based on a convolutional neural network;
the image classification module is used for classifying the infrared images of the to-be-detected power transformation equipment by using an image classification model based on an extreme learning machine according to the image characteristics of the infrared images of the to-be-detected power transformation equipment to obtain the equipment category corresponding to the infrared images of the to-be-detected power transformation equipment;
the image detection module is used for matching the infrared image of the to-be-detected power transformation equipment by using a normal case and a fault case in a preset expert database under the equipment category and checking whether the infrared image of the to-be-detected power transformation equipment is normal or not;
and the equipment load current value prediction module is used for acquiring the load current value of the equipment in the infrared image of the to-be-detected power transformation equipment by using an equipment load current prediction model based on an extreme learning machine according to the highest temperature value, the lowest temperature value, the environmental temperature data and the environmental humidity data.
The equipment infrared detection report generation module is used for generating a corresponding equipment infrared detection report according to the highest temperature value, the lowest temperature value, the environment temperature data, the environment humidity data, the load current value and the detection result of the to-be-detected substation equipment infrared image
A storage medium having stored thereon a computer program, which when executed by a processor, implements the steps of the aforementioned method for infrared detection of a power transformation device.
A computer device comprises a storage medium, a processor and a computer program which is stored in the storage medium and can be executed by the processor, wherein the computer program realizes the steps of the infrared detection method of the power transformation device when being executed by the processor.
Drawings
Fig. 1 is a flowchart illustrating steps of an infrared detection method for a power transformation device according to embodiment 1 of the present invention;
fig. 2 is a flowchart illustrating steps of a method for detecting infrared radiation of a power transformation device according to embodiment 2 of the present invention;
FIG. 3 is a schematic diagram of a super-resolution test sequence network used in an embodiment of the present invention;
fig. 4 is a schematic diagram of an infrared detection system of a power transformation device according to an embodiment of the present invention;
description of reference numerals: 1. a data acquisition processing module; 2. an image classification module; 3. an image detection module; 4. an equipment load current value prediction module; 5. and a device infrared detection report generation module.
Detailed Description
The drawings are for illustrative purposes only and are not to be construed as limiting the patent;
it should be understood that the embodiments described are only some embodiments of the present application, and not all embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments in the present application without any creative effort belong to the protection scope of the embodiments in the present application.
The terminology used in the embodiments of the present application is for the purpose of describing particular embodiments only and is not intended to be limiting of the embodiments of the present application. As used in the examples of this application and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.
When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the application, as detailed in the appended claims. In the description of the present application, it is to be understood that the terms "first," "second," "third," and the like are used solely to distinguish one from another and are not necessarily used to describe a particular order or sequence, nor are they to be construed as indicating or implying relative importance. The specific meaning of the above terms in the present application can be understood by those of ordinary skill in the art as appropriate.
Further, in the description of the present application, "a plurality" means two or more unless otherwise specified. "and/or" describes the association relationship of the associated objects, meaning that there may be three relationships, e.g., a and/or B, which may mean: a exists alone, A and B exist simultaneously, and B exists alone. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. The invention is further illustrated below with reference to the figures and examples.
In order to solve the limitation of the prior art, the present embodiment provides a technical solution, and the technical solution of the present invention is further described below with reference to the accompanying drawings and embodiments.
Referring to fig. 1, an infrared detection method for substation equipment includes the following steps:
s01, acquiring an infrared image of the to-be-detected power transformation equipment, and environmental temperature data and environmental humidity data obtained when the infrared image of the to-be-detected power transformation equipment is shot, and extracting image characteristics, a highest temperature value and a lowest temperature value of the infrared image of the to-be-detected power transformation equipment by using an image characteristic extraction model based on a convolutional neural network;
s02, classifying the infrared images of the to-be-detected power transformation equipment by using an image classification model based on an extreme learning machine according to the image characteristics of the infrared images of the to-be-detected power transformation equipment to obtain the equipment category corresponding to the infrared images of the to-be-detected power transformation equipment;
s03, matching the infrared image of the to-be-detected power transformation equipment by using a normal case and a fault case in a preset expert library under the equipment category, and checking whether the infrared image of the to-be-detected power transformation equipment is normal;
s04, obtaining a load current value of equipment in the infrared image of the to-be-detected power transformation equipment by using an equipment load current prediction model based on an extreme learning machine according to the highest temperature value, the lowest temperature value, the environmental temperature data and the environmental humidity data;
and S05, generating a corresponding equipment infrared detection report according to the highest temperature value, the lowest temperature value, the environment temperature data, the environment humidity data, the load current value and the detection result of the infrared image of the to-be-detected substation equipment.
Compared with the prior art, the method has the advantages that the real infrared image characteristics of the power transformation equipment are accurately extracted through the image characteristic extraction model based on the convolutional neural network, whether the equipment is abnormal or not is accurately judged, the abnormal infrared image can be diagnosed through matching of the expert database, the reason of the abnormality can be analyzed, and the load current of the equipment can be predicted according to the infrared image characteristics and the temperature and the humidity of the environment; and finally, an infrared temperature measurement report is automatically generated, so that the workload of testing personnel is reduced, the working efficiency is improved, and an accurate and reliable detection result is obtained.
Specifically, the infrared image of the power transformation equipment is influenced by a complex environment, shooting angles, shooting distances and light intensity in the shooting process, the problems are not well solved in the shooting process, and misjudgment of results can be caused. In the embodiment, the infrared image is subjected to feature extraction by using a deep learning algorithm, and the algorithm can perform multi-level feature learning on the original infrared image.
Convolutional Neural Networks (CNN) are a class of feed forward neural networks that contain convolutional computations and have a deep structure, and are one of the typical algorithms of deep learning algorithms. The convolutional layer of CNN is subjected to feature extraction and mapping through a convolutional kernel, the shallow layer convolutional layer extracts low-level features such as edges, lines and corners, and the higher layer convolutional layer extracts higher-level features. Assuming that an input feature matrix of a previous layer of the CNN network is T ∈ n × n, a learnable convolution kernel of the layer is K ∈ K × K, and an output feature matrix of the layer obtained after the convolved data passes through an activation function is shown as the following formula:
wherein l represents the first convolutional layer, K represents the convolutional kernel, Bia represents the offset matrix,
the output characteristic of the l-th layer is represented,
representing the input features of the l-th layer. f (-) represents an activation function, typical CNN activation functions include eLU function, lreuu function, Sigmoid function, etc., in order to improve the performance of convolutional layer feature extraction and the adaptability to input images, the present case proposes a mixed activation function, which is composed of linear and nonlinear activation functions, and the expression is shown in the following formula):
fmix(x)＝ρflrelu(x)+(1-ρ)felu(x)；
wherein rho belongs to [0,1] as a mixing coefficient, and represents the weight occupied by each activation function.
The essence of the pooling layer is that the down-sampling, which follows the convolutional layer, is also composed of a plurality of feature planes, each of which uniquely corresponds to a feature plane of the layer above it, without changing the number of feature planes. The pooling layer aims to obtain the features with space invariance by reducing the resolution of the feature plane, and plays a role of secondary feature extraction, and each neuron of the pooling layer performs pooling operation on a local receptive field. The pooling operation is calculated as shown in the following equation:
wherein l represents the first convolutional layer, K represents the convolutional kernel, Bia represents the offset matrix,
the output characteristic of the l-th layer is represented,
representing the input characteristics of the l layer, k is the size of the pooling kernel, and down (·) is the maximum pooling method, and the expression is as follows:
down(x)＝max(x)；
where max represents the maximum value.
In this embodiment, in step S02, the infrared image is classified by using a wavelet extreme learning machine (wellm) method. An Extreme Learning Machine (ELM) is a single hidden layer forward neural network, and is composed of an input layer, a hidden layer and an output layer. The ELM has the biggest characteristic that the weight from the input layer to the hidden layer and the bias matrix of the hidden layer are randomly assigned, so that model calculation is greatly simplified, and the ELM is widely applied to the fields of regression analysis and image classification.
Given a training data set of N different samples
Wherein xi＝[xi1,xi2,Λ,xim]∈RmAs an input vector, yiAre labeled with the corresponding category. For the packet containing nhAn extreme learning machine mathematical model with an activation function g (x) can be expressed as:
wherein wi＝[wi1,wi2,Λ,wim]TTo connect the ith hidden layer neuron to the weights of the input layer,
is the deviation of the ith implicit node. Beta is ai＝[βi1,βi2,Λ,βim]TAre weights connecting the ith hidden layer neuron and the output layer.
The extreme learning machine mathematical model with activation function g (x) can be in the form of a matrix, as shown in the following equation:
Hβ＝Y；
in the formula (I), the compound is shown in the specification,
in order to have a hidden layer output matrix,
Solving the output weight value is to ensure that the loss function obtains the minimum value, and introduce an adjusting coefficient C to balance the training error and the output weight value, wherein the target loss function is as follows:
e＝Y-Hβ；
constructing the lagrange equation yields:
wherein alpha isi(i ═ 1,2, Λ, N) denotes lagrangian.
Respectively solving the partial derivatives of all variables in the Lagrange equation and making the partial derivative function be zero to obtain the following formulas:
from this, the mathematical expression of the output weight matrix can be derived as follows:
in the formula, I represents an identity matrix.
In this embodiment, the activation function g (x) is a discrete wavelet function, and its expression is as follows:
wherein eta, mu and theta are parameters of the wavelet kernel function, and the correct selection of the sizes of the three parameters has important significance for improving the classification accuracy.
From the above analysis, an extreme learning machine-based image classification model can be obtained:
in step S02, when a new image x is inputiThe category to which the image belongs can be predicted according to the above formula.
Specifically, in step S04, obtaining the load current value of the device in the infrared image of the substation device to be detected is still implemented by using a wavelet kernel function-based extreme learning machine (WELM) method. The input of the model is an infrared image Ima and the highest temperature T of the equipmentmaxAnd a minimum temperature TminAverage temperature TmeanTemperature of the environment TenvAnd humidity HenvThe model input [ Ima, T ] is represented in the form of a matrixmax,Tmin,Tmean,Tevn,Hevn]The output of the model is the load current value I of the device. The implementation of the palm model in step S04 is similar to the image classification in step S02, and is not described here again.
Specifically, in step S05, when the device infrared detection report is generated, the maximum temperature value, the minimum temperature value, the load current value, and the to-be-detected substation device infrared image are automatically filled into corresponding positions in the template according to a preset template.
As a preferred embodiment, please refer to fig. 2, which further includes the following steps:
and S031, if the inspection result of the to-be-inspected substation equipment infrared image is normal, adding the to-be-inspected substation equipment infrared image to the expert database as a normal case under the equipment category.
As a preferred embodiment, the method further comprises the following steps:
s032, if the detection result of the infrared image of the to-be-detected power transformation equipment is abnormal, acquiring a fault reason according to a matching result of matching the infrared image of the to-be-detected power transformation equipment;
and S033, adding the infrared image of the to-be-detected power transformation equipment to the expert database as a fault case under the equipment category.
Specifically, in step S03, the infrared diagnosis of the electrical equipment is performed by using "infrared diagnosis standard for live equipment" (DL/T664-2016) as a determination rule, and a normal case and a file of a fault case are established according to the equipment abnormal phenomenon found by the previous infrared temperature measurement, and the method for objectively analyzing the infrared temperature measurement abnormal data by combining with the daily experience of a tester to accurately evaluate the health condition of the equipment is implemented. In addition, the expert database containing normal cases and fault cases is formed by using the infrared temperature measurement images of the conventional equipment, and the integrity of the expert database is ensured by adaptively updating the expert database through the steps S031 and S033.
As a preferred embodiment, the convolutional neural network in step S01 employs a super-resolution test sequence network.
The super-resolution test sequence (VGG) network is shown in fig. 3, and has a total of 19 layers, including an input layer, a pooling layer, a convolutional layer, and an output layer.
As a preferred embodiment, the image feature extraction model is obtained by performing image feature extraction training on a preset sample data set by a convolutional neural network; the sample data set comprises a front side, a back side, a left side, a right side, a height-changing sleeve and a sample infrared image of the height-changing sleeve, environmental temperature data and environmental humidity data when the sample infrared image is shot and load current of a power transformation device; the sample infrared image is marked with the highest temperature value and the lowest temperature value of equipment in the sample infrared image;
and the convolutional neural network outputs the highest temperature value and the lowest temperature value of the equipment in the sample infrared image after the sample infrared image is trained.
Specifically, the size of the sample infrared image is 640 × 480.
Further, the image classification model is obtained by an extreme learning machine after image classification training is carried out on the image features of the sample infrared images and the expert database.
Further, the equipment load current prediction model is obtained by performing equipment load current prediction training on the maximum temperature value and the minimum temperature value of the equipment in the sample infrared image, the environment temperature data and the environment humidity data when the sample infrared image is shot and the load current of the power transformation equipment by an extreme learning machine.
The embodiment further includes the following contents:
referring to fig. 4, the transformer equipment infrared detection system includes a data acquisition processing module 1, an image classification module 2, an image detection module 3, an equipment load current value prediction module 4, and an equipment infrared detection report generation module 5; the image classification module 2 is connected with the data acquisition processing module 1; the image detection module 3 is connected with the data acquisition processing module 1 and the image classification module 2; the equipment load current value prediction module 4 is connected with the data acquisition processing module 1; the device infrared detection report generation module 5 is connected with the data acquisition and processing module 1, the image detection module 3 and the device load current value prediction module 4; wherein:
the data acquisition and processing module 1 is used for acquiring an infrared image of a to-be-detected power transformation device, and environmental temperature data and environmental humidity data when the infrared image of the to-be-detected power transformation device is shot, and extracting image features, a highest temperature value and a lowest temperature value of the infrared image of the to-be-detected power transformation device by using an image feature extraction model based on a convolutional neural network;
the image classification module 2 is used for classifying the infrared images of the to-be-detected power transformation equipment by using an image classification model based on an extreme learning machine according to the image characteristics of the infrared images of the to-be-detected power transformation equipment to obtain the equipment category corresponding to the infrared images of the to-be-detected power transformation equipment;
the image detection module 3 is used for matching the infrared image of the power transformation equipment to be detected by using a normal case and a fault case in the equipment category in a preset expert database, and checking whether the infrared image of the power transformation equipment to be detected is normal;
the equipment load current value prediction module 4 is used for acquiring a load current value of equipment in the infrared image of the to-be-detected power transformation equipment by using an equipment load current prediction model based on an extreme learning machine according to the highest temperature value, the lowest temperature value, the environmental temperature data and the environmental humidity data;
and the equipment infrared detection report generating module 5 is used for generating a corresponding equipment infrared detection report according to the highest temperature value, the lowest temperature value, the environmental temperature data, the environmental humidity data, the load current value and the detection result of the to-be-detected transformer equipment infrared image.
A storage medium having stored thereon a computer program that, when executed by a processor, implements the steps of the power transformation device infrared detection method of embodiment 1.
A computer device comprising a storage medium, a processor, and a computer program stored in the storage medium and executable by the processor, the computer program when executed by the processor implementing the steps of the substation device infrared detection method of embodiment 1.
It should be understood that the above-described embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. Other variations and modifications will be apparent to persons skilled in the art in light of the above description. And are neither required nor exhaustive of all embodiments. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the claims of the present invention.
Claims (10)
1. The infrared detection method for the power transformation equipment is characterized by comprising the following steps of:
s01, acquiring an infrared image of the to-be-detected power transformation equipment, and environmental temperature data and environmental humidity data obtained when the infrared image of the to-be-detected power transformation equipment is shot, and extracting image characteristics, a highest temperature value and a lowest temperature value of the infrared image of the to-be-detected power transformation equipment by using an image characteristic extraction model based on a convolutional neural network;
s02, classifying the infrared images of the to-be-detected power transformation equipment by using an image classification model based on an extreme learning machine according to the image characteristics of the infrared images of the to-be-detected power transformation equipment to obtain the equipment category corresponding to the infrared images of the to-be-detected power transformation equipment;
s03, matching the infrared image of the to-be-detected power transformation equipment by using a normal case and a fault case in a preset expert library under the equipment category, and checking whether the infrared image of the to-be-detected power transformation equipment is normal;
s04, obtaining a load current value of equipment in the infrared image of the to-be-detected power transformation equipment by using an equipment load current prediction model based on an extreme learning machine according to the highest temperature value, the lowest temperature value, the environmental temperature data and the environmental humidity data;
and S05, generating a corresponding equipment infrared detection report according to the highest temperature value, the lowest temperature value, the environment temperature data, the environment humidity data, the load current value and the detection result of the infrared image of the to-be-detected substation equipment.
2. The substation equipment infrared detection method according to claim 1, further comprising the steps of:
and S031, if the inspection result of the to-be-inspected substation equipment infrared image is normal, adding the to-be-inspected substation equipment infrared image to the expert database as a normal case under the equipment category.
3. The substation equipment infrared detection method according to claim 1, further comprising the steps of:
s032, if the detection result of the infrared image of the to-be-detected power transformation equipment is abnormal, acquiring a fault reason according to a matching result of matching the infrared image of the to-be-detected power transformation equipment;
and S033, adding the infrared image of the to-be-detected power transformation equipment to the expert database as a fault case under the equipment category.
4. The infrared detection method for substation equipment as claimed in claim 1, wherein said convolutional neural network in step S01 employs a super-resolution test sequence network.
5. The infrared detection method for the power transformation equipment as recited in claim 1, wherein the image feature extraction model is obtained by performing image feature extraction training on a preset sample data set by a convolutional neural network; the sample data set comprises a front side, a back side, a left side, a right side, a height-changing sleeve and a sample infrared image of the height-changing sleeve, environmental temperature data and environmental humidity data when the sample infrared image is shot and load current of a power transformation device; the sample infrared image is marked with the highest temperature value and the lowest temperature value of equipment in the sample infrared image;
and the convolutional neural network outputs the highest temperature value and the lowest temperature value of the equipment in the sample infrared image after the sample infrared image is trained.
6. The infrared detection method for the power transformation equipment as recited in claim 5, wherein the image classification model is obtained by an extreme learning machine after image classification training of the image features of the sample infrared images and the expert database.
7. A transformation equipment infrared detection method as claimed in claim 5, wherein the equipment load current prediction model is obtained by an extreme learning machine after performing equipment load current prediction training on a highest temperature value and a lowest temperature value of equipment in the sample infrared image, environmental temperature data and environmental humidity data when the sample infrared image is shot, and load current of the transformation equipment.
8. The infrared detection system for the power transformation equipment is characterized by comprising a data acquisition and processing module (1), an image classification module (2), an image detection module (3), an equipment load current value prediction module (4) and an equipment infrared detection report generation module (5); the image classification module (2) is connected with the data acquisition and processing module (1); the image detection module (3) is connected with the data acquisition and processing module (1) and the image classification module (2); the equipment load current value prediction module (4) is connected with the data acquisition and processing module (1); the device infrared detection report generation module (5) is connected with the data acquisition and processing module (1), the image detection module (3) and the device load current value prediction module (4); wherein:
the data acquisition and processing module (1) is used for acquiring an infrared image of the to-be-detected power transformation equipment, and environmental temperature data and environmental humidity data when the infrared image of the to-be-detected power transformation equipment is shot, and extracting image characteristics, a highest temperature value and a lowest temperature value of the infrared image of the to-be-detected power transformation equipment by using an image characteristic extraction model based on a convolutional neural network;
the image classification module (2) is used for classifying the infrared images of the to-be-detected power transformation equipment by using an image classification model based on an extreme learning machine according to the image characteristics of the infrared images of the to-be-detected power transformation equipment to obtain the equipment category corresponding to the infrared images of the to-be-detected power transformation equipment;
the image detection module (3) is used for matching the infrared image of the to-be-detected power transformation equipment by using a normal case and a fault case in a preset expert database under the equipment category and checking whether the infrared image of the to-be-detected power transformation equipment is normal or not;
the equipment load current value prediction module (4) is used for acquiring the load current value of equipment in the infrared image of the to-be-detected power transformation equipment by using an equipment load current prediction model based on an extreme learning machine according to the highest temperature value, the lowest temperature value, the environmental temperature data and the environmental humidity data;
and the equipment infrared detection report generating module (5) is used for generating a corresponding equipment infrared detection report according to the highest temperature value, the lowest temperature value, the environment temperature data, the environment humidity data, the load current value and the detection result of the infrared image of the to-be-detected power transformation equipment.
9. A storage medium having a computer program stored thereon, the computer program comprising: the computer program when executed by a processor implements the steps of the method for infrared detection of a power transformation device as claimed in any one of claims 1 to 7.
10. A computer device, characterized by: comprising a storage medium, a processor and a computer program stored in the storage medium and executable by the processor, the computer program, when executed by the processor, implementing the steps of the method for infrared detection of power transformation equipment as claimed in any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011630610.1A CN112766251B (en) | 2020-12-30 | 2020-12-30 | Infrared detection method and system for power transformation equipment, storage medium and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011630610.1A CN112766251B (en) | 2020-12-30 | 2020-12-30 | Infrared detection method and system for power transformation equipment, storage medium and computer equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112766251A true CN112766251A (en) | 2021-05-07 |
| CN112766251B CN112766251B (en) | 2022-06-14 |
Family
ID=75699530
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011630610.1A Active CN112766251B (en) | 2020-12-30 | 2020-12-30 | Infrared detection method and system for power transformation equipment, storage medium and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112766251B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116804706A (en) * | 2023-06-06 | 2023-09-26 | 淮阴工学院 | Temperature prediction method and device for lithium battery of electric automobile |
| WO2024001309A1 (en) * | 2022-06-30 | 2024-01-04 | 武汉高德智感科技有限公司 | Method and apparatus for generating and producing template for infrared thermal image analysis report |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104331521A (en) * | 2014-11-27 | 2015-02-04 | 国家电网公司 | Automatic power transformation equipment abnormity identification method based on image processing |
| CN105976383A (en) * | 2016-05-16 | 2016-09-28 | 国网河南省电力公司电力科学研究院 | Power transmission equipment fault diagnosis method based on limit learning machine image recognition |
| US20190331301A1 (en) * | 2016-12-30 | 2019-10-31 | Du Yuchuan | Method for leakage detection of underground pipeline corridor based on dynamic infrared thermal image processing |
| CN111242123A (en) * | 2020-01-07 | 2020-06-05 | 西安交通大学 | Power equipment fault diagnosis method based on infrared image |
| WO2020244134A1 (en) * | 2019-06-05 | 2020-12-10 | 华南理工大学 | Multi-task feature sharing neural network-based intelligent fault diagnosis method |
-
2020
- 2020-12-30 CN CN202011630610.1A patent/CN112766251B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104331521A (en) * | 2014-11-27 | 2015-02-04 | 国家电网公司 | Automatic power transformation equipment abnormity identification method based on image processing |
| CN105976383A (en) * | 2016-05-16 | 2016-09-28 | 国网河南省电力公司电力科学研究院 | Power transmission equipment fault diagnosis method based on limit learning machine image recognition |
| US20190331301A1 (en) * | 2016-12-30 | 2019-10-31 | Du Yuchuan | Method for leakage detection of underground pipeline corridor based on dynamic infrared thermal image processing |
| WO2020244134A1 (en) * | 2019-06-05 | 2020-12-10 | 华南理工大学 | Multi-task feature sharing neural network-based intelligent fault diagnosis method |
| CN111242123A (en) * | 2020-01-07 | 2020-06-05 | 西安交通大学 | Power equipment fault diagnosis method based on infrared image |
Non-Patent Citations (2)
| Title |
|---|
| 张洋 等: "基于LSTM神经网络的短期高压负荷电流预测方法", 《计算机科学》 * |
| 张洋 等: "基于LSTM神经网络的短期高压负荷电流预测方法", 《计算机科学》, vol. 46, no. 6, 30 June 2019 (2019-06-30), pages 49 - 54 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2024001309A1 (en) * | 2022-06-30 | 2024-01-04 | 武汉高德智感科技有限公司 | Method and apparatus for generating and producing template for infrared thermal image analysis report |
| CN116804706A (en) * | 2023-06-06 | 2023-09-26 | 淮阴工学院 | Temperature prediction method and device for lithium battery of electric automobile |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112766251B (en) | 2022-06-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112766251B (en) | Infrared detection method and system for power transformation equipment, storage medium and computer equipment | |
| CN111242123B (en) | Power equipment fault diagnosis method based on infrared image | |
| CN111751002B (en) | Intelligent charged equipment fault diagnosis method based on infrared thermal imaging | |
| CN112070134A (en) | Power equipment image classification method and device, power equipment and storage medium | |
| US20220323030A1 (en) | Probabilistic image analysis | |
| CN111382029B (en) | Mainboard abnormity diagnosis method and device based on PCA and multidimensional monitoring data | |
| CN111340116A (en) | Converter flame identification method and system, electronic equipment and medium | |
| WO2018044041A1 (en) | Method for learning latest data considering external influences in early warning system and system for same | |
| CN109145752A (en) | For assessing the method, apparatus, equipment and medium of object detection and track algorithm | |
| Brust et al. | Not just a matter of semantics: the relationship between visual similarity and semantic similarity | |
| Ettalibi et al. | AI and Computer Vision-based Real-time Quality Control: A Review of Industrial Applications | |
| CN117173147A (en) | Surface treatment equipment and method for steel strip processing | |
| CN115760822B (en) | Image quality detection model building method and system | |
| CN111238927A (en) | Fatigue durability evaluation method and device, electronic equipment and computer readable medium | |
| Yu et al. | Avocado ripeness classification using graph neural network | |
| CN111027569A (en) | Degraded insulator identification and diagnosis method and device based on infrared image | |
| CN114022698A (en) | Multi-tag behavior identification method and device based on binary tree structure | |
| WO2022139402A1 (en) | Diagnostic classification device and method | |
| CN117335409B (en) | Power consumer load prediction system based on artificial intelligence | |
| WO2023101375A1 (en) | Method, device, and system for optional artificial intelligence engine-based nondestructive inspection of object | |
| CN115083442B (en) | Data processing method, device, electronic equipment and computer readable storage medium | |
| AU2021282310B2 (en) | Identification of effect pigments in a target coating | |
| Jensen et al. | Deep learning-based anomaly detection on X-ray images of fuel cell electrodes | |
| CN118167426A (en) | Intelligent monitoring equipment and method for mine safety management | |
| Yujiang et al. | Ynet: Deep learning model-based detection of insulators and self-detonation defects |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |