CN116467579B

CN116467579B - Power equipment health grading method and system based on feature mining technology

Info

Publication number: CN116467579B
Application number: CN202310388237.0A
Authority: CN
Inventors: 莫建国; 王露民; 张锋; 郑瑞云; 俞佳捷; 卢俊; 陈家栋; 涂智恒; 郑南; 刘典; 蔡一骏; 林才春; 张贵中; 方凯伦; 周行; 周冬升
Original assignee: Ningbo Power Transmission And Transformation Construction Co ltd Operation And Maintenance Branch; Ningbo Power Supply Co of State Grid Zhejiang Electric Power Co Ltd
Current assignee: Ningbo Power Transmission And Transformation Construction Co ltd Operation And Maintenance Branch; Ningbo Power Supply Co of State Grid Zhejiang Electric Power Co Ltd
Priority date: 2023-04-12
Filing date: 2023-04-12
Publication date: 2023-11-14
Anticipated expiration: 2043-04-12
Also published as: CN116467579A

Abstract

The disclosure relates to a power equipment health grading method and system based on a feature mining technology, and relates to the technical field of power equipment evaluation. The health grading method comprises the following steps: acquiring first operation parameter vectors corresponding to a plurality of electric power mechanisms in the electric power equipment and second operation parameter vectors corresponding to connecting lines between adjacent electric power mechanisms; selecting a first characteristic parameter vector and a second characteristic parameter vector related to equipment health grading from the first operation parameter vector and the second operation parameter vector respectively based on a preset characteristic selection model; fusing the first characteristic parameter vector and the second characteristic parameter vector to obtain a fused parameter characteristic; and based on the fusion parameter characteristics, completing the health grading of the power equipment by utilizing the preset classification model. The embodiment of the disclosure can realize the health grading of the power equipment.

Description

Power equipment health grading method and system based on feature mining technology

Technical Field

The disclosure relates to the technical field of power equipment evaluation, in particular to a power equipment health grading method and system based on a feature mining technology.

Background

The power equipment is a basic unit or a subunit forming a power grid, and is a key for improving the resource optimal allocation capacity, the economic operation efficiency, the safety level and the intelligent level of the power grid. Current electrical equipment, comprising: and the power mechanisms such as transformers, buses, circuit breakers, knife switches and the like are connected by circuits.

Artificial intelligence is a branch of computer science that attempts to understand the essence of intelligence and to produce a new intelligent machine that can react in a similar way to human intelligence. Artificial intelligence has received increased attention and has found wide application in robotics, politics, decision-making, control systems, and simulation systems.

In addition, feature selection refers to the process of selecting corresponding features from existing features to optimize specific indexes of a system, and is a process of selecting some most effective features from original features to reduce the dimension of a data set, and is an important means for improving the performance of a learning algorithm and a key data preprocessing step in pattern recognition.

At present, the traditional power equipment health grading requires a great deal of labor cost, and lacks related technical schemes or technical means of automatic power equipment health grading. Therefore, it is necessary to grasp the health condition of all the power equipment in the current supervision range in real time based on the artificial intelligence and the feature selection technology so as to stop detection and update of the power equipment.

Disclosure of Invention

The disclosure provides a technical scheme of a power equipment health grading method and system based on a feature mining technology.

According to an aspect of the present disclosure, there is provided a power equipment health grading method based on a feature mining technology, which is characterized by comprising:

acquiring first operation parameter vectors corresponding to a plurality of electric power mechanisms in the electric power equipment and second operation parameter vectors corresponding to connecting lines between adjacent electric power mechanisms;

selecting a first characteristic parameter vector and a second characteristic parameter vector related to equipment health grading from the first operation parameter vector and the second operation parameter vector respectively based on a preset characteristic selection model;

fusing the first characteristic parameter vector and the second characteristic parameter vector to obtain a fused parameter characteristic;

and based on the fusion parameter characteristics, completing the health grading of the power equipment by utilizing the preset classification model.

Preferably, the method for fusing the first feature parameter vector and the second feature parameter vector to obtain the fused parameter feature includes:

splicing the first characteristic parameter vector and the second characteristic parameter vector to obtain a fusion parameter characteristic;

Or alternatively, the first and second heat exchangers may be,

splicing the first characteristic parameter vector and the second characteristic parameter vector to obtain a spliced characteristic parameter vector;

respectively fusing the first characteristic parameter vector, the second characteristic parameter vector and the spliced characteristic parameter vector by using a preset characteristic fusion model to obtain a first fused characteristic parameter vector, a second fused characteristic parameter vector and a third fused characteristic parameter vector;

and splicing the first fusion characteristic parameter vector, the second fusion characteristic parameter vector and the third fusion characteristic parameter vector to obtain fusion parameter characteristics.

Preferably, the method for completing the health grading of the power equipment by using the preset classification model based on the fusion parameter features comprises the following steps:

acquiring a plurality of preset classification models;

based on the fusion parameter characteristics, respectively utilizing a plurality of preset classification models to obtain a plurality of corresponding grading results;

and voting the plurality of grading results, and configuring the grading result with the highest voting as the healthy grading of the power equipment.

Or alternatively, the first and second heat exchangers may be,

the method for completing the health grading of the power equipment based on the fusion parameter characteristics by utilizing the preset classification model comprises the following steps:

Acquiring a plurality of preset classification models;

based on the fusion parameter characteristics, respectively utilizing a plurality of preset classification models to obtain a plurality of corresponding grading probability values;

and performing logistic regression on the plurality of grading probability values to finish the health grading of the power equipment.

Preferably, the power equipment health grading method further comprises the following steps: constructing a parameter vector matrix diagram by using the first operation parameter vector and the second operation parameter vector;

carrying out convolution processing on the parameter vector matrix diagram by using a preset convolution neural network to obtain a convolution characteristic parameter vector;

based on a preset feature selection model, selecting third feature parameter vectors related to equipment health grading from the convolution feature parameter vectors respectively;

fusing the first characteristic parameter vector, the second characteristic parameter vector and the third characteristic parameter vector to obtain a fused parameter characteristic;

Preferably, the method for fusing the first feature parameter vector, the second feature parameter vector and the third feature parameter vector to obtain a fused parameter feature includes:

Splicing the first characteristic parameter vector, the second characteristic parameter vector and the third characteristic parameter vector to obtain a fusion parameter characteristic;

or alternatively, the first and second heat exchangers may be,

splicing the first characteristic parameter vector, the second characteristic parameter vector and the third characteristic parameter vector to obtain a spliced characteristic parameter vector;

respectively fusing the first characteristic parameter vector, the second characteristic parameter vector, the third characteristic parameter vector and the spliced characteristic parameter vector by using a preset characteristic fusion model to obtain a first fused characteristic parameter vector, a second fused characteristic parameter vector, a third fused characteristic parameter vector and a fourth fused characteristic parameter vector;

and splicing the first fusion characteristic parameter vector, the second fusion characteristic parameter vector, the third fusion characteristic parameter vector and the fourth fusion characteristic parameter vector to obtain fusion characteristic parameters.

Preferably, the status of the electrical device is monitored prior to healthy grading of the electrical device;

if the state is in a first risk state, the power equipment health is adjusted up by a first set level in a rated manner;

if the state is in a second risk state, the health of the power equipment is graded down by a second set grade; wherein the second risk state has a higher risk level than the first risk state.

Preferably, the method of monitoring the status of the electrical device comprises:

acquiring an early warning risk value of the power equipment;

and monitoring the state of the power equipment based on the early warning risk value and a preset early warning risk value.

According to an aspect of the present disclosure, there is provided a power equipment health grading system based on feature mining technology, which is characterized by comprising:

the acquisition unit is used for acquiring first operation parameter vectors corresponding to a plurality of electric power mechanisms in the electric power equipment and second operation parameter vectors corresponding to connecting lines between adjacent electric power mechanisms;

the selection unit is used for selecting a first characteristic parameter vector and a second characteristic parameter vector which are related to the equipment health grade from the first operation parameter vector and the second operation parameter vector respectively based on a preset characteristic selection model;

the fusion unit is used for fusing the first characteristic parameter vector and the second characteristic parameter vector to obtain fusion parameter characteristics;

and the grading unit is used for completing the health grading of the power equipment by utilizing the preset classification model based on the fusion parameter characteristics.

According to an aspect of the present disclosure, there is provided a power equipment health grading system based on feature mining technology, which is characterized by comprising: a processor;

A memory for storing processor-executable instructions;

wherein the processor is configured to invoke the instructions stored by the memory to perform the power device health grading method based on feature mining techniques described above.

According to an aspect of the present disclosure, there is provided a power equipment health grading system based on feature mining technology, including: a computer readable storage medium having stored thereon computer program instructions, wherein the computer program instructions, when executed by a processor, implement the above-described power equipment health grading method based on feature mining techniques.

In the embodiment of the disclosure, the automatic health grading of the power equipment can be realized, so that the problems that a large amount of labor cost is required for traditional health grading of the power equipment and related technical schemes or technical means of the automatic health grading of the power equipment are lacked are solved.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.

Other features and aspects of the present disclosure will become apparent from the following detailed description of exemplary embodiments, which proceeds with reference to the accompanying drawings.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the disclosure and together with the description, serve to explain the technical aspects of the disclosure.

FIG. 1 illustrates a flow chart of a power plant health grading method based on feature mining techniques in accordance with an embodiment of the present disclosure;

FIG. 2 illustrates a network architecture diagram of a preset convolutional neural network in accordance with an embodiment of the present disclosure;

FIG. 3 illustrates a schematic diagram of a topological relationship between a plurality of power mechanisms according to an embodiment of the present disclosure;

FIG. 4 is a block diagram of an electronic device 800, shown in accordance with an exemplary embodiment;

fig. 5 is a block diagram illustrating an electronic device 1900 according to an example embodiment.

Detailed Description

Various exemplary embodiments, features and aspects of the disclosure will be described in detail below with reference to the drawings. In the drawings, like reference numbers indicate identical or functionally similar elements. Although various aspects of the embodiments are illustrated in the accompanying drawings, the drawings are not necessarily drawn to scale unless specifically indicated.

The word "exemplary" is used herein to mean "serving as an example, embodiment, or illustration. Any embodiment described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments.

The term "and/or" is herein merely an association relationship describing an associated object, meaning that there may be three relationships, e.g., a and/or B, may represent: a exists alone, A and B exist together, and B exists alone. In addition, the term "at least one" herein means any one of a plurality or any combination of at least two of a plurality, for example, including at least one of A, B, C, and may mean including any one or more elements selected from the group consisting of A, B and C.

Furthermore, numerous specific details are set forth in the following detailed description in order to provide a better understanding of the present disclosure. It will be understood by those skilled in the art that the present disclosure may be practiced without some of these specific details. In some instances, methods, means, elements, and circuits well known to those skilled in the art have not been described in detail in order not to obscure the present disclosure.

It will be appreciated that the above-mentioned method embodiments of the present disclosure may be combined with each other to form a combined embodiment without departing from the principle logic, and are limited to the description of the present disclosure.

In addition, the disclosure further provides a power equipment health grading device, an electronic device, a computer readable storage medium and a program based on the feature mining technology, and any one of the power equipment health grading methods based on the feature mining technology provided by the disclosure can be realized, and corresponding technical schemes and descriptions and corresponding descriptions of method parts are omitted.

Fig. 1 illustrates a flow chart of a power device health grading method based on feature mining techniques in accordance with an embodiment of the present disclosure. As shown in fig. 1, the method for grading the health of the power equipment based on the feature mining technology comprises the following steps: step S101: acquiring first operation parameter vectors corresponding to a plurality of electric power mechanisms in the electric power equipment and second operation parameter vectors corresponding to connecting lines between adjacent electric power mechanisms; step S102: selecting a first characteristic parameter vector and a second characteristic parameter vector related to equipment health grading from the first operation parameter vector and the second operation parameter vector respectively based on a preset characteristic selection model; step S103: fusing the first characteristic parameter vector and the second characteristic parameter vector to obtain a fused parameter characteristic; step S104: and based on the fusion parameter characteristics, completing the health grading of the power equipment by utilizing the preset classification model. The automatic health grading of the power equipment can be realized, so that the problems that a large amount of labor cost is required for the traditional health grading of the power equipment and related technical schemes or technical means of the automatic health grading of the power equipment are lacked are solved.

Step S101: and acquiring first operation parameter vectors corresponding to the plurality of electric power mechanisms in the electric power equipment and second operation parameter vectors corresponding to connecting lines between the adjacent electric power mechanisms.

In embodiments of the present disclosure and other possible embodiments, the first operating parameter vector (first operating parameter) may include: historical overhaul times, defect grades, tripping times, abnormal times and other information corresponding to a plurality of power mechanisms in the power equipment. The second plurality of operating parameter vectors (second operating parameters): historical overhaul times, defect levels, abnormal times and other information corresponding to a plurality of power mechanisms in the power equipment. The historical overhaul times, defect levels, trip times and anomaly times are conventional parameters for the man skilled in the art to overhaul the power mechanism and the connection lines between the power mechanisms, and the disclosure will not be described in detail.

In embodiments of the present disclosure and other possible embodiments, the power mechanism may include: one or more of power mechanisms such as transformers, buses, circuit breakers, knife switches and the like.

Step S102: and selecting a first characteristic parameter vector and a second characteristic parameter vector related to equipment health grading from the first operation parameter vector and the second operation parameter vector respectively based on a preset characteristic selection model.

In embodiments of the present disclosure and other possible embodiments, the preset feature selection model may be a minimum absolute contraction and selection operator (Least absolute shrinkage and selection operator, lasso) model that compresses variables that are larger by parameter estimation to less than 0; models may also be selected for other existing features. For example, LAR, bolasso, feaLectdent, etc. Wherein FeaLect scores all features according to regression coefficient combination analysis; another popular approach is a recursive feature elimination (Recursive Feature Elimination) algorithm, commonly used in support vector machines SVM, that removes low weight features by iteratively constructing the same model.

In the embodiment of the present disclosure, a mathematical expression corresponding to the Lasso model is given as shown in formula (1).

In formula (2.1), x _ij Representing an argument, i.e. normalized first/second operating parameter vector, y _i Representing a dependent variable (device health rating), a representing a first/second eigenvector associated with the device health rating; lambda represents penalty parameter (lambda.gtoreq.0), beta _j Representing regression coefficients, i.e. [1, n ]]，j∈[0,p]。

In embodiments of the present disclosure and other possible embodiments, the first/second feature parameter vector is a subset of the first/second operating parameter vector, respectively.

In the embodiments of the present disclosure and other possible embodiments, if the first feature parameter vector and the second feature parameter vector related to the equipment health grading have been determined by using the preset feature selection model, the corresponding first feature parameter vector and second feature parameter vector may be selected from the first operation parameter vector and the second operation parameter vector, and the corresponding preset feature selection model is not required to be executed.

Step S103: and fusing the first characteristic parameter vector and the second characteristic parameter vector to obtain a fused parameter characteristic.

In an embodiment of the disclosure, the method for fusing the first feature parameter vector and the second feature parameter vector to obtain a fused parameter feature includes: and splicing the first characteristic parameter vector and the second characteristic parameter vector to obtain a fusion parameter characteristic. For example, if the dimensions of the first feature parameter vector and the second feature parameter vector are n×n and m×n, respectively, the dimensions of the concatenated feature parameter vector are (n+m) ×n.

Or, in an embodiment of the present disclosure, the method for fusing the first feature parameter vector and the second feature parameter vector to obtain a fused parameter feature includes: splicing the first characteristic parameter vector and the second characteristic parameter vector to obtain a spliced characteristic parameter vector; respectively fusing the first characteristic parameter vector, the second characteristic parameter vector and the spliced characteristic parameter vector by using a preset characteristic fusion model to obtain a first fused characteristic parameter vector, a second fused characteristic parameter vector and a third fused characteristic parameter vector; and splicing the first fusion characteristic parameter vector, the second fusion characteristic parameter vector and the third fusion characteristic parameter vector to obtain fusion parameter characteristics.

The predetermined feature fusion model may be a principal component analysis (Principal Component Analysis, PCA) model or a neural network, for example BP is a neural network. Wherein, the neural network includes at least: an input layer, an hidden layer and an output layer; and training parameters corresponding to an input layer, an hidden layer and an output layer of the neural network by using the first characteristic parameter vector, the second characteristic parameter vector and the spliced characteristic parameter vector respectively to obtain three trained neural networks. After the trained neural network is obtained, the first characteristic parameter vector, the second characteristic parameter vector and the splicing characteristic parameter vector which correspond to the power equipment to be determined and are in health grading are fused by utilizing the correspondingly trained neural network, and then the first fused characteristic parameter vector, the second fused characteristic parameter vector and the third fused characteristic parameter vector are obtained respectively.

In the embodiments of the present disclosure and other possible embodiments, the process of splicing the first fused feature parameter vector, the second fused feature parameter vector, and the third fused feature parameter vector to obtain the fused feature parameter may refer to the process of splicing the first feature parameter vector and the second feature parameter vector to obtain the fused feature parameter, which is not described in detail herein.

Step S104: and based on the fusion parameter characteristics, completing the health grading of the power equipment by utilizing the preset classification model.

in embodiments of the present disclosure and other possible embodiments, the preset classification model may be a Machine Learning (ML) classification model, including: support vector machine SVM, decision tree DT, random forest RF, K neighbor KNN, logistic regression LR, self-adaptive enhancement Ada, linear discriminant analysis LDA, cluster analysis, multi-layer perceptron and the like.

In embodiments of the present disclosure and other possible embodiments, the preset classification model is trained using the fusion parameter features and corresponding health grading (labeling). And then, completing the health grading of the health grading power equipment to be determined by utilizing the fusion parameter characteristics corresponding to the health grading power equipment to be determined and the trained preset classification model.

In embodiments of the present disclosure and other possible embodiments, the health grading is at least 2 levels, but may be multiple levels. For example, the level 2 may be a health level and a non-health level, or a health level and a replacement level; the multiple levels may be a first level (health level), a second level (sub-health level), and a third level (replacement level). Those skilled in the art can configure the number of health grading stages according to actual needs.

Or, in an embodiment of the present disclosure, a plurality of preset classification models are obtained; based on the fusion parameter characteristics, respectively utilizing a plurality of preset classification models to obtain a plurality of corresponding grading results; and voting the plurality of grading results, and configuring the grading result with the highest voting as the healthy grading of the power equipment. Wherein the number of the preset classification models is at least 3 or more than 3 odd numbers.

In embodiments of the present disclosure and other possible embodiments, the preset classification model may be a Machine Learning (ML) classification model, including: at least 3 or more than 3 odd numbers in a Support Vector Machine (SVM), a Decision Tree (DT), a Random Forest (RF), a K Nearest Neighbor (KNN), a Logistic Regression (LR), an adaptive enhancement (Ada), a Linear Discriminant Analysis (LDA), a cluster analysis, a multi-layer perceptron and the like.

For example, taking 2 health grading levels as an example for explanation, 3 preset classification models are obtained and are respectively a support vector machine SVM, a decision tree DT and a random forest RF; respectively utilizing 3 preset classification models to obtain corresponding 3 grading results which are respectively a health level, a health level and a replacement level; voting on the plurality of grading results, the health level having 2 votes and the replacement level having only 1 vote; the power device health rating is configured to be a health rating.

Or, in an embodiment of the disclosure, the method for completing the health grading of the power equipment by using the preset classification model based on the fusion parameter features includes: acquiring a plurality of preset classification models; based on the fusion parameter characteristics, respectively utilizing a plurality of preset classification models to obtain a plurality of corresponding grading probability values; and performing logistic regression on the plurality of grading probability values to finish the health grading of the power equipment. Wherein, the logistic regression adopts LR or logistic regression model.

Logistic regression (Logistic regression, LR) is a very widely used classification machine learning algorithm that fits data into a logistic function to enable prediction of the probability of occurrence of events ^[161] . Logistic regression is a supervised generalized linear model, mainly used to classify samples. The logistic regression analysis comprises (1) linear regression, wherein y is a quantitative variable, and y has corresponding predicted values for different independent variables; (2) Logistic regression, y is a qualitative variable, e.g. the quantitative variable y can only be 0 or 1. The logistic regression is actually a linear regression function output followed by sigmoid processing.

Taking two classification as an example, let Y denote the class of the two classification problem, y=1 denote positive samples, y=0 denote negative samples, and the probability of positive samples is q. Y obeys bernoulli distribution, denoted as P (y=y) =q ^y (1-q) ^1-y Y is 0 or 1. In addition, a relationship between the mean value of Y and the independent variable, that is, a mathematical expression of the LR model is established as shown in expression (2).

Wherein, beta is the linear coefficient combination corresponding to the independent variable x.

Solving a parameter beta by using a maximum likelihood estimation method so as to obtain an estimated quantity q of q, wherein the estimated quantity q is shown as a formula (3); and when the estimated quantity is larger than 0.5, judging positive, otherwise judging negative.

The LR model has the advantages that data distribution is not required to be assumed in advance, and the problem of inaccurate classification caused by inaccurate assumption is avoided. In addition, the algorithm has good mathematical properties, and the optimal solution is easy to calculate.

In an embodiment of the disclosure, the power equipment health grading method further includes: constructing a parameter vector matrix diagram by using the first operation parameter vector and the second operation parameter vector; carrying out convolution processing on the parameter vector matrix diagram by using a preset convolution neural network to obtain a convolution characteristic parameter vector; based on a preset feature selection model, selecting third feature parameter vectors related to equipment health grading from the convolution feature parameter vectors respectively; fusing the first characteristic parameter vector, the second characteristic parameter vector and the third characteristic parameter vector to obtain a fused parameter characteristic; and based on the fusion parameter characteristics, completing the health grading of the power equipment by utilizing the preset classification model.

In embodiments of the present disclosure and other possible embodiments, there is provided: splicing the first operation parameter vector and the second operation parameter vector to obtain a spliced vector; determining the number M of the splicing vectors; and arranging the spliced vectors in a set mode to generate an MxM image histology feature map. In an embodiment of the present disclosure and other possible embodiments, the method for generating an mxm image histology feature map by arranging the stitching vectors in a set manner includes: taking M characteristic elements of the spliced vector as characteristic vectors of a first row; shifting the feature vector of the next row to the right of the feature vector of the previous row, wherein the last feature element is used as the first feature element of the row; finally, an m×m parameter vector matrix map is generated.

For example, the splice vector is [1,2,3,4], the number of splice vectors is 4, the feature vector of the first row is [1,2,3,4], the feature vector of the second row is [4,1,2,3], the feature vector of the third row is [3,4,1,2], and the feature vector of the fourth row is [2,3,4,1]; finally, a 4×4 parametric vector matrix diagram is generated, the first to fourth rows of the 4×4 parametric vector matrix diagram being [1,2,3,4], [4,1,2,3], [3,4,1,2] and [2,3,4,1], respectively.

In embodiments of the present disclosure and other possible embodiments, a preset convolutional neural network includes: a convolution layer, an exponential linear unit/gaussian error linear unit (activation function layer) ELU, and four stride convolution layers. Wherein the feature map is reduced in size by half per stride of the convolutional layers. The convolutional neural network can be configured according to actual needs by a person skilled in the art, and the structure of the convolutional neural network is not particularly limited in the present disclosure.

Fig. 2 illustrates a network structure diagram of a preset convolutional neural network according to an embodiment of the present disclosure. As shown in fig. 2, the preset convolutional neural network includes: the method comprises the following steps of sequentially connecting a plurality of convolution units or sequentially connecting the plurality of convolution units by adopting residual errors. Specifically, a plurality of convolution units connected in sequence, including: a first convolution layer (5×5×5 Conv-ELU) a second convolution layer (2×2×2Conv S2), a first activation function layer (ELU) a second convolution layer (2×2×2ConvS2) a first activation function layer (ELU) a fifth convolution layer (5×5×5Conv-ELU×4), a fourth activation function layer (ELU) a sixth convolution layer (2×2×2ConvS2) sixth convolution layer (2) x 2Conv S2). Wherein S2 and x 2 each represent 2 identical layer structures, and similarly x 4 represents 4 identical layer structures.

In embodiments of the present disclosure and other possible embodiments, the training of the preset convolutional neural network is performed in advance using the parameter vector matrix diagram. And inputting a parameter vector matrix diagram corresponding to the healthy rated power equipment to be determined into a trained preset convolutional neural network to carry out convolutional processing, so as to obtain a convolutional characteristic parameter vector. And then, fusing the first characteristic parameter vector, the second characteristic parameter vector and the third characteristic parameter vector by using a preset characteristic fusion model to obtain fusion parameter characteristics.

In an embodiment of the disclosure, the method for fusing the first feature parameter vector, the second feature parameter vector and the third feature parameter vector to obtain a fused parameter feature includes: and splicing the first characteristic parameter vector, the second characteristic parameter vector and the third characteristic parameter vector to obtain a fusion parameter characteristic.

Or, in an embodiment of the present disclosure, the first feature parameter vector, the second feature parameter vector, and the third feature parameter vector are spliced to obtain a spliced feature parameter vector; respectively fusing the first characteristic parameter vector, the second characteristic parameter vector, the third characteristic parameter vector and the spliced characteristic parameter vector by using a preset characteristic fusion model to obtain a first fused characteristic parameter vector, a second fused characteristic parameter vector, a third fused characteristic parameter vector and a fourth fused characteristic parameter vector; and splicing the first fusion characteristic parameter vector, the second fusion characteristic parameter vector, the third fusion characteristic parameter vector and the fourth fusion characteristic parameter vector to obtain fusion characteristic parameters.

In an embodiment of the present disclosure, the status of the electrical device is monitored prior to healthy grading of the electrical device; if the state is in a first risk state (low risk state), up-regulating the power equipment health rating by a first set level; if the state is in a second risk state (high risk state), downgrading the power equipment health by a second set level; wherein the second risk state has a higher risk level than the first risk state.

In embodiments of the present disclosure and other possible embodiments, the first setting level and the first setting level may be configurable to 1 level or 2 levels. The first setting level can be configured by a person skilled in the art according to actual needs.

For example, the health-rating number is a first level (health level), a second level (sub-health level), and a third level (replacement level) of the health rating number; in addition, the first setting level and the first setting level are respectively configured as 1 level; if the state is in a first risk state (low risk state), the power equipment health grade is the highest health grade, and no adjustment is performed; if the state is in a first risk state (low risk state), the power equipment health rating is a sub-health rating, and then the power equipment health rating is up-regulated by a first set rating to a health rating; if the state is in a second risk state (high risk state), grading the power equipment health to a replacement level of the lowest level, and not adjusting; and if the state is in a second risk state (high risk state), grading the power equipment health into a sub-health grade, and downwards regulating the power equipment health grade to a second set grade to a replacement grade.

In an embodiment of the disclosure, the method of monitoring a status of the electrical device includes: acquiring an early warning risk value of the power equipment; and monitoring the state of the power equipment based on the early warning risk value and a preset early warning risk value.

In an embodiment of the present disclosure and other possible embodiments, a method for monitoring a state of a power device based on a topology analysis technology is provided, including: establishing or acquiring a topological relation among a plurality of electric power mechanisms, and determining an early warning risk vector corresponding to each topological branch based on a preset risk weight corresponding to an adjacent electric power mechanism of each topological branch in the topological relation; determining an early warning risk value of the power equipment corresponding to the topological relation according to the early warning risk vector corresponding to each topological branch; and monitoring the state of the power equipment based on the early warning risk value and a preset early warning risk value. The method solves the problem that the prior power equipment lacks necessary state monitoring and control, improves the resource optimal configuration capacity, economic operation efficiency, safety level and intelligent level of the power grid, achieves the aims of reliability, safety, economy, high efficiency and safe use of the power grid, and further ensures the optimal and high-efficiency operation of the power grid so as to provide the electric energy quality meeting the demands of users.

Establishing or acquiring a topological relation among a plurality of electric power mechanisms, and determining an early warning risk vector corresponding to each topological branch based on a preset risk weight corresponding to an adjacent electric power mechanism of each topological branch in the topological relation.

In an embodiment of the present invention, the method for establishing a topological relation between a plurality of power institutions includes: and establishing a topological relation among the plurality of electric power mechanisms by taking the plurality of electric power mechanisms as nodes and connecting lines among the plurality of electric power mechanisms as topological connecting lines.

In an embodiment of the present invention and other possible embodiments, the power mechanism includes at least: one or more of power mechanisms such as transformers, buses, circuit breakers, knife switches and the like. For example, the plurality of power mechanisms includes: the transformer, the circuit breaker and the disconnecting link are used as nodes, and the connecting lines of the transformer, the circuit breaker and the disconnecting link are used as topological connecting lines, so that the topological relation of the transformer, the circuit breaker and the disconnecting link is established. The connection lines of the transformer, the circuit breaker and the knife switch can be selected by a person skilled in the art according to actual needs, and the disclosure will not be described in detail herein.

In embodiments of the present invention and other possible embodiments, establishing or obtaining a topological relationship between a plurality of power mechanisms may include a plurality of topological branches, where each of the topological branches may include a plurality of power mechanisms of the same type or different types, or a plurality of power mechanisms of the same type but different models. For example, the path of each topological branch can comprise one or more of a transformer, a bus, a circuit breaker, a disconnecting link and other power mechanisms. Wherein the same type is the same power mechanism, such as a transformer.

Specifically, in the embodiment of the present invention and other possible embodiments, the topology relationship between the plurality of power mechanisms is a tree topology relationship or a network topology relationship or a combination of tree topology relationship and network topology relationship. In the tree topology relationship, the nodes formed by power mechanisms of different levels in the tree structure are different in position, the nodes at the tree root part are configured as main nodes, the nodes extending downwards from the main nodes are configured as intermediate nodes, and the nodes at the tail end are configured as leaf nodes. In the tree topology, no loop is generated between two nodes in all nodes, and all paths can carry out bidirectional transmission. If the tree topology and the set of the net topology are combined, the net topology is built between the intermediate nodes or the leaf nodes, and thus loops are generated between the intermediate nodes or the leaf nodes.

More specifically, the power mechanisms corresponding to the main node, the intermediate node and the leaf node may be multiple power mechanisms of the same type or different types, or multiple power mechanisms of the same type but different types, and how to configure the topology relationship between the multiple power mechanisms is required to be configured by those skilled in the art according to actual needs. For example, in the topology of the power equipment of the power transformation mechanism, the main node and the intermediate node need to be configured with transformers (the same type of power mechanism), and even the leaf node needs to be configured with transformers (the same type of power mechanism), however, the types of the transformers of the main node, the intermediate node and the leaf node are different, and meanwhile, the power mechanisms such as a bus, a circuit breaker and a disconnecting link need to be matched, so that the final topology is formed by using a connecting line.

In the embodiments of the present disclosure and other possible embodiments, the starting point of the topology branch is an electric power input terminal, and the ending point of the topology branch is an electric power consumption terminal. Wherein the power input terminal may be a power plant (plant), such as a thermal power plant, a wind power plant, a nuclear power plant, or a corresponding power plant within a thermal power plant, a wind power plant, a nuclear power plant; wherein the power device further comprises a plurality of power mechanisms; the electricity terminal may be a substation (house), a factory, a residential user, a public electricity consumer or other electricity consumers, for example: electric equipment for urban traffic illumination, subway or light rail and the like.

In the embodiment of the present invention and other possible embodiments, the preset risk weight corresponding to the adjacent power mechanism is a weight between the adjacent power mechanisms in each topology branch. For example, the number of adjacent power mechanisms in each topological branch is N, and the number of weight correspondences between adjacent power mechanisms in each topological branch is N-1. Therefore, the early warning risk vector is a first early warning risk vector with 1 dimension, and the element of the first early warning risk vector is the weight between adjacent power mechanisms in each topological branch.

Fig. 3 shows a schematic diagram of a topological relationship between a plurality of power institutions according to an embodiment of the present disclosure. As shown in fig. 3, the topological relation among the plurality of power mechanisms includes: 3 topological branches, namely a first topological branch, a second topological branch and a third topological branch; the first topology leg, further comprising: the power input terminal 1, the first transformer 21 and the second transformer 22 which are connected in parallel, the first breaker 31, the second breaker 32 and the third breaker 33 which are connected in parallel, the first disconnecting link 41, the second disconnecting link 42 and the third disconnecting link 43 which are connected in parallel, and the first power terminal 51; a second topology leg, further comprising: a third transformer 23, …, a second power use terminal 52; a third topology leg, further comprising: fourth transformer 24, …, third electrical terminal 53.

In an embodiment of the present invention, the method for determining the early warning risk vector corresponding to each topology branch based on the preset risk weight corresponding to the adjacent power mechanism of each topology branch in the topology relationship includes: and calculating a preset risk weight corresponding to the adjacent power mechanism of each topological branch in the topological relation by using the obtained preset nonlinear function, and determining an early warning risk vector corresponding to each topological branch.

In the embodiment of the present invention and other possible embodiments, the preset nonlinear function may be configured as one or more of Sigmoid function, hyperbolic tangent function Tanh, modified linear unit function ReLu, gaussian error linear unit function GeLu, softmax function, so as to convert the preset risk weight corresponding to the adjacent power mechanism of each topological branch in the topological relation into an early warning risk vector in the 0-1 interval.

For example, the preset nonlinear function may be configured as a Sigmoid function, and the preset risk weight corresponding to the adjacent power mechanism of each topological branch in the topological relation is calculated by using the Sigmoid function, so as to obtain an early warning risk vector in the interval 0-1, so as to determine the early warning risk vector corresponding to each topological branch. Specifically, the preset risk weights corresponding to adjacent power mechanisms of the Sigmoid function are compressed into a range from 0 to 1, the preset risk weights corresponding to adjacent power mechanisms corresponding to the adjacent power mechanisms with larger negative absolute values are converted to approach 0, the preset risk weights corresponding to adjacent power mechanisms corresponding to the adjacent power mechanisms with larger positive absolute values are converted to approach 1, and then the preset risk weights corresponding to the adjacent power mechanisms of each topological branch in the topological relation are further quantized.

And determining the early warning risk value of the power equipment corresponding to the topological relation according to the early warning risk vector corresponding to each topological branch. Based on the above, the disclosure proposes a method for determining a preset risk weight corresponding to an adjacent power mechanism of each topological branch in the topological relation.

In an embodiment of the present invention, before the pre-warning risk vector corresponding to each topology branch is determined based on the preset risk weights corresponding to the neighboring power mechanisms of each topology branch in the topology relationship, a determination method of the pre-warning risk vector corresponding to the neighboring power mechanisms of each topology branch in the topology relationship is determined, including: acquiring a preset survival analysis model, first multi-time operation parameters corresponding to adjacent power mechanisms of each topological branch and second multi-time operation parameters corresponding to connecting lines between the adjacent power mechanisms; based on the first multi-time operation parameter and the second multi-time operation parameter, constructing a weight setting model by using a preset survival analysis model; based on the constructed weight setting model, the preset risk weights corresponding to the adjacent power mechanisms of each topological branch in the topological relation are determined by utilizing the first multi-time operation parameters corresponding to the adjacent power mechanisms and the second multi-time operation parameters corresponding to the connecting lines between the adjacent power mechanisms.

In the embodiment of the present invention and other possible embodiments, the adjacent power mechanisms for constructing the weight setting model and the connection lines between the adjacent power mechanisms need to determine the corresponding power mechanisms and the connection lines thereof in the preset risk weights corresponding to the adjacent power mechanisms of each topology branch. For example, it is required to determine a corresponding power mechanism and a connection line thereof in a preset risk weight corresponding to an adjacent power mechanism of each topology branch, including: and if the adjacent power mechanisms are used for constructing the weight setting model, the adjacent power mechanisms also need to comprise the power mechanisms such as the transformer, the circuit breaker and the disconnecting link, and the connecting lines between the adjacent power mechanisms are matched with the types and the types of the connecting lines of the power mechanisms in the preset risk weights corresponding to the adjacent power mechanisms needing to be determined for each topological branch. And the topological branch circuit formed by the adjacent power mechanisms and the connecting circuits between the adjacent power mechanisms for constructing the weight setting model is identical or same with the topological branch circuit formed by the corresponding power mechanism and the connecting circuits thereof in the preset risk weight corresponding to the adjacent power mechanism for determining each topological branch circuit.

The point starting time of the first multiple time and the second multiple time can be configured as the time when the adjacent power mechanism and the connection line between the adjacent power mechanisms start to be used, and the ending time of the first multiple time and the second multiple time needs to be configured as the failure time (the time corresponding to the fault) corresponding to the topology branch.

In the embodiments of the present invention and other possible embodiments, the first multiple times and the second multiple times correspond to each other, for example, the first multiple times are 1 st day, 2 nd day, 3 rd day, and 4 th day, and the second multiple times should also be 1 st day, 2 nd day, 3 rd day, and 4 th day; for example, the first multiple time is 1 month, 2 months, 3 months, and 4 months, and the second multiple time should be 1 month, 2 months, 3 months, and 4 months. The configuration at the first multiple times and the configuration at the second multiple times need to be configured according to the attribute types corresponding to the electric power mechanism and the connecting line between the electric power mechanisms.

In one embodiment of the present invention and other possible embodiments, the first multi-time operation parameters corresponding to the adjacent power mechanism of each topology branch may include: the first time corresponds to the information such as historical overhaul times, defect levels, tripping times and abnormal times. The second multi-time operation parameters corresponding to the connection lines between the adjacent power mechanisms of each topological branch can comprise: and the second time is corresponding to the historical overhaul times, defect levels, abnormal times and other information. The historical overhaul times, defect levels, trip times and anomaly times are conventional parameters for the man skilled in the art to overhaul the power mechanism and the connection lines between the power mechanisms, and the disclosure will not be described in detail.

More specifically, the first multi-time operation parameters corresponding to the adjacent power mechanisms of each topological branch and the second multi-time operation parameters corresponding to the connecting lines between the adjacent power mechanisms can be obtained by experimental testing under set experimental conditions in a laboratory. For example, the set experimental conditions may include setting a power facility and setting a climate condition of a region where a connection line between the power facilities is located, for example, the climate condition may be an ambient temperature, an altitude, an ambient humidity, a wind level, a rain, snow, ice fog, or the like.

In embodiments of the present invention and other possible embodiments, the survival analysis model (Proportional hazards model, cox model) is a probabilistic model describing time-based system failure or life death.

The mathematical expression corresponding to the Cox model is shown as formula (1).

/>

Wherein x represents a study variable, a first multi-time operation parameter corresponding to an adjacent power mechanism of each topological branch and a second multi-time operation parameter corresponding to a connecting line between the adjacent power mechanisms of the present disclosure; t represents a first multi-moment or a second multi-moment, wherein the first multi-moment and the second multi-moment correspond relatively, for example, the first multi-moment is 1 st, 2 nd, 3 rd and 4 th day, and the second multi-moment should be 1 st, 2 nd, 3 rd and 4 th day; lambda (lambda) ₀ (t) represents a baseline risk function or a baseline risk rate; beta ^T x represents a log-risk function; beta represents the partial regression coefficient of the independent variable estimated from the sample data (study variable); the T topology leg represents the failure time (time corresponding to the failure time).

In an embodiment of the invention, a method for determining an early warning risk value of power equipment corresponding to a topological relation based on an early warning risk vector corresponding to a topological branch is provided. Specifically, the method for determining the early warning risk value of the power equipment corresponding to the topological relation according to the early warning risk vector corresponding to each topological branch comprises the following steps: determining or acquiring a relationship between power mechanisms in sub-topology branches in the topology branches; wherein the relationship is a parallel relationship or a serial relationship; if the relation of the power mechanisms in the sub-topology branch circuits is the parallel relation, only the power mechanism corresponding to the maximum preset risk weight of the early warning risk vector is reserved in the sub-topology branch circuits, and the power mechanisms of the rest parallel relation are deleted; until the relations of the power mechanisms in the sub-topology branches are serial relations, and updating the early warning risk vector; and calculating the updated early warning risk vector (the early warning risk elements in the power equipment) based on the acquired preset formula, and determining the early warning risk value of the power equipment corresponding to the topological relation. The updated early warning risk vector is only a preset risk weight corresponding to the power mechanism with the serial relation.

In the embodiment of the invention and other possible embodiments, the concept of parallel relationship and serial relationship uses the concept of parallel connection and serial connection in a circuit, and is specific to adversary, a plurality of power mechanisms in a topological branch of the parallel relationship do not respond to each other, and a plurality of power mechanisms in a topological branch of the serial relationship are in a cascade state.

Based on the above, for example, the topological relation among the plurality of power mechanisms comprises a plurality of topological branches L ₁ 、L ₂ 、L ₃ 、...、L _n Parallel relations and serial relations exist among the power mechanisms in the topological branches; if the power mechanism L in the sub-topology branch of the topology branch is ₁₁ 、L ₂₁ If the relation is the parallel relation, only the power mechanism corresponding to the maximum preset risk weight of the early warning risk vector is reserved in the sub-topology branch; and until the relations of the power mechanisms in the sub-topology branches are serial relations, and updating the early warning risk vector.

As shown in fig. 3, the topological relation among the plurality of power mechanisms comprises 3 topological branches L ₁ 、L ₂ 、L ₃ The description will be given taking part of the first topology branch as an example. For example, a first topological leg L ₁ Comprises only: the power input terminal 1, the first transformer 21 and the second circuit breaker are connected in sequence 32. The first disconnecting link 41, the second disconnecting link 42, the third disconnecting link 43 and the first electric terminal 51 are connected in parallel; at this time, the sub-topology branch L in the first topology branch ₁ The first disconnecting link 41, the second disconnecting link 42 and the third disconnecting link (electric mechanism) are in parallel connection, and the largest early warning risk vector is selected from the early warning risk vectors of the corresponding multiple topological branches; for example, if the early warning risk vectors corresponding to the first, second and third disconnecting links 41, 42 and 43 in the parallel relationship are respectively 0.4, 0.5 and 0.8, only the electric power mechanism (the third disconnecting link 43) corresponding to the preset risk weight of 0.8 of the early warning risk vector is reserved in the sub-topology branch, and the first disconnecting link 41, that is, the second disconnecting link 42 in the parallel relationship in the sub-topology branch is deleted; then, the relations of the power mechanisms in the sub-topology branches are serial relations; that is, the power input terminal 1, the first transformer 21, the second circuit breaker 32, and the third switch 43, which are sequentially connected, are all in serial relationship, and the early warning risk vector is updated, and at this time, the updated early warning risk vector is only a preset risk weight among the power input terminal 1, the first transformer 21, the second circuit breaker 32, and the third switch 43 in serial relationship.

For example, in the embodiment of the present invention and other possible embodiments, the preset formula may be configured as a sum calculation formula, and the early warning risk elements in the updated early warning risk vector are summed to obtain the early warning risk value of the power device corresponding to the topological relation. For another example, in the embodiment of the present invention and other possible embodiments, the preset formula may be configured as a mean value calculation formula, and the mean value of the early warning risk elements in the updated early warning risk vector is performed, so as to obtain an early warning risk value of the power device corresponding to the topological relation.

In an embodiment of the present disclosure, a method for determining an early warning risk value of a power device corresponding to the topological relation based on an artificial intelligence technology is provided. Specifically, the method for determining the early warning risk value of the power equipment corresponding to the topological relation according to the early warning risk vector corresponding to each topological branch comprises the following steps: acquiring a preset classification model; and determining the early warning risk value of the power equipment corresponding to the topological relation by utilizing the preset classification model based on the early warning risk vector corresponding to each topological branch.

In embodiments of the present disclosure and other possible embodiments, the preset classification model may be one or more of a Support Vector Machine (SVM), a multi-layer perceptron (MLP), a Random Forest (RF), a K-nearest neighbor (KNN), a Logistic Regression (LR), a Decision Tree (DT), a gradient lifting (GB), a Linear Discriminant Analysis (LDA), and the like; and meanwhile, determining the early warning risk value of the power equipment corresponding to the topological relation as a corresponding early warning risk probability value by utilizing the preset classification model.

In an embodiment of the disclosure, the method for monitoring the state of the power equipment based on the early warning risk value and a preset early warning risk value includes: if the early warning risk value is larger than or equal to a preset early warning risk value, determining that the power equipment state is in a high risk state; otherwise, determining that the power device state is in a low risk state. The preset early warning risk value can be configured to be 0.5, and a person skilled in the art can also perform configuration of the preset early warning risk value according to actual needs, for example, 0.4, 0.7, 0.8, etc.

Meanwhile, the disclosure also provides a power equipment control method based on the topology analysis technology, which comprises the following steps: according to the power equipment state monitoring method, after the power equipment state is monitored, if the power equipment state is in a high-risk state, the early warning risk vector corresponding to each topological branch is respectively determined again, and the relation of the power mechanism in the sub-topological branch in the topological branch is determined; and if the relation of the power mechanisms is a parallel relation, controlling the connection circuit of the power mechanism corresponding to the maximum preset risk weight in the early warning risk vector to be disconnected.

In an embodiment of the present disclosure and other possible embodiments, the method for redefining the early warning risk vector corresponding to each topology branch includes: acquiring a preset survival analysis model, first multi-time operation parameters corresponding to adjacent power mechanisms of each topological branch and second multi-time operation parameters corresponding to connecting lines between the adjacent power mechanisms; based on the first multi-time operation parameter and the second multi-time operation parameter, constructing a weight setting model by using a preset survival analysis model; based on the constructed weight setting model, utilizing a first multi-time operation parameter corresponding to adjacent power mechanisms to be determined and a second multi-time operation parameter corresponding to a connecting line between the adjacent power mechanisms to be determined, determining preset risk weights corresponding to the adjacent power mechanisms of each topological branch in the topological relation, and determining the preset risk weights corresponding to the adjacent power mechanisms of each topological branch in the topological relation; and determining an early warning risk vector corresponding to each topological branch based on a preset risk weight corresponding to an adjacent power mechanism of each topological branch in the topological relation. Reference is made in particular to the detailed description of the state monitoring method based on topology analysis techniques.

In the embodiments of the present disclosure and other possible embodiments, before the redetermining the pre-warning risk vector corresponding to each topological branch, determining or acquiring a relationship between power mechanisms in topological branches in the topological branch; wherein the relationship is a parallel relationship or a serial relationship; and if the relation of the power mechanisms in the sub-topology branch circuits is the parallel relation, deleting the power mechanism corresponding to the maximum preset risk weight of the early warning risk vector in the sub-topology branch circuits.

In the embodiment of the disclosure, after the connection line of the power mechanism corresponding to the maximum early warning risk vector is controlled to be disconnected, the topological relation among the plurality of power mechanisms in the connection state and the early warning risk vector corresponding to each topological branch are reestablished or acquired, and the state of the power equipment is re-monitored; if the power equipment state is still in a high risk state, controlling the power equipment to disconnect all connection lines; otherwise, the power equipment is controlled to maintain the current connection line.

The execution subject of the feature mining technology-based power device health grading method may be a feature mining technology-based power device health grading apparatus or system, for example, the feature mining technology-based power device health grading method may be executed by a terminal device or a server or other processing device, where the terminal device may be a User Equipment (UE), a mobile device, a User terminal, a cellular phone, a cordless phone, a personal digital processing (Personal Digital Assistant, PDA), a handheld device, a computing device, a vehicle-mounted device, a wearable device, or the like. In some possible implementations, the power device health grading method based on feature mining techniques may be implemented by way of a processor invoking computer readable instructions stored in a memory.

It will be appreciated by those skilled in the art that in the above-described method of the specific embodiments, the written order of steps is not meant to imply a strict order of execution but rather should be construed according to the function and possibly inherent logic of the steps.

In addition, the disclosure also provides a power equipment health grading system based on the feature mining technology, which comprises: the acquisition unit is used for acquiring first operation parameter vectors corresponding to a plurality of electric power mechanisms in the electric power equipment and second operation parameter vectors corresponding to connecting lines between adjacent electric power mechanisms; the selection unit is used for selecting a first characteristic parameter vector and a second characteristic parameter vector which are related to the equipment health grade from the first operation parameter vector and the second operation parameter vector respectively based on a preset characteristic selection model; the fusion unit is used for fusing the first characteristic parameter vector and the second characteristic parameter vector to obtain fusion parameter characteristics; and the grading unit is used for completing the health grading of the power equipment by utilizing the preset classification model based on the fusion parameter characteristics. The automatic health grading of the power equipment can be realized, so that the problems that a large amount of labor cost is required for the traditional health grading of the power equipment and related technical schemes or technical means of the automatic health grading of the power equipment are lacked are solved.

In addition, the disclosure also provides a power equipment health grading system based on the feature mining technology, which comprises: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to invoke the instructions stored in the memory to perform the method described above. The automatic health grading of the power equipment can be realized, so that the problems that a large amount of labor cost is required for the traditional health grading of the power equipment and related technical schemes or technical means of the automatic health grading of the power equipment are lacked are solved.

In addition, the disclosure also provides a power equipment health grading system based on the feature mining technology, which comprises: a computer readable storage medium having stored thereon computer program instructions, characterized in that the computer program instructions, when executed by a processor, implement the method as described above. The automatic health grading of the power equipment can be realized, so that the problems that a large amount of labor cost is required for the traditional health grading of the power equipment and related technical schemes or technical means of the automatic health grading of the power equipment are lacked are solved.

In some embodiments, functions or modules included in an apparatus provided by the embodiments of the present disclosure may be used to perform a method described in the foregoing method embodiments, and specific implementations thereof may refer to descriptions of the foregoing method embodiments, which are not repeated herein for brevity.

The disclosed embodiments also provide a computer readable storage medium having stored thereon computer program instructions which, when executed by a processor, implement the above-described method. The computer readable storage medium may be a non-volatile computer readable storage medium.

The embodiment of the disclosure also provides an electronic device, which comprises: a processor; a memory for storing processor-executable instructions; wherein the processor is configured as the method described above. The electronic device may be provided as a terminal, server or other form of device.

Fig. 4 is a block diagram of an electronic device 800, according to an example embodiment. For example, electronic device 800 may be a mobile phone, computer, digital broadcast terminal, messaging device, game console, tablet device, medical device, exercise device, personal digital assistant, or the like.

Referring to fig. 4, the electronic device 800 may include one or more of the following components: a processing component 802, a memory 804, a power component 806, a multimedia component 808, an audio component 810, an input/output (I/O) interface 812, a sensor component 814, and a communication component 816.

The processing component 802 generally controls overall operation of the electronic device 800, such as operations associated with display, telephone calls, data communications, camera operations, and recording operations. The processing component 802 may include one or more processors 820 to execute instructions to perform all or part of the steps of the methods described above. Further, the processing component 802 can include one or more modules that facilitate interactions between the processing component 802 and other components. For example, the processing component 802 can include a multimedia module to facilitate interaction between the multimedia component 808 and the processing component 802.

The memory 804 is configured to store various types of data to support operations at the electronic device 800. Examples of such data include instructions for any application or method operating on the electronic device 800, contact data, phonebook data, messages, pictures, videos, and so forth. The memory 804 may be implemented by any type or combination of volatile or nonvolatile memory devices such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disk.

The power supply component 806 provides power to the various components of the electronic device 800. The power components 806 may include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power for the electronic device 800.

The multimedia component 808 includes a screen between the electronic device 800 and the user that provides an output interface. In some embodiments, the screen may include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from a user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensor may sense not only the boundary of a touch or slide action, but also the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 808 includes a front camera and/or a rear camera. When the electronic device 800 is in an operational mode, such as a shooting mode or a video mode, the front camera and/or the rear camera may receive external multimedia data. Each front camera and rear camera may be a fixed optical lens system or have focal length and optical zoom capabilities.

The audio component 810 is configured to output and/or input audio signals. For example, the audio component 810 includes a Microphone (MIC) configured to receive external audio signals when the electronic device 800 is in an operational mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signals may be further stored in the memory 804 or transmitted via the communication component 816. In some embodiments, audio component 810 further includes a speaker for outputting audio signals.

The I/O interface 812 provides an interface between the processing component 802 and peripheral interface modules, which may be a keyboard, click wheel, buttons, etc. These buttons may include, but are not limited to: homepage button, volume button, start button, and lock button.

The sensor assembly 814 includes one or more sensors for providing status assessment of various aspects of the electronic device 800. For example, the sensor assembly 814 may detect an on/off state of the electronic device 800, a relative positioning of the components, such as a display and keypad of the electronic device 800, the sensor assembly 814 may also detect a change in position of the electronic device 800 or a component of the electronic device 800, the presence or absence of a user's contact with the electronic device 800, an orientation or acceleration/deceleration of the electronic device 800, and a change in temperature of the electronic device 800. The sensor assembly 814 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor assembly 814 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 814 may also include an acceleration sensor, a gyroscopic sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

The communication component 816 is configured to facilitate communication between the electronic device 800 and other devices, either wired or wireless. The electronic device 800 may access a wireless network based on a communication standard, such as WiFi,2G, or 3G, or a combination thereof. In one exemplary embodiment, the communication component 816 receives broadcast signals or broadcast related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, the communication component 816 further includes a Near Field Communication (NFC) module to facilitate short range communications. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, ultra Wideband (UWB) technology, bluetooth (BT) technology, and other technologies.

In an exemplary embodiment, the electronic device 800 may be implemented by one or more Application Specific Integrated Circuits (ASICs), digital Signal Processors (DSPs), digital Signal Processing Devices (DSPDs), programmable Logic Devices (PLDs), field Programmable Gate Arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic elements for executing the methods described above.

In an exemplary embodiment, a non-transitory computer readable storage medium is also provided, such as memory 804 including computer program instructions executable by processor 820 of electronic device 800 to perform the above-described methods.

Fig. 5 is a block diagram illustrating an electronic device 1900 according to an example embodiment. For example, electronic device 1900 may be provided as a server. Referring to FIG. 5, electronic device 1900 includes a processing component 1922 that further includes one or more processors and memory resources represented by memory 1932 for storing instructions, such as application programs, that can be executed by processing component 1922. The application programs stored in memory 1932 may include one or more modules each corresponding to a set of instructions. Further, processing component 1922 is configured to execute instructions to perform the methods described above.

The electronic device 1900 may also include a power component 1926 configured to perform power management of the electronic device 1900, a wired or wireless network interface 1950 configured to connect the electronic device 1900 to a network, and an input/output (I/O) interface 1958. The electronic device 1900 may operate based on an operating system stored in memory 1932, such as Windows Server, mac OS XTM, unixTM, linuxTM, freeBSDTM, or the like.

In an exemplary embodiment, a non-transitory computer readable storage medium is also provided, such as memory 1932, including computer program instructions executable by processing component 1922 of electronic device 1900 to perform the methods described above.

The present disclosure may be a system, method, and/or computer program product. The computer program product may include a computer readable storage medium having computer readable program instructions embodied thereon for causing a processor to implement aspects of the present disclosure.

The computer readable storage medium may be a tangible device that can hold and store instructions for use by an instruction execution device. The computer readable storage medium may be, for example, but not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium would include the following: portable computer disks, hard disks, random Access Memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static Random Access Memory (SRAM), portable compact disk read-only memory (CD-ROM), digital Versatile Disks (DVD), memory sticks, floppy disks, mechanical coding devices, punch cards or in-groove structures such as punch cards or grooves having instructions stored thereon, and any suitable combination of the foregoing. Computer-readable storage media, as used herein, are not to be construed as transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., optical pulses through fiber optic cables), or electrical signals transmitted through wires.

The computer readable program instructions described herein may be downloaded from a computer readable storage medium to a respective computing/processing device or to an external computer or external storage device over a network, such as the internet, a local area network, a wide area network, and/or a wireless network. The network may include copper transmission cables, fiber optic transmissions, wireless transmissions, routers, firewalls, switches, gateway computers and/or edge servers. The network interface card or network interface in each computing/processing device receives computer readable program instructions from the network and forwards the computer readable program instructions for storage in a computer readable storage medium in the respective computing/processing device.

Computer program instructions for performing the operations of the present disclosure can be assembly instructions, instruction Set Architecture (ISA) instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, c++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The computer readable program instructions may be executed entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer (for example, through the Internet using an Internet service provider). In some embodiments, aspects of the present disclosure are implemented by personalizing electronic circuitry, such as programmable logic circuitry, field Programmable Gate Arrays (FPGAs), or Programmable Logic Arrays (PLAs), with state information of computer readable program instructions, which can execute the computer readable program instructions.

Various aspects of the present disclosure are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the disclosure. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer-readable program instructions.

These computer readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks. These computer readable program instructions may also be stored in a computer readable storage medium that can direct a computer, programmable data processing apparatus, and/or other devices to function in a particular manner, such that the computer readable medium having the instructions stored therein includes an article of manufacture including instructions which implement the function/act specified in the flowchart and/or block diagram block or blocks.

The computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer, other programmable apparatus or other devices implement the functions/acts specified in the flowchart and/or block diagram block or blocks.

The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The foregoing description of the embodiments of the present disclosure has been presented for purposes of illustration and description, and is not intended to be exhaustive or limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the various embodiments described. The terminology used herein was chosen in order to best explain the principles of the embodiments, the practical application, or the technical improvement of the technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. The utility model provides a power equipment health grading method based on characteristic excavation technique which is characterized by comprising the following steps:

acquiring first operation parameter vectors corresponding to a plurality of electric power mechanisms in the electric power equipment and second operation parameter vectors corresponding to connecting lines between adjacent electric power mechanisms; wherein the first operating parameter vector comprises: one or more of historical overhaul times, defect levels, tripping times and abnormal times corresponding to the plurality of electric power mechanisms; wherein the second operating parameter vector comprises: one or more of historical overhaul times, defect grades and abnormal times corresponding to connecting lines among the plurality of power mechanisms;

based on the fusion parameter characteristics, completing the health grading of the power equipment by using a preset classification model; the method for completing the health grading of the power equipment based on the fusion parameter characteristics by utilizing the preset classification model comprises the following steps: acquiring a plurality of preset classification models; based on the fusion parameter characteristics, respectively utilizing a plurality of preset classification models to obtain a plurality of corresponding grading results; voting the plurality of grading results, and configuring the grading result with the highest voting as the healthy grading of the power equipment; or, the method for completing the health grading of the power equipment by using the preset classification model based on the fusion parameter characteristics comprises the following steps: acquiring a plurality of preset classification models; based on the fusion parameter characteristics, respectively utilizing a plurality of preset classification models to obtain a plurality of corresponding grading probability values; performing logistic regression on the plurality of grading probability values to finish the health grading of the power equipment; the preset classification model is configured as one of a support vector machine, a decision tree, a random forest, a K neighbor, logistic regression, self-adaptive enhancement, linear discriminant analysis, cluster analysis and a multi-layer perceptron.

2. The method for health grading of electrical equipment according to claim 1, wherein the method for fusing the first feature parameter vector and the second feature parameter vector to obtain the fused parameter feature comprises:

or alternatively, the first and second heat exchangers may be,

3. The power equipment health grading method according to any of claims 1-2, further comprising: constructing a parameter vector matrix diagram by using the first operation parameter vector and the second operation parameter vector;

4. The method for health grading of electrical equipment according to claim 3, wherein the method for fusing the first feature parameter vector, the second feature parameter vector and the third feature parameter vector to obtain fused parameter features comprises:

or alternatively, the first and second heat exchangers may be,

5. The power plant health grading method according to any of claims 1-2, 4, characterized in that the status of the power plant is monitored before the power plant health grading;

6. A method of power plant health grading according to claim 3, characterized in that the status of the power plant is monitored before the power plant health grading;

7. The power device health grading method according to claim 5, wherein the method of monitoring the status of the power device comprises:

acquiring an early warning risk value of the power equipment;

8. The power device health grading method according to claim 6, wherein the method of monitoring the status of the power device comprises:

acquiring an early warning risk value of the power equipment;

9. The utility model provides a healthy grading system of power equipment based on feature mining technique which characterized in that includes:

the acquisition unit is used for acquiring first operation parameter vectors corresponding to a plurality of electric power mechanisms in the electric power equipment and second operation parameter vectors corresponding to connecting lines between adjacent electric power mechanisms; wherein the first operating parameter vector comprises: one or more of historical overhaul times, defect levels, tripping times and abnormal times corresponding to the plurality of electric power mechanisms; wherein the second operating parameter vector comprises: one or more of historical overhaul times, defect grades and abnormal times corresponding to connecting lines among the plurality of power mechanisms;

the grading unit is used for completing the health grading of the power equipment by utilizing a preset classification model based on the fusion parameter characteristics; the method for completing the health grading of the power equipment based on the fusion parameter characteristics by utilizing the preset classification model comprises the following steps: acquiring a plurality of preset classification models; based on the fusion parameter characteristics, respectively utilizing a plurality of preset classification models to obtain a plurality of corresponding grading results; voting the plurality of grading results, and configuring the grading result with the highest voting as the healthy grading of the power equipment; or, the method for completing the health grading of the power equipment by using the preset classification model based on the fusion parameter characteristics comprises the following steps: acquiring a plurality of preset classification models; based on the fusion parameter characteristics, respectively utilizing a plurality of preset classification models to obtain a plurality of corresponding grading probability values; performing logistic regression on the plurality of grading probability values to finish the health grading of the power equipment; the preset classification model is configured as one of a support vector machine, a decision tree, a random forest, a K neighbor, logistic regression, self-adaptive enhancement, linear discriminant analysis, cluster analysis and a multi-layer perceptron.

10. The utility model provides a healthy grading system of power equipment based on feature mining technique which characterized in that includes:

a processor;

a memory for storing processor-executable instructions;

wherein the processor is configured to invoke the instructions stored in the memory to perform the method of any of claims 1 to 8.

11. A feature mining technology-based power equipment health grading system, comprising: a computer readable storage medium having stored thereon computer program instructions, which when executed by a processor, implement the method of any of claims 1 to 8.