CN115689403A - Method and device for constructing risk evaluation model of distribution network equipment - Google Patents
Method and device for constructing risk evaluation model of distribution network equipment Download PDFInfo
- Publication number
- CN115689403A CN115689403A CN202211422220.4A CN202211422220A CN115689403A CN 115689403 A CN115689403 A CN 115689403A CN 202211422220 A CN202211422220 A CN 202211422220A CN 115689403 A CN115689403 A CN 115689403A
- Authority
- CN
- China
- Prior art keywords
- distribution network
- network equipment
- risk
- risk evaluation
- evaluation model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method and a device for constructing a risk evaluation model of distribution network equipment, wherein the method comprises the following steps: the method comprises the steps of firstly establishing a risk evaluation index set of the distribution network equipment, determining risk characteristics of the distribution network equipment corresponding to the risk evaluation index set of the distribution network equipment, constructing a risk evaluation model of the distribution network equipment, training the risk evaluation model of the distribution network equipment by using a preset training method based on the risk characteristics of the distribution network equipment to obtain the trained risk evaluation model of the distribution network equipment, providing a risk evaluation model of the distribution network equipment based on a neural network and an attention mechanism, providing support for risk prediction of the subsequent distribution network equipment, greatly reducing labor cost and time cost, and improving evaluation reliability.
Description
Technical Field
The invention belongs to the field of equipment safety, and particularly relates to a method and a device for constructing a risk evaluation model of distribution network equipment.
Background
The distribution network is used as a terminal link for connecting the power system and the users, the overall efficiency of the system and the power utilization quality and reliability of the users are directly affected, and most of power failure accidents of the users are caused by power distribution system faults. The distribution network equipment is numerous and has dispersed geographic positions; the equipment state is various, the state parameter is many, the evaluation work load is big, and its running state has important influence to the power supply reliability, so it is necessary to carry out the risk evaluation to distribution network equipment.
In the prior art, the traditional risk evaluation process of the distribution network equipment only aims at single distribution network equipment, the state parameters of most of the equipment are difficult to obtain in the practical application process, the state evaluation efficiency is low, and the manpower and material resources are wasted, which is a problem to be solved urgently.
Disclosure of Invention
The invention aims to provide a method and a device for constructing a risk evaluation model of distribution network equipment, which are used for solving the defects in the prior art.
An embodiment of the present application provides a method for constructing a distribution network device risk evaluation model, where the method includes:
establishing a risk evaluation index set of distribution network equipment, wherein the distribution network equipment is obtained by dividing equipment groups according to each distribution network equipment;
determining the risk characteristics of the distribution network equipment corresponding to the risk evaluation index set of the distribution network equipment;
constructing a risk evaluation model of the distribution network equipment;
and training the risk evaluation model of the distribution network equipment by using a preset training method based on the risk characteristics of the distribution network equipment to obtain the trained risk evaluation model of the distribution network equipment, wherein the preset training method comprises a neural network and attention mechanism mixing method.
Optionally, the establishing a risk evaluation index set of the distribution network device includes:
acquiring behavior log data and risk log data of distribution network equipment, wherein the behavior log data comprise log data generated by using behaviors of the distribution network equipment, and the risk log data comprise log data recorded when the distribution network equipment generates risks;
and respectively determining the weights of the behavior log data and the risk log data of the distribution network equipment to obtain a risk evaluation index set of the distribution network equipment.
Optionally, the determining the risk characteristics of the distribution network device corresponding to the risk evaluation index set of the distribution network device includes:
acquiring service information of the distribution network equipment in the distribution network equipment risk evaluation index set, wherein the service information comprises distribution network equipment position information and distribution network equipment performance information;
clustering the position information of the distribution network equipment, and fusing the clustered position information of the distribution network equipment with the performance information of the distribution network equipment to acquire an association relation between the distribution network equipment;
and calculating the risk characteristics of the distribution network equipment based on the incidence relation and a preset risk evaluation function.
Optionally, the constructing a distribution network device risk evaluation model includes:
generating a plurality of distribution network equipment risk evaluation initial models based on a preset method, wherein the preset method comprises one or a combination of a logistic regression method, a random forest method, a nearest neighbor classification method, a K-means classification method and a Bayesian method;
and evaluating the performance of the plurality of distribution network equipment risk evaluation initial models, selecting two distribution network equipment risk evaluation initial models with the best performance, and combining to form the finally constructed distribution network equipment risk evaluation model.
Optionally, the distribution network device risk evaluation model includes:
the convolution layer, the pooling layer, the first full-connection layer and the second full-connection layer are connected in sequence;
the convolutional layer is used for extracting the characteristics of the input risk characteristic data of the distribution network equipment to obtain a preprocessed risk characteristic high-dimensional vector;
the pooling layer is used for carrying out dimensionality reduction treatment on the risk feature high-dimensional vectors and finishing maximum pooling to obtain a mixed risk feature index;
the first full connection layer is used for converting the mixed risk characteristic index into a fusion characteristic vector;
and the second full-connection layer is used for receiving the output from the first full-connection layer and updating the dimension of the fusion characteristic vector so as to obtain two classification probability values corresponding to the fusion characteristic vector.
Optionally, the calculation formula for obtaining the mixed risk characteristic index is as follows:
H t ＝δ(t i W i +H t-1 W i-1 )
wherein H t And H t-1 The output weights, W, of the pooling layers at times t and t-1, respectively i And W i-1 Output weights, t, of feature node i and feature node i-1, respectively i δ is a predetermined loss function for the output data of the feature node i.
Optionally, after the risk characteristic of the distribution network device is used to train the distribution network device risk evaluation model by using a preset training method, and the trained distribution network device risk evaluation model is obtained, the method further includes:
and inputting the risk evaluation index set of the distribution network equipment to be evaluated into the trained distribution network equipment risk evaluation model to obtain a risk evaluation result of the distribution network equipment to be evaluated.
Another embodiment of the present application provides a device for constructing a distribution network equipment risk evaluation model, where the device includes:
the system comprises an establishing module, a risk evaluation index collecting module and a risk evaluation index collecting module, wherein the risk evaluation index collecting module is used for establishing a risk evaluation index collecting set of distribution network equipment, and the distribution network equipment is obtained by dividing equipment groups according to the distribution network equipment;
the determining module is used for determining the risk characteristics of the distribution network equipment corresponding to the risk evaluation index set of the distribution network equipment;
the construction module is used for constructing a distribution network equipment risk evaluation model;
and the training module is used for training the distribution network equipment risk evaluation model by using a preset training method based on the risk characteristics of the distribution network equipment to obtain the trained distribution network equipment risk evaluation model, wherein the preset training method comprises a neural network and attention mechanism mixing method.
Optionally, the establishing module includes:
the system comprises an acquisition unit, a risk management unit and a risk management unit, wherein the acquisition unit is used for acquiring behavior log data and risk log data of the distribution network equipment, the behavior log data comprises log data generated by using the behavior of the distribution network equipment, and the risk log data comprises log data recorded when the distribution network equipment generates risks;
and the determining unit is used for respectively determining the weights of the behavior log data and the risk log data of the distribution network equipment so as to obtain a risk evaluation index set of the distribution network equipment.
Optionally, the determining module includes:
the device comprises an obtaining unit and a processing unit, wherein the obtaining unit is used for obtaining service information of the distribution network equipment in a distribution network equipment risk evaluation index set, and the service information comprises distribution network equipment position information and distribution network equipment performance information;
the clustering unit is used for clustering the position information of the distribution network equipment and fusing the clustered position information of the distribution network equipment with the performance information of the distribution network equipment so as to obtain the incidence relation among the distribution network equipment;
and the calculating unit is used for calculating the risk characteristics of the distribution network equipment based on the incidence relation and a preset risk evaluation function.
Optionally, the building module includes:
the generating unit is used for generating a plurality of distribution network equipment risk evaluation initial models based on a preset method, wherein the preset method comprises one or a combination of a logistic regression method, a random forest method, a nearest neighbor classification method, a K-means classification method and a Bayesian method;
and the evaluation unit is used for evaluating the performance of the plurality of distribution network equipment risk evaluation initial models, selecting two distribution network equipment risk evaluation initial models with the best performance, and combining to form the finally constructed distribution network equipment risk evaluation model.
Optionally, the apparatus further comprises:
and the evaluation module is used for inputting the risk evaluation index set of the distribution network equipment to be evaluated into the trained distribution network equipment risk evaluation model to obtain the risk evaluation result of the distribution network equipment to be evaluated.
A further embodiment of the application provides a storage medium having a computer program stored thereon, wherein the computer program is arranged to perform the method of any of the above when executed.
Yet another embodiment of the present application provides an electronic device comprising a memory having a computer program stored therein and a processor configured to execute the computer program to perform the method of any of the above.
Compared with the prior art, the method comprises the steps of firstly establishing a risk evaluation index set of the distribution network equipment, determining the risk characteristics of the distribution network equipment corresponding to the risk evaluation index set of the distribution network equipment, constructing a distribution network equipment risk evaluation model, training the distribution network equipment risk evaluation model based on the risk characteristics of the distribution network equipment by using a preset training method to obtain the trained distribution network equipment risk evaluation model, providing a distribution network equipment risk evaluation model based on a neural network and an attention mechanism to support the risk prediction of the subsequent distribution network equipment, greatly reducing the labor cost and the time cost, and improving the reliability of the evaluation.
Drawings
Fig. 1 is a hardware structure block diagram of a computer terminal of a method for constructing a distribution network device risk evaluation model according to an embodiment of the present invention;
fig. 2 is a schematic flow chart of a method for constructing a distribution network device risk evaluation model according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a device for constructing a distribution network equipment risk evaluation model according to an embodiment of the present invention.
Detailed Description
The embodiments described below with reference to the drawings are illustrative only and should not be construed as limiting the invention.
The embodiment of the invention firstly provides a method for constructing a risk evaluation model of distribution network equipment, and the method can be applied to electronic equipment, such as a computer terminal, in particular to a common computer, a quantum computer and the like.
This will be described in detail below by way of example as it would run on a computer terminal. Fig. 1 is a hardware structure block diagram of a computer terminal of a method for constructing a distribution network device risk evaluation model according to an embodiment of the present invention. As shown in fig. 1, the computer terminal may include one or more (only one shown in fig. 1) processors 102 (the processor 102 may include, but is not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, and optionally, a transmission device 106 for communication functions and an input-output device 108. It will be understood by those skilled in the art that the structure shown in fig. 1 is only an illustration and is not intended to limit the structure of the computer terminal. For example, the computer terminal may also include more or fewer components than shown in FIG. 1, or have a different configuration than shown in FIG. 1.
The memory 104 may be configured to store software programs and modules of application software, such as program instructions/modules corresponding to the method for constructing the distribution network device risk evaluation model in the embodiment of the present application, and the processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, so as to implement the method described above. The memory 104 may include high speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 can further include memory located remotely from the processor 102, which can be connected to a computer terminal over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The transmission device 106 is used for receiving or transmitting data via a network. Specific examples of the network described above may include a wireless network provided by a communication provider of the computer terminal. In one example, the transmission device 106 includes a Network adapter (NIC) that can be connected to other Network devices through a base station to communicate with the internet. In one example, the transmission device 106 can be a Radio Frequency (RF) module, which is used to communicate with the internet in a wireless manner.
Referring to fig. 2, fig. 2 is a schematic flow chart of a method for constructing a risk evaluation model of distribution network equipment according to an embodiment of the present invention, which may include the following steps:
s201: and establishing a risk evaluation index set of the distribution network equipment, wherein the distribution network equipment is obtained by dividing equipment groups according to each distribution network equipment.
Specifically, establishing a risk evaluation index set of the distribution network device may include:
1. the method comprises the steps of obtaining behavior log data and risk log data of the distribution network equipment, wherein the behavior log data comprise log data generated by using the behavior of the distribution network equipment, and the risk log data comprise log data recorded when the distribution network equipment generates risks.
2. And respectively determining the weights of the behavior log data and the risk log data of the distribution network equipment to obtain a risk evaluation index set of the distribution network equipment.
In an optional implementation mode, the terminal communicates with the log server through a network, and the terminal acquires the behavior log data and the risk log data in the log server. The terminal performs integrated screening processing on the behavior log data and the risk log data, and can determine the weights of the behavior log data and the risk log data of the distribution network equipment by combining the intentions of management personnel or by an entropy weight method to obtain a risk evaluation index set of the distribution network equipment.
The terminal can also determine a risk evaluation index set of the distribution network equipment according to a risk triggering condition, and obtain behavior log data and risk log data, wherein the behavior log data includes log data generated by using the behavior of the distribution network equipment, and the behavior log data can be corresponding log data generated by certain behavior operations, such as query and deletion, performed by a user using the distribution network equipment system. The risk log data comprises log data recorded when the distribution network equipment is used for generating risks, and the risk log data is generally log data recorded in a system when the distribution network equipment generates the risks. The risk may include: and the equipment has risks such as equipment failure, offline abnormal events, overload of distribution network equipment and the like. The risk of the distribution network equipment is usually triggered by the fact that a manager logs in a certain distribution network equipment management system due to the fact that some point or multiple points are abnormal.
It should be noted that, when the distribution network device management system is used by a manager, the manager remotely monitors the external distribution network device or performs analog monitoring in a digital twin manner, so as to generate behavior log data and risk log data of the corresponding distribution network device, then performs integrated screening processing on the behavior log data and the risk log data, determines weights of the behavior log data and the risk log data of the distribution network device, and obtains a risk evaluation index set of the distribution network device. The integration and screening processing can be a processing mode of integrating behavior log data and risk log data and screening the log data obtained after integration so as to meet the requirement of modeling data for subsequently constructing a distribution network equipment risk evaluation model. The behavior log data and the risk log data may be integrated, for example, based on identification information or other data that may bind the behavior log data and the risk log data together in some relationship. And screening the integrated data to obtain data which accords with the construction of the risk evaluation model of the distribution network equipment.
S202: and determining the risk characteristics of the distribution network equipment corresponding to the risk evaluation index set of the distribution network equipment.
Specifically, determining the risk characteristics of the distribution network device corresponding to the risk evaluation index set of the distribution network device may include:
step 1: and obtaining service information of the distribution network equipment in the distribution network equipment risk evaluation index set, wherein the service information comprises distribution network equipment position information and distribution network equipment performance information.
Step 2: clustering the position information of the distribution network equipment, and fusing the clustered position information of the distribution network equipment with the performance information of the distribution network equipment to acquire the association relationship between the distribution network equipment.
And step 3: and calculating the risk characteristics of the distribution network equipment based on the incidence relation and a preset risk evaluation function.
For example, the service information of the distribution network device may include various data such as the number of devices, the model of devices, the status of devices, the operation data of devices, the location information of devices, the service information of devices, and the performance information of devices. For convenience of understanding, the device information is not further limited in the present application, and only the distribution network device location information and the distribution network device performance information used in the embodiments of the present application are exemplarily described.
For example, the expression of the service information of the distribution network device may be
Wherein the content of the first and second substances,
device location information indicating the ith distribution network device at time t,
and indicating the device performance information of the ith device at the time t. And clustering the position information of the distribution network equipment to obtain a plurality of position cluster clusters.
In an optional embodiment, a DBSCAN (sensitivity-Based spot Clustering of Applications with Noise) method is used to cluster the position information of the multiple devices, and a plurality of position cluster clusters are obtained. Each location cluster corresponds to one device cluster, and therefore each location cluster contains device location information of each device in the device cluster.
In another alternative embodiment, some network distribution devices are geographically distant but perform closely related or identical services. In order to realize the fusion of the service information of the distribution network equipment and find the incidence relation among different distribution network equipment, the position information of the distribution network equipment after the clustering is finished and the performance information of the distribution network equipment can be fused by adopting an encoder. An Auto Encoder (AE) is a type of artificial neural network used in semi-supervised learning and unsupervised learning, and functions to perform characterization learning on input information by using the input information as a learning target; and finally, calculating the risk characteristics of the distribution network equipment based on the incidence relation and a preset risk evaluation function. The preset risk evaluation function may be:
v j denotes the jth distribution network device, v j ∈N(v i ) Denotes that the jth distribution network device is adjacent to the ith distribution network device, w ij And representing the association weight between the equipment service information of the ith distribution network equipment and the equipment service information of the jth distribution network equipment.
It should be noted that the risk characteristics of the distribution network equipment include a line overload risk characteristic, a transformer overload risk characteristic, a low voltage risk characteristic, an overvoltage risk characteristic, a load loss risk characteristic, a voltage out-of-limit risk characteristic, a voltage collapse risk characteristic, a transient risk characteristic, and the like, which are not listed herein.
S203: and constructing a distribution network equipment risk evaluation model.
Specifically, constructing a distribution network equipment risk evaluation model may include:
s2031: generating a plurality of distribution network equipment risk evaluation initial models based on a preset method, wherein the preset method comprises one or a combination of a logistic regression method, a random forest method, a nearest neighbor classification method, a K-means classification method and a Bayesian method.
S2032: and evaluating the performance of the plurality of distribution network equipment risk evaluation initial models, selecting two distribution network equipment risk evaluation initial models with the best performance, and combining to form the finally constructed distribution network equipment risk evaluation model.
Illustratively, a plurality of distribution network equipment risk evaluation initial models are obtained based on one or a combination of a preset logistic regression method, a random forest method, a nearest neighbor classification method, a K-means classification method and a Bayesian method.
The method comprises the steps of carrying out normalization index processing on feature information output by a feature layer of a plurality of distribution network equipment risk evaluation initial models to obtain a probability distribution interval, then carrying out normalization processing to obtain a maximum mean difference to form a target function, finally respectively processing the plurality of distribution network equipment risk evaluation initial models according to the target function, and jointly forming a finally constructed distribution network equipment risk evaluation model by comparing the performance of the processed plurality of distribution network equipment risk evaluation initial models and selecting two distribution network equipment risk evaluation initial models with the best performance.
In an optional implementation manner, the distribution network device risk evaluation model may include: the convolution layer, the pooling layer, the first full-connection layer and the second full-connection layer are connected in sequence;
the convolutional layer is used for extracting the characteristics of the input risk characteristic data of the distribution network equipment to obtain a preprocessed risk characteristic high-dimensional vector; the pooling layer is used for carrying out dimensionality reduction treatment on the risk feature high-dimensional vectors and finishing maximum pooling to obtain a mixed risk feature index; the first full connection layer is used for converting the mixed risk characteristic index into a fusion characteristic vector; and the second full-connection layer is used for receiving the output from the first full-connection layer and updating the dimension of the fusion characteristic vector so as to obtain two classification probability values corresponding to the fusion characteristic vector.
In an optional implementation manner, the distribution network device risk evaluation model may include: the sequentially connected convolution layers comprise a first convolution layer and a second convolution layer; the pooling layers include a first pooling layer and a second pooling layer, a first fully-connected layer and a second fully-connected layer.
The formula for performing convolution operation on the first convolution layer and the second convolution layer is as follows:
β i ＝f(Uz i +h-1+b j )
in the formula beta i Representing the eigenvectors obtained by the i-th convolution operation, f being the activation function, U representing the weight matrix of the convolution kernel, z i Representing the input of the ith neuron of the risk evaluation model of the distribution network equipment, h is the height of a convolution kernel, b j A bias parameter representing a jth convolution kernel; and (3) completing the maximum pooling through the operation of a pooling layer, and obtaining the final characteristic expression of the mixed risk characteristic index as follows: t = max { beta { 1 、β 2 、…、β n -h +1}; and finally, the prediction evaluation results output by the first full-connection layer and the second full-connection layer are as follows:
wherein y represents the result of the prediction evaluation,
weight matrix representing a first fully-connected layer and a second fully-connected layer, b m Representing the bias parameters of the first fully-connected layer and the second fully-connected layer.
Wherein, the calculation formula for obtaining the mixed risk characteristic index is as follows:
H t ＝δ(t i W i +H t-1 W i-1 )
wherein H t And H t-1 The output weights, W, of the pooling layers at times t and t-1, respectively i And W i-1 Output weights, t, of feature node i and feature node i-1, respectively i δ is a predetermined loss function for the output data of the feature node i.
S204: and training the distribution network equipment risk evaluation model by using a preset training method based on the risk characteristics of the distribution network equipment to obtain the trained distribution network equipment risk evaluation model, wherein the preset training method comprises a neural network and attention mechanism mixing method.
Specifically, a neural network and attention mechanism hybrid method is used for risk evaluation of the distribution network equipment, in this embodiment, a residual error module may be provided for a convolution layer of a distribution network equipment risk evaluation model, the distribution network equipment risk evaluation model provided with the residual error module is trained through training samples in training data, specifically, each training sample may be input into the distribution network equipment risk evaluation model, and an evaluation result corresponding to the training sample output by the distribution network equipment risk evaluation model is obtained; calculating a residual error function value according to the evaluation result and the sample label of the training sample; and adjusting network parameters of the distribution network equipment risk evaluation model according to the residual error function value. Whether the model is trained or not can be judged by judging that the training times of the distribution network equipment risk evaluation model are larger than a preset training time threshold, or judging that the prediction accuracy of the distribution network equipment risk evaluation model is larger than a preset accuracy threshold, or by judging that the residual function of the distribution network equipment risk evaluation model is convergent, and the residual function is not limited herein.
Exemplarily, a residual module is provided for the convolutional layer of the distribution network equipment risk evaluation model, and a residual function is required to be provided for this purpose, as follows:
F(x)＝H(x)-x
as long as F (x) =0, an identity map H (x) = x can be constructed, where H (x) is the network map of the inputs to the summation and F (x) is the residual function.
In an optional implementation manner, after the risk characteristic of the distribution network device is used to train the risk evaluation model of the distribution network device by using a preset training method, and the trained risk evaluation model of the distribution network device is obtained, the method further includes: and inputting the risk evaluation index set of the distribution network equipment to be evaluated into the trained distribution network equipment risk evaluation model to obtain a risk evaluation result of the distribution network equipment to be evaluated. For example, if the risk evaluation of the distribution network device to be evaluated is greater than a preset risk threshold, a risk alarm condition that the risk evaluation result of the distribution network device satisfies may be further determined.
The risk evaluation model of the distribution network equipment is trained by a preset training method based on the risk characteristics of the distribution network equipment to obtain the trained risk evaluation model of the distribution network equipment.
Another embodiment of the present application provides a device for constructing a risk evaluation model of a distribution network device, such as a schematic structural diagram of the device for constructing a risk evaluation model of a distribution network device shown in fig. 3, where the device includes:
the establishing module 301 is configured to establish a risk evaluation index set of distribution network devices, where the distribution network devices perform device group division according to each distribution network device;
a determining module 302, configured to determine a risk characteristic of the distribution network device corresponding to the risk evaluation index set of the distribution network device;
the building module 303 is used for building a distribution network equipment risk evaluation model;
the training module 304 is configured to train the distribution network equipment risk evaluation model by using a preset training method based on the risk characteristics of the distribution network equipment to obtain the trained distribution network equipment risk evaluation model, where the preset training method includes a neural network and attention system mixing method.
Specifically, the establishing module includes:
the system comprises an acquisition unit, a risk management unit and a risk management unit, wherein the acquisition unit is used for acquiring behavior log data and risk log data of the distribution network equipment, the behavior log data comprises log data generated by using the behavior of the distribution network equipment, and the risk log data comprises log data recorded when the distribution network equipment generates risks;
and the determining unit is used for respectively determining the weights of the behavior log data and the risk log data of the distribution network equipment so as to obtain a risk evaluation index set of the distribution network equipment.
Specifically, the determining module includes:
the device comprises an obtaining unit and a processing unit, wherein the obtaining unit is used for obtaining service information of the distribution network equipment in a distribution network equipment risk evaluation index set, and the service information comprises distribution network equipment position information and distribution network equipment performance information;
the clustering unit is used for clustering the position information of the distribution network equipment and fusing the clustered position information of the distribution network equipment with the performance information of the distribution network equipment so as to obtain the association relationship among the distribution network equipment;
and the calculating unit is used for calculating the risk characteristics of the distribution network equipment based on the incidence relation and a preset risk evaluation function.
Specifically, the building module includes:
the generating unit is used for generating a plurality of distribution network equipment risk evaluation initial models based on a preset method, wherein the preset method comprises one or a combination of a logistic regression method, a random forest method, a nearest neighbor classification method, a K-means classification method and a Bayesian method;
and the evaluation unit is used for evaluating the performance of the plurality of distribution network equipment risk evaluation initial models, selecting two distribution network equipment risk evaluation initial models with the best performance, and combining to form the finally constructed distribution network equipment risk evaluation model.
Specifically, the apparatus further comprises:
and the evaluation module is used for inputting the risk evaluation index set of the distribution network equipment to be evaluated into the trained distribution network equipment risk evaluation model to obtain the risk evaluation result of the distribution network equipment to be evaluated.
Compared with the prior art, the risk evaluation method of the distribution network equipment comprises the steps of firstly establishing a risk evaluation index set of the distribution network equipment, determining the risk characteristics of the distribution network equipment corresponding to the risk evaluation index set of the distribution network equipment, constructing a risk evaluation model of the distribution network equipment, training the risk evaluation model of the distribution network equipment based on the risk characteristics of the distribution network equipment by using a preset training method to obtain the trained risk evaluation model of the distribution network equipment.
An embodiment of the present invention further provides a storage medium, in which a computer program is stored, where the computer program is configured to execute the steps in any of the above method embodiments when running.
Specifically, in the present embodiment, the storage medium may be configured to store a computer program for executing the steps of:
s201: establishing a risk evaluation index set of distribution network equipment, wherein the distribution network equipment is obtained by dividing equipment groups according to each distribution network equipment;
s202: determining the risk characteristics of the distribution network equipment corresponding to the risk evaluation index set of the distribution network equipment;
s203: constructing a risk evaluation model of the distribution network equipment;
s204: and training the distribution network equipment risk evaluation model by using a preset training method based on the risk characteristics of the distribution network equipment to obtain the trained distribution network equipment risk evaluation model, wherein the preset training method comprises a neural network and attention mechanism mixing method.
Specifically, in this embodiment, the storage medium may include, but is not limited to: various media capable of storing computer programs, such as a usb disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic disk, or an optical disk.
Compared with the prior art, the method comprises the steps of firstly establishing a risk evaluation index set of the distribution network equipment, determining the risk characteristics of the distribution network equipment corresponding to the risk evaluation index set of the distribution network equipment, constructing a distribution network equipment risk evaluation model, training the distribution network equipment risk evaluation model based on the risk characteristics of the distribution network equipment by using a preset training method to obtain the trained distribution network equipment risk evaluation model, providing a distribution network equipment risk evaluation model based on a neural network and an attention mechanism to support the risk prediction of the subsequent distribution network equipment, greatly reducing the labor cost and the time cost, and improving the reliability of the evaluation.
An embodiment of the present invention further provides an electronic apparatus, which includes a memory and a processor, where the memory stores a computer program, and the processor is configured to execute the computer program to perform the steps in any of the above method embodiments.
Specifically, the electronic apparatus may further include a transmission device and an input/output device, wherein the transmission device is connected to the processor, and the input/output device is connected to the processor.
Specifically, in this embodiment, the processor may be configured to execute the following steps by a computer program:
s201: establishing a risk evaluation index set of distribution network equipment, wherein the distribution network equipment is obtained by dividing equipment groups according to each distribution network equipment;
s202: determining the risk characteristics of the distribution network equipment corresponding to the risk evaluation index set of the distribution network equipment;
s203: constructing a risk evaluation model of the distribution network equipment;
s204: and training the distribution network equipment risk evaluation model by using a preset training method based on the risk characteristics of the distribution network equipment to obtain the trained distribution network equipment risk evaluation model, wherein the preset training method comprises a neural network and attention mechanism mixing method.
Compared with the prior art, the method comprises the steps of firstly establishing a risk evaluation index set of the distribution network equipment, determining the risk characteristics of the distribution network equipment corresponding to the risk evaluation index set of the distribution network equipment, constructing a distribution network equipment risk evaluation model, training the distribution network equipment risk evaluation model based on the risk characteristics of the distribution network equipment by using a preset training method to obtain the trained distribution network equipment risk evaluation model, providing a distribution network equipment risk evaluation model based on a neural network and an attention mechanism to support the risk prediction of the subsequent distribution network equipment, greatly reducing the labor cost and the time cost, and improving the reliability of the evaluation.
It should be noted that, for simplicity of description, the above-mentioned method embodiments are described as a series of acts or combination of acts, but those skilled in the art will recognize that the present invention is not limited by the order of acts, as some steps may occur in other orders or concurrently in accordance with the invention. Further, those skilled in the art should also appreciate that the embodiments described in the specification are preferred embodiments and that the acts and modules referred to are not necessarily required by the invention.
In the foregoing embodiments, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.
In the embodiments provided in the present invention, it should be understood that the disclosed apparatus can be implemented in other manners. For example, the above-described embodiments of the apparatus are merely illustrative, and for example, the above-described division of the units is only one type of division of logical functions, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of some interfaces, devices or units, and may be an electric or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit may be stored in a computer readable memory if it is implemented in the form of a software functional unit and sold or used as a stand-alone product. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a memory and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the above methods according to the embodiments of the present invention. And the aforementioned memory comprises: a U-disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic or optical disk, and other various media capable of storing program codes.
The above embodiments of the present invention are described in detail, and the principle and the implementation of the present invention are explained by applying specific embodiments, and the above description of the embodiments is only used to help understanding the method of the present invention and the core idea thereof; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
Claims (10)
1. A method for constructing a risk evaluation model of distribution network equipment is characterized by comprising the following steps:
establishing a risk evaluation index set of distribution network equipment, wherein the distribution network equipment is obtained by dividing equipment groups according to each distribution network equipment;
determining the risk characteristics of the distribution network equipment corresponding to the risk evaluation index set of the distribution network equipment;
constructing a risk evaluation model of the distribution network equipment;
and training the distribution network equipment risk evaluation model by using a preset training method based on the risk characteristics of the distribution network equipment to obtain the trained distribution network equipment risk evaluation model, wherein the preset training method comprises a neural network and attention mechanism mixing method.
2. The method of claim 1, wherein establishing the set of risk assessment indicators for the distribution network device comprises:
acquiring behavior log data and risk log data of distribution network equipment, wherein the behavior log data comprise log data generated by using the behavior of the distribution network equipment, and the risk log data comprise log data recorded when the distribution network equipment generates risks;
and respectively determining the weights of the behavior log data and the risk log data of the distribution network equipment to obtain a risk evaluation index set of the distribution network equipment.
3. The method of claim 2, wherein the determining the risk characteristics of the distribution network device corresponding to the set of risk evaluation indicators of the distribution network device comprises:
acquiring service information of the distribution network equipment in the distribution network equipment risk evaluation index set, wherein the service information comprises distribution network equipment position information and distribution network equipment performance information;
clustering the position information of the distribution network equipment, and fusing the clustered position information of the distribution network equipment with the performance information of the distribution network equipment to obtain an incidence relation between the distribution network equipment;
and calculating the risk characteristics of the distribution network equipment based on the incidence relation and a preset risk evaluation function.
4. The method of any one of claims 1 to 3, wherein the constructing of the distribution network equipment risk assessment model comprises:
generating a plurality of distribution network equipment risk evaluation initial models based on a preset method, wherein the preset method comprises one or the combination of a logistic regression method, a random forest method, a nearest neighbor classification method, a K-means classification method and a Bayesian method;
and evaluating the performance of the plurality of distribution network equipment risk evaluation initial models, selecting two distribution network equipment risk evaluation initial models with the best performance, and combining to form the finally constructed distribution network equipment risk evaluation model.
5. The method of claim 4, wherein the distribution network equipment risk assessment model comprises:
the convolution layer, the pooling layer, the first full-connection layer and the second full-connection layer are connected in sequence;
the convolutional layer is used for extracting the characteristics of the input risk characteristic data of the distribution network equipment to obtain a preprocessed risk characteristic high-dimensional vector;
the pooling layer is used for carrying out dimensionality reduction treatment on the risk feature high-dimensional vectors and finishing maximum pooling to obtain a mixed risk feature index;
the first full connection layer is used for converting the mixed risk characteristic index into a fusion characteristic vector;
and the second full-connection layer is used for receiving the output from the first full-connection layer and updating the dimension of the fusion characteristic vector so as to obtain two classification probability values corresponding to the fusion characteristic vector.
6. The method according to claim 5, wherein the calculation formula for obtaining the hybrid risk characteristic indicator is:
H t ＝δ(t i W i +H t-1 W i-1 )
wherein H t And H t-1 The output weights, W, of the pooling layers at times t and t-1, respectively i And W i-1 Output weights, t, of feature node i and feature node i-1, respectively i δ is a predetermined loss function for the output data of the feature node i.
7. The method according to claim 1, wherein after the risk evaluation model of the distribution network equipment is trained by using a preset training method based on the risk characteristics of the distribution network equipment and the trained risk evaluation model of the distribution network equipment is obtained, the method further comprises:
and inputting the risk evaluation index set of the distribution network equipment to be evaluated into the trained distribution network equipment risk evaluation model to obtain a risk evaluation result of the distribution network equipment to be evaluated.
8. A device for constructing a risk evaluation model of distribution network equipment is characterized by comprising:
the system comprises an establishing module, a risk evaluation index setting module and a risk evaluation index setting module, wherein the risk evaluation index setting module is used for establishing a risk evaluation index set of distribution network equipment, and the distribution network equipment is obtained by dividing equipment groups according to each distribution network equipment;
the determining module is used for determining the risk characteristics of the distribution network equipment corresponding to the risk evaluation index set of the distribution network equipment;
the construction module is used for constructing a distribution network equipment risk evaluation model;
and the training module is used for training the distribution network equipment risk evaluation model by using a preset training method based on the risk characteristics of the distribution network equipment to obtain the trained distribution network equipment risk evaluation model, wherein the preset training method comprises a neural network and attention mechanism mixing method.
9. A storage medium, in which a computer program is stored, wherein the computer program is arranged to perform the method of any of claims 1 to 7 when executed.
10. An electronic device comprising a memory and a processor, wherein the memory has stored therein a computer program, and wherein the processor is arranged to execute the computer program to perform the method of any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211422220.4A CN115689403A (en) | 2022-11-14 | 2022-11-14 | Method and device for constructing risk evaluation model of distribution network equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211422220.4A CN115689403A (en) | 2022-11-14 | 2022-11-14 | Method and device for constructing risk evaluation model of distribution network equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115689403A true CN115689403A (en) | 2023-02-03 |
Family
ID=85052343
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211422220.4A Pending CN115689403A (en) | 2022-11-14 | 2022-11-14 | Method and device for constructing risk evaluation model of distribution network equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115689403A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117273516A (en) * | 2023-09-15 | 2023-12-22 | 国网湖北省电力有限公司黄石供电公司 | Performance evaluation method based on attention mechanism neural network |
-
2022
- 2022-11-14 CN CN202211422220.4A patent/CN115689403A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117273516A (en) * | 2023-09-15 | 2023-12-22 | 国网湖北省电力有限公司黄石供电公司 | Performance evaluation method based on attention mechanism neural network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104601438A (en) | Friend recommendation method and device | |
| CN111611488B (en) | Information recommendation method and device based on artificial intelligence and electronic equipment | |
| WO2023185539A1 (en) | Machine learning model training method, service data processing method, apparatuses, and systems | |
| WO2021068513A1 (en) | Abnormal object recognition method and apparatus, medium, and electronic device | |
| CN111652661B (en) | Mobile phone client user loss early warning processing method | |
| CN115689403A (en) | Method and device for constructing risk evaluation model of distribution network equipment | |
| WO2019209231A2 (en) | System and method for creating recommendation of splitting and merging microservice | |
| CN111797320A (en) | Data processing method, device, equipment and storage medium | |
| CN112085869A (en) | Civil aircraft flight safety analysis method based on flight parameter data | |
| CN114091610A (en) | Intelligent decision method and device | |
| CN110378739B (en) | Data traffic matching method and device | |
| CN115965058A (en) | Neural network training method, entity information classification method, device and storage medium | |
| CN113541986B (en) | Fault prediction method and device for 5G slice and computing equipment | |
| CN116541166A (en) | Super-computing power scheduling server and resource management method | |
| CN112528132A (en) | Method for managing network and network management system | |
| CN114880406A (en) | Data management method and device | |
| CN114330720A (en) | Knowledge graph construction method and device for cloud computing and storage medium | |
| CN115907056A (en) | Prediction model training method, information prediction method and corresponding devices | |
| CN113965475B (en) | Network slice engineering acceptance method and system | |
| CN117273963B (en) | Risk identification method and device based on car insurance scene | |
| US20230018893A1 (en) | Multitask distributed learning system and method based on lottery ticket neural network | |
| CN112104467B (en) | Cutover operation risk rating method and device and computing equipment | |
| LIU et al. | A case study on intelligent operation system for wireless networks | |
| CN116611506B (en) | User analysis model training method, user label determining method and device | |
| CN116720167B (en) | Role management system for multiple isolation management |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |