Power facility monitoring method, system, device and storage medium

Abstract

Embodiments of the present invention disclose a method, system, device, and storage medium for monitoring electric power facilities, belonging to the field of electric power facility technology. The method includes obtaining grid location information; the grid location information includes target location information of at least one target facility; classifying and processing the grid location information using a trained settlement discrimination model to obtain settlement information of the target facility; the settlement discrimination model is trained using a grid location training set; if the settlement information is determined to match a preset settlement condition, generating a settlement status corresponding to the target facility and using the settlement status as a monitoring result. Using a neural network model, namely a settlement discrimination model, the grid location information is processed to obtain the settlement status of each target facility, making it less likely that a target facility will be missed, thereby improving the quality of settlement monitoring of electric power facilities.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power facilities, and in particular to an electric power facility monitoring method, system, device and storage medium. Background Art

[0002] Power facilities can be divided into power generation facilities, substation facilities, and transmission facilities. Transmission facilities include poles, transmission lines, transformers, and grounding devices. Due to the geographical dispersion of power facilities and the numerous dynamic factors influencing them, especially the mountainous, densely forested environment where transmission lines are located and the frequent occurrence of geological disasters, power facilities are affected by these factors and experience subsidence, posing a significant risk to their normal operation.

[0003] In order to detect the subsidence of power facilities in a timely manner, traditional monitoring methods generally use manual inspections, in which professional inspectors conduct regular inspections to determine the location and working conditions of power facilities. However, manual inspections are inefficient, have poor real-time performance, and are prone to missed inspections, resulting in a decline in the monitoring quality of power facilities. Summary of the Invention

[0004] In view of this, the present invention provides a method, system, device, and storage medium for monitoring electric power facilities, which are used to solve the problem of poor monitoring quality in the prior art. To achieve one, part, or all of the above-mentioned objectives or other objectives, the present invention provides a method, system, device, and storage medium for monitoring electric power facilities. In a first aspect,

[0005] A method for monitoring electric power facilities, comprising:

[0006] Acquiring power grid positioning information; the power grid positioning information includes target positioning information of at least one target facility;

[0007] Using the trained settlement discrimination model to classify the power grid location information to obtain the settlement information of the target facility; the settlement discrimination model is trained using the power grid location training set;

[0008] If it is determined that the settlement information matches the preset settlement condition, a settlement status corresponding to the target facility is generated and the settlement status is used as a monitoring result.

[0009] Preferably, before classifying the power grid location information using the trained settlement discrimination model, the monitoring method further includes:

[0010] Utilizing the autoencoding neural network model in the settlement discrimination model to calculate the power grid positioning training set, to obtain settlement training parameters;

[0011] Calculating the settlement training parameters using the support vector machine classification model in the settlement discrimination model to obtain settlement state parameters;

[0012] When the settlement state parameters match the preset training conditions, it is determined that the settlement discrimination model training is completed.

[0013] Preferably, the step of classifying the power grid location information using the trained settlement discrimination model to obtain the settlement information of the target facility includes:

[0014] Obtaining a positioning steering vector parameter in the power grid positioning information; each target positioning information corresponds to the positioning steering vector parameter;

[0015] Using the trained autoencoder neural network model to encode and decode the positioning guidance vector parameters to obtain feature guidance vector parameters;

[0016] The trained support vector machine classification model is used to classify the feature-oriented vector parameters to obtain the settlement information of each target facility.

[0017] Preferably, the step of encoding and decoding the positioning steering vector parameters using the trained autoencoder neural network model to obtain feature steering vector parameters includes:

[0018] Encoding the positioning and steering vector parameters using an encoder in the autoencoding neural network model to obtain encoding feature parameters;

[0019] Decoding the training feature parameters using a decoder in the autoencoding neural network model to obtain imitation vector parameters;

[0020] The encoding feature parameters are extracted and used as the feature-guided vector parameters.

[0021] Preferably, the step of obtaining grid location information includes:

[0022] Acquiring the target positioning information of each target facility;

[0023] Extracting facility coding parameters and longitude and latitude coordinate parameters from the target positioning information;

[0024] Calculate the number of facilities based on all the facility coding parameters;

[0025] Arranging the longitude and latitude coordinate parameters according to the corresponding relationship between the number of facilities and the facility coding parameters to obtain a signal source array;

[0026] Obtaining distribution direction parameters of each of the target facilities based on the latitude and longitude coordinate parameters;

[0027] The signal source array and the distribution direction parameters are calculated using a multi-source information fusion positioning matrix model to obtain the power grid positioning information.

[0028] Preferably, before classifying the power grid location information using the trained settlement discrimination model, the method further includes:

[0029] Constructing identification space parameters according to the signal source array in the power grid positioning information; the identification space parameters include the positioning steering vector parameters;

[0030] Generate characteristic vector parameters and use the characteristic vector parameters to project within the identification space parameters to obtain projection vector parameters;

[0031] The spatial coefficient parameters are obtained by processing the eigenvector parameters, the projection vector parameters and the positioning guidance vector parameters;

[0032] Determining whether the spatial coefficient parameter of each of the positioning and steering vector parameters matches a preset discrimination value;

[0033] If so, the corresponding positioning steering vector parameter is determined to be noise and is excluded; if not, the corresponding positioning steering vector parameter is retained to denoise and update the power grid positioning information.

[0034] Preferably, the step of obtaining the spatial coefficient parameters by processing the eigenvector parameters, the projection vector parameters and the positioning guidance vector parameters comprises:

[0035] When the inner product of the projection vector parameter and any of the positioning steering vector parameters is equal to the inner product of the eigenvector parameter and the corresponding positioning steering vector parameter, the spatial coefficient parameter corresponding to the positioning steering vector parameter is calculated using the corresponding positioning steering vector parameter and the eigenvector parameter.

[0036] Second aspect:

[0037] A power facility monitoring system includes an acquisition module for acquiring power grid positioning information; the power grid positioning information includes target positioning information of at least one target facility;

[0038] a settlement module, configured to classify the power grid location information using a trained settlement discrimination model to obtain settlement information of the target facility; the settlement discrimination model is trained using a power grid location training set;

[0039] The judgment module is used to generate a settlement status corresponding to the target facility if it is judged that the settlement information matches the preset settlement condition, and use the settlement status as a monitoring result.

[0040] Preferably, the system further comprises a training module for calculating the power grid location training set using the autoencoder neural network model in the subsidence discrimination model to obtain subsidence training parameters before classifying the power grid location information using the trained subsidence discrimination model;

[0041] Calculating the settlement training parameters using the support vector machine classification model in the settlement discrimination model to obtain settlement state parameters;

[0042] When the settlement state parameters match the preset training conditions, it is determined that the settlement discrimination model training is completed.

[0043] Preferably, the settlement module includes an extraction unit for obtaining positioning steering vector parameters in the power grid positioning information; each of the target positioning information corresponds to the positioning steering vector parameters;

[0044] an encoding and decoding unit, configured to encode and decode the positioning steering vector parameters using the trained autoencoding neural network model to obtain feature steering vector parameters;

[0045] A discrimination unit is used to classify the feature-oriented vector parameters using the trained support vector machine classification model to obtain the settlement information of each target facility.

[0046] Preferably, the encoding and decoding unit includes an encoding subunit, which is used to encode the positioning steering vector parameters using an encoder in the autoencoding neural network model to obtain encoding feature parameters;

[0047] A decoding subunit, configured to decode the training feature parameters using a decoder in the autoencoding neural network model to obtain imitation vector parameters;

[0048] The extraction subunit is configured to extract the encoding feature parameters and use the encoding feature parameters as the feature-guided vector parameters.

[0049] Preferably, the acquisition module includes a target positioning unit, configured to acquire the target positioning information of each target facility;

[0050] A parameter unit, configured to extract facility coding parameters and latitude and longitude coordinate parameters from the target positioning information;

[0051] a summing unit, configured to calculate the number of facilities according to all the facility coding parameters;

[0052] An array unit, configured to arrange the longitude and latitude coordinate parameters according to a correspondence between the number of facilities and the facility coding parameters to obtain a signal source array;

[0053] a direction unit, configured to obtain distribution direction parameters of each of the target facilities based on the latitude and longitude coordinate parameters;

[0054] A calculation unit is used to calculate the signal source array and the distribution direction parameters using a multi-source information fusion positioning matrix model to obtain the power grid positioning information.

[0055] Preferably, the system further comprises a denoising module, which constructs identification space parameters according to the signal source array in the grid positioning information before classifying the grid positioning information using the trained settlement discrimination model; the identification space parameters include the positioning steering vector parameters;

[0056] Generate characteristic vector parameters and use the characteristic vector parameters to project within the identification space parameters to obtain projection vector parameters;

[0057] The spatial coefficient parameters are obtained by processing the eigenvector parameters, the projection vector parameters and the positioning guidance vector parameters;

[0058] Determining whether the spatial coefficient parameter of each of the positioning and steering vector parameters matches a preset discrimination value;

[0059] If so, the corresponding positioning steering vector parameter is determined to be noise and is excluded; if not, the corresponding positioning steering vector parameter is retained to denoise and update the power grid positioning information.

[0060] Preferably, the denoising module includes a processing unit for calculating the spatial coefficient parameter corresponding to the positioning steering vector parameter using the corresponding positioning steering vector parameter and the eigenvector parameter when the inner product of the projection vector parameter and any positioning steering vector parameter is equal to the inner product of the eigenvector parameter and the corresponding positioning steering vector parameter.

[0061] The third aspect:

[0062] An electric power facility monitoring device includes a memory and a processor. The memory stores an electric power facility monitoring method. The processor adopts the above-mentioned method when executing the electric power facility monitoring method.

[0063] Fourth aspect:

[0064] A storage medium stores a computer program that can be loaded by a processor and execute the above method.

[0065] The implementation of the present invention will have the following beneficial effects:

[0066] By integrating the target location information of all target facilities into grid location information, and then classifying and processing the grid location information using a neural network model, known as a subsidence discrimination model, the system determines whether each target facility has experienced subsidence, thereby generating monitoring results for each target facility. This approach, coupled with the fact that the number of target facilities is known, makes it less likely that a target facility will be missed when integrating the grid location information. Furthermore, the entire monitoring process requires no human intervention. Therefore, compared to manual inspections and monitoring, this approach improves the efficiency and quality of power facility monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0068] in:

[0069] Figure 1 FIG. 4 is a flow chart of a method for monitoring electric power facilities in one embodiment.

[0070] Figure 2 A flowchart of a method for monitoring electric power facilities in one embodiment for training a settlement discrimination model.

[0071] Figure 3 The figure is a flowchart of obtaining grid location information of a method for monitoring electric power facilities in one embodiment.

[0072] Figure 4 FIG. 4 is a structural block diagram of a power facility monitoring system in one embodiment.

[0073] Figure 5 FIG. 1 is a schematic diagram of the structure of a power facility monitoring device in one embodiment. DETAILED DESCRIPTION

[0074] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0075] In the following description, reference is made to “some embodiments”, which describes a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0076] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention pertains. The terms used herein are for the purpose of describing embodiments of the present invention only and are not intended to limit the present invention.

[0077] The embodiment of the present application discloses a method for monitoring power facilities. Power facilities are the basis for transporting electricity. Regular monitoring of power facilities not only helps to ensure the stable transmission of electricity, but also helps to avoid damage to power facilities. Due to geographical influences, most power facilities are located in high mountains and dense forests. Once subsidence occurs, it will not only endanger the transmission of electricity, but also make it more prone to accidents such as leakage and fire. The existing monitoring methods for power facilities generally use manual inspections, in which professional inspectors conduct regular inspections to determine the location and working conditions of power facilities. However, the working method of manual inspections is inefficient, has poor real-time performance, and is prone to missed inspections, resulting in a decline in the monitoring quality of power facilities.

[0078] In order to overcome the above-mentioned defects, the present application discloses a method for monitoring electric power facilities, such as Figure 1 Shown, including:

[0079] 101. Acquire power grid positioning information; the power grid positioning information includes target positioning information of at least one target facility.

[0080] In one embodiment, the power grid positioning information is composed of target positioning information of several target facilities. The target facilities refer to power facilities that require settlement monitoring, such as poles and towers, and the target positioning information refers to the location information of the target facilities. In one embodiment, the target positioning information is the latitude and longitude information of the corresponding power facility; in another embodiment, the target positioning information is the coordinate information of the corresponding power facility in a preset coordinate system. It should be noted that each target facility corresponds to only one piece of target positioning information, and in one embodiment, to facilitate identification of the target facility corresponding to the target positioning information, each target facility is assigned a unique number, which is regarded as the target number, and the target positioning information includes the corresponding target number.

[0081] In one embodiment, the power grid positioning information is an array of target positioning information. In another embodiment, the power grid positioning information is a spatial positioning map based on the target positioning information, where each positioning point in the spatial positioning map represents corresponding target positioning information. The goal is to integrate all target positioning information into the power grid positioning information.

[0082] 102. Classify the power grid location information using the trained settlement discrimination model to obtain settlement information of the target facility.

[0083] Specifically, the settlement discrimination model is completed using the power grid positioning training set. In one embodiment, the settlement discrimination model is to judge the settlement of each target facility through the power grid positioning information to obtain the corresponding settlement information. Among them, in one application scenario, the settlement information is a numerical value, and there are two numerical values. The settlement discrimination model outputs the numerical value corresponding to each target facility to achieve classification processing. Each numerical value corresponding to the target facility is the settlement information. In another application scenario, the settlement information is a symbol or a character string, and there are at least two types, or there can be multiple types. After the power grid positioning information is processed using the settlement discrimination model, the symbol or character string corresponding to each target facility is obtained, and each symbol or character string is the settlement information of the corresponding target facility.

[0084] 103. If it is determined that the settlement information matches the preset settlement condition, a settlement status corresponding to the target facility is generated, and the settlement status is used as a monitoring result.

[0085] Depending on the type of settlement information, a corresponding settlement condition is preset. For example, in one embodiment, if the settlement information is a numerical value, the settlement condition can be a numerical range or a specific numerical value. When the settlement condition is a numerical range, a determination is made as to whether the settlement information falls within the numerical range to determine whether the settlement information matches the settlement condition. When the settlement information is a symbol, a determination is made as to whether the settlement condition is identical to the settlement information to determine whether the settlement information matches the settlement condition.

[0086] In one embodiment, the settlement status includes settled and non-settled, and the settlement status of each target facility is the monitoring result.

[0087] By integrating the target location information of all target facilities into grid location information, and then classifying and processing the grid location information using a neural network model, known as a subsidence discrimination model, the system determines whether each target facility has experienced subsidence, thereby generating monitoring results for each target facility. This approach, coupled with the fact that the number of target facilities is known, makes it less likely that a target facility will be missed when integrating the grid location information. Furthermore, the entire monitoring process requires no human intervention. Therefore, compared to manual inspections and monitoring, this approach improves the efficiency and quality of power facility monitoring.

[0088] In another embodiment of the present application, Figure 2 As shown, before the step of classifying the power grid location information using the trained settlement discrimination model, the monitoring method further includes:

[0089] 201. Utilize the autoencoding neural network model in the settlement discrimination model to calculate the power grid positioning training set to obtain settlement training parameters.

[0090] The autoencoder neural network model, also known as an autoencoder neural network (autoencoder), is an unsupervised learning algorithm that uses backpropagation to ensure that the output value of the objective function equals the input value. Specifically, in one embodiment, the autoencoder neural network model includes an encoder and a decoder. During learning, the decoder outputs a target value r close to the encoder input value x. The trained network attempts to replicate the input to the output.

[0091] 202. Calculate the settlement training parameters using the support vector machine classification model in the settlement discrimination model to obtain settlement state parameters.

[0092] In one embodiment, the support vector machine classification model uses a nonlinear support vector machine classifier with a Gaussian radial basis kernel function. The support vector machine classification model, namely SVM (support vector machine), is a binary classification model.

[0093] 203. When the settlement state parameter matches a preset training condition, it is determined that the settlement discrimination model training is completed.

[0094] In one embodiment, the training process includes training an autoencoder neural network model and training a support vector machine classification model. In one application scenario, after the autoencoder neural network model is trained, the sedimentation training parameters are calculated to train the support vector machine classification model. When the sedimentation state parameters match the training conditions, the sedimentation discrimination model training is determined to be complete. In another application scenario, the autoencoder neural network model and the support vector machine classification model are trained simultaneously. When the sedimentation state parameters do not match the training conditions, the autoencoder neural network model is iteratively updated simultaneously.

[0095] In one embodiment, the training conditions include known and correct settlement state parameters. When the settlement state parameters are consistent with the settlement state parameters in the training conditions, it is determined that the settlement state parameters match the training conditions; if they are inconsistent, it is determined that they do not match.

[0096] In addition, it should be noted that the completion of training of the sedimentation discrimination model can also be determined based on the number of training iterations. For example, when the number of iterations is greater than a preset iteration threshold, the training of the sedimentation discrimination model is determined to be complete. The completion of training can also be determined based on the loss function, which will not be further described.

[0097] The settlement discrimination model includes an autoencoder neural network model and a support vector machine classification model. The two models are used to classify and process the grid positioning information, which facilitates efficient and accurate monitoring of the settlement of each target facility and improves the monitoring quality of power facilities.

[0098] In another embodiment of the present application, the step of classifying the power grid location information using the trained settlement discrimination model to obtain the settlement information of the target facility includes:

[0099] 301. Obtain positioning steering vector parameters in the power grid positioning information; each target positioning information corresponds to the positioning steering vector parameters.

[0100] In one embodiment, the power grid positioning information is a multi-source information fusion positioning matrix constructed based on the target positioning information of all target facilities and each target facility. Each target facility, i.e., the corresponding target positioning information, has a positioning steering vector parameter. Because each target facility has a unique target number, the corresponding positioning steering vector parameter can be retrieved from the power grid positioning information based on the target number.

[0101] 302. Use the trained autoencoder neural network model to encode and decode the positioning steering vector parameters to obtain feature steering vector parameters.

[0102] After the trained autoencoder neural network model encodes and decodes the positioning guidance vector parameter, it generates a feature guidance vector parameter. The feature guidance vector parameter is the key feature information of the positioning guidance vector parameter calculated by the autoencoder neural network model.

[0103] 303. Use the trained support vector machine classification model to classify the feature-oriented vector parameters to obtain the settlement information of each target facility.

[0104] After the feature-oriented vector parameters are input into the support vector machine classification model, the corresponding settlement information is calculated. In one embodiment, the settlement information is 0 or 1, which realizes the classification of the feature-oriented vector parameters.

[0105] In the actual use of the settlement discrimination model, an autoencoder neural network model is first used to obtain feature-oriented vector parameters, which reduces the amount of computation, improves efficiency, and reduces computing resource usage. A support vector machine classification model is then used to output settlement information, identifying and classifying each positioning-oriented vector parameter. This helps determine the settlement status of each target facility based on the settlement information. The entire process requires no human intervention, improving the automation level of power facility settlement monitoring, making it less likely to miss a target facility and eliminating operational errors, thereby improving the monitoring quality of power facilities.

[0106] In another embodiment of the present application, the step of encoding and decoding the positioning steering vector parameters using the trained autoencoder neural network model to obtain the feature steering vector parameters includes:

[0107] 401. Encode the positioning guidance vector parameters using the encoder in the autoencoding neural network model to obtain encoding feature parameters.

[0108] 402. Decode the training feature parameters using a decoder in the autoencoding neural network model to obtain imitation vector parameters.

[0109] 403. Extract the encoding feature parameters and use the encoding feature parameters as the feature-guided vector parameters.

[0110] The autoencoder neural network model includes an encoder and a decoder. The encoder encodes the positioning guidance vector parameters and calculates the eigenvalues ​​within them to obtain the most representative encoded feature parameters. The decoder decodes the encoded feature parameters to obtain imitation vector parameters that approximate the positioning guidance vector parameters. During training of the autoencoder neural network model, the goal is to improve the similarity between the imitation vector parameters and the positioning guidance vector parameters. Training is iteratively performed until the number of iterations or training requirements are met. After encoding and decoding, the encoded feature parameters are extracted as the feature guidance vector parameters.

[0111] Through the encoding and decoding process, the encoding feature parameters in the positioning guidance vector parameters are calculated, and the encoding feature parameters are used as the feature guidance vector parameters as the subsequent calculation target, which helps to reduce resource usage and improve the classification efficiency of subsequent classification.

[0112] In another embodiment of the present application, Figure 3 As shown, the steps for obtaining grid location information include:

[0113] 501. Obtain the target positioning information of each target facility.

[0114] In one embodiment, each target facility is equipped with a Beidou positioning device, which includes a Beidou positioning module, a data transmission module, and a microprocessor. The Beidou positioning module receives Beidou positioning signals and transmits them to the microprocessor. The microprocessor processes the Beidou positioning signals, including adding the target facility's target number. The microprocessor then outputs the Beidou positioning signals via the data transmission module, allowing the current executing entity to obtain the target facility's target location information. The Beidou positioning signals provide the target facility's longitude and latitude information.

[0115] 502. Extract facility coding parameters and longitude and latitude coordinate parameters from the target positioning information.

[0116] The facility code parameter is the target number of the target facility.

[0117] 503. Calculate the number of facilities based on all the facility coding parameters.

[0118] Since the facility code parameters for each target facility are unique, the number of facility code parameters is equal to the number of target facilities. The number of facilities can be calculated by summing the facility code parameters.

[0119] 504. Arrange the longitude and latitude coordinate parameters according to the corresponding relationship between the number of facilities and the facility coding parameters to obtain a signal source array.

[0120] In one embodiment, the signal source array is s k (t); where t refers to the time, k = 1, ..., K, and K is equal to the number of facilities.

[0121] 505. Obtain distribution direction parameters of each of the target facilities based on the longitude and latitude coordinate parameters.

[0122] The direction of the target facility relative to the reference point can be calculated using the longitude and latitude in the latitude and longitude coordinate parameters. The reference point can be the latitude and longitude coordinates of the current execution entity or a custom location. After calculating the direction of the latitude and longitude coordinate parameters relative to the reference point, the distribution direction parameter is obtained.

[0123] 506. Calculate the signal source array and the distribution direction parameters using a multi-source information fusion positioning matrix model to obtain the power grid positioning information.

[0124] In one embodiment, the multi-source information fusion positioning matrix model is: Among them, y(t)=[y1(t),…,y M (t)] T , s k (t)=[s1(t),…,s K (t)] T , where M is a multi-source information channel. is an array steering matrix, which represents the overall positioning situation; is the distribution direction parameter. n(t)=[n1(t),…,n M (t)] T , n(t) is an independent and identically distributed Gaussian noise, which is used to represent the positioning noise caused by environmental factors.

[0125] The grid positioning information is automatically calculated through the multi-source information fusion positioning matrix model, which is less likely to miss the target positioning information of the target facility and is highly efficient, helping to improve the settlement monitoring efficiency of power facilities.

[0126] In another embodiment of the present application, before the step of classifying the power grid location information using the trained settlement discrimination model, the method further includes:

[0127] 601. Construct identification space parameters according to the signal source array in the power grid positioning information; the identification space parameters include the positioning steering vector parameters.

[0128] In one embodiment, s k (t) obeys independent Gaussian distribution, studies the second-order information of each parameter in the signal source array, and constructs the recognition space Ar using the positioning and guidance vector parameters corresponding to the n target facilities obtained by a single projection of the spatial spectrum. Specifically, Ar=[a(θ1), a(θ2), ..., a(θ n )]; wherein, there are k target facility positioning steering vector parameters, and (nk) noise steering vector parameters corresponding to the positioning noise angle. In one embodiment, the k target facility positioning steering vector parameters are arranged first, followed by (nk) noise steering vector parameters.

[0129] 602. Generate characteristic vector parameters and use the characteristic vector parameters to project within the recognition space parameters to obtain projection vector parameters.

[0130] In one embodiment, the eigenvector parameter u Si is the eigenvector corresponding to any signal in the signal subspace after eigendecomposition, and u Si =max(u S1 ,u S2 ,...,u Sn ). Ar only contains k positioning guidance vector parameters a(θ1)~a(θ k ), Ar is a set of complete bases in the space. Si Projection in the space where Ar is located to obtain the projection vector parameters It can be obtained from a(θ1)~a(θ k ) is uniquely represented by a linear combination of:

[0131] Among them, B=[β1, β2,...,β k ] T , B is the coefficient matrix, β k is the spatial coefficient parameter.

[0132] 603. Utilize the eigenvector parameters, the projection vector parameters, and the positioning guidance vector parameters to obtain spatial coefficient parameters.

[0133] For the projection vector parameters, we have <Ar·B,a(θi )>= Si , a(θ i )>, meaning u Si Projection vector parameters in Ar space With any space basis a(θ i ) is equal to u Si With the space basis a(θ i ). The spatial coefficient parameter β i =(a(θ i ) H ·a(θ i )) -1 ·a(θ i ) H u Sj . After simplification, it is Thus, the spatial coefficient parameter β is calculated by the characteristic vector parameter, the projection vector parameter and the positioning guidance vector parameter. i .

[0134] 604. Determine whether the spatial coefficient parameter of each positioning and steering vector parameter matches a preset judgment value.

[0135] If so, the corresponding positioning steering vector parameter is determined to be noise and is excluded; if not, the corresponding positioning steering vector parameter is retained to denoise and update the power grid positioning information.

[0136] In one embodiment, because the spatial complete basis represents the vector in the space uniquely, the coefficient corresponding to the vector outside the complete basis is 0, so the discriminant value is 0. When the spatial coefficient parameter is equal to 0, it is judged to match the discriminant value.

[0137] By removing positioning noise, the quality of power grid positioning information is improved, thereby improving the accuracy of power facility settlement monitoring results.

[0138] In another embodiment of the present application, the step of obtaining the spatial coefficient parameter by processing the eigenvector parameter, the projection vector parameter, and the positioning guidance vector parameter includes:

[0139] When the inner product of the projection vector parameter and any of the positioning steering vector parameters is equal to the inner product of the eigenvector parameter and the corresponding positioning steering vector parameter, the spatial coefficient parameter corresponding to the positioning steering vector parameter is calculated using the corresponding positioning steering vector parameter and the eigenvector parameter.

[0140] According to the characteristics of the space complete basis, the space coefficient parameters are calculated, which improves the accuracy of the space coefficient parameters.

[0141] ​The multi-source information fusion positioning matrix model is used to fuse the target location information of each target facility into grid location information. The settlement discrimination model is then used to automatically calculate the settlement status of each target facility to obtain monitoring results. This process removes positioning noise from the grid location information, eliminating the need for manual intervention. This reduces the risk of missing target facilities and improves the quality of settlement monitoring for power facilities.

[0142] This application also provides a power facility monitoring system, such as Figure 4 As shown, it includes an acquisition module 1, which is used to acquire power grid positioning information; the power grid positioning information includes target positioning information of at least one target facility;

[0143] Settlement module 2, configured to classify the power grid location information using a trained settlement discrimination model to obtain settlement information of the target facility; the settlement discrimination model is trained using a power grid location training set;

[0144] The judgment module 3 is used to generate a settlement state corresponding to the target facility if it is determined that the settlement information matches the preset settlement condition, and use the settlement state as a monitoring result.

[0145] Preferably, the system further comprises a training module for calculating the power grid location training set using the autoencoder neural network model in the subsidence discrimination model to obtain subsidence training parameters before classifying the power grid location information using the trained subsidence discrimination model;

[0146] Calculating the settlement training parameters using the support vector machine classification model in the settlement discrimination model to obtain settlement state parameters;

[0147] When the settlement state parameters match the preset training conditions, it is determined that the settlement discrimination model training is completed.

[0148] Preferably, the settlement module 2 includes an extraction unit for obtaining positioning steering vector parameters in the power grid positioning information; each target positioning information corresponds to the positioning steering vector parameters;

[0149] an encoding and decoding unit, configured to encode and decode the positioning steering vector parameters using the trained autoencoding neural network model to obtain feature steering vector parameters;

[0150] A discrimination unit is used to classify the feature-oriented vector parameters using the trained support vector machine classification model to obtain the settlement information of each target facility.

[0151] Preferably, the encoding and decoding unit includes an encoding subunit, which is used to encode the positioning steering vector parameters using an encoder in the autoencoding neural network model to obtain encoding feature parameters;

[0152] A decoding subunit, configured to decode the training feature parameters using a decoder in the autoencoding neural network model to obtain imitation vector parameters;

[0153] The extraction subunit is configured to extract the encoding feature parameters and use the encoding feature parameters as the feature-guided vector parameters.

[0154] Preferably, the acquisition module 1 includes a target positioning unit, configured to acquire the target positioning information of each target facility;

[0155] A parameter unit, configured to extract facility coding parameters and latitude and longitude coordinate parameters from the target positioning information;

[0156] a summing unit, configured to calculate the number of facilities according to all the facility coding parameters;

[0157] An array unit, configured to arrange the longitude and latitude coordinate parameters according to a correspondence between the number of facilities and the facility coding parameters to obtain a signal source array;

[0158] a direction unit, configured to obtain distribution direction parameters of each of the target facilities based on the latitude and longitude coordinate parameters;

[0159] A calculation unit is used to calculate the signal source array and the distribution direction parameters using a multi-source information fusion positioning matrix model to obtain the power grid positioning information.

[0160] Preferably, the system further comprises a denoising module, which constructs identification space parameters according to the signal source array in the grid positioning information before classifying the grid positioning information using the trained settlement discrimination model; the identification space parameters include the positioning steering vector parameters;

[0161] Generate characteristic vector parameters and use the characteristic vector parameters to project within the identification space parameters to obtain projection vector parameters;

[0162] The spatial coefficient parameters are obtained by processing the eigenvector parameters, the projection vector parameters and the positioning guidance vector parameters;

[0163] Determining whether the spatial coefficient parameter of each of the positioning and steering vector parameters matches a preset discrimination value;

[0164] If so, the corresponding positioning steering vector parameter is determined to be noise and is excluded; if not, the corresponding positioning steering vector parameter is retained to denoise and update the power grid positioning information.

[0165] Preferably, the denoising module includes a processing unit for calculating the spatial coefficient parameter corresponding to the positioning steering vector parameter using the corresponding positioning steering vector parameter and the eigenvector parameter when the inner product of the projection vector parameter and any positioning steering vector parameter is equal to the inner product of the eigenvector parameter and the corresponding positioning steering vector parameter.

[0166] After acquiring the grid location information from acquisition module 1, settlement module 2 processes it using a settlement discrimination model to determine the settlement status of each target facility. Determination module 3 then compares this settlement information with settlement conditions and outputs the settlement status of each target facility. This process reduces the risk of missing a target facility, improving the quality of settlement monitoring for power facilities.

[0167] It should be noted that the description of the above embodiment of the power facility monitoring system is similar to the description of the above method and has the same beneficial effects as the method embodiment. For technical details not disclosed in the embodiment of the power facility monitoring system of the present invention, those skilled in the art should refer to the description of the method embodiment of the present invention for understanding.

[0168] It should be noted that, in the embodiment of the present invention, if the above method is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program codes, such as a U disk, a mobile hard disk, a read-only memory (ROM, Read Only Memory), a magnetic disk or an optical disk. In this way, the embodiment of the present invention is not limited to any specific combination of hardware and software.

[0169] Correspondingly, an embodiment of the present application further discloses a storage medium storing a computer program that can be loaded by a processor and execute the above method.

[0170] The present application also discloses a power facility monitoring device, such as Figure 5As shown, the system includes a processor 100, at least one communication bus 200, a user interface 300, at least one external communication interface 400, and a memory 500. The communication bus 200 is configured to enable communication between these components. The user interface 300 may include a display screen, and the external communication interface 400 may include a standard wired interface and a wireless interface. The memory 500 stores a method for monitoring electric power facilities. The processor 100 is configured to employ the method when executing the method stored in the memory 500.

[0171] The above description of the embodiments of the power facility monitoring device and storage medium is similar to the description of the above-mentioned method embodiment and has similar beneficial effects as the method embodiment. For technical details not disclosed in the embodiments of the power facility monitoring device and storage medium of the present invention, please refer to the description of the method embodiment of the present invention for understanding.

[0172] It should be understood that "one embodiment" or "an embodiment" mentioned throughout the specification means that the specific features, structures or characteristics related to the embodiment are included in at least one embodiment of the present invention. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner. It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention. The serial numbers of the above-mentioned embodiments of the present invention are for description only and do not represent the advantages and disadvantages of the embodiments.

[0173] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or apparatus comprising the element.

[0174] In the several embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0175] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed across multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0176] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0177] Those skilled in the art will understand that all or part of the steps of implementing the above-mentioned method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above-mentioned method embodiment; and the aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROMs, magnetic disks or optical disks.

[0178] Alternatively, if the integrated units of the present invention are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of the present invention, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes a number of instructions for enabling a device to perform all or part of the methods described in various embodiments of the present invention. The aforementioned storage media include various media capable of storing program code, such as mobile storage devices, ROMs, magnetic disks, or optical disks.

[0179] The above disclosure is merely a preferred embodiment of the present invention and certainly cannot be used to limit the scope of the present invention. Therefore, equivalent changes made according to the claims of the present invention are still within the scope of the present invention.

Claims

1. A method for monitoring electric power facilities, characterized in that: include: Acquiring power grid positioning information; the power grid positioning information includes target positioning information of at least one target facility; Using the trained settlement discrimination model to classify the power grid location information, and obtain settlement information of the target facility; The settlement discrimination model is trained using a power grid positioning training set; If it is determined that the settlement information matches the preset settlement condition, a settlement status corresponding to the target facility is generated and the settlement status is used as a monitoring result; The step of classifying the power grid location information using the trained settlement discrimination model to obtain the settlement information of the target facility includes: Obtaining a positioning steering vector parameter in the power grid positioning information; each target positioning information corresponds to the positioning steering vector parameter; Using the trained autoencoding neural network model in the settlement discrimination model to encode and decode the positioning guidance vector parameters to obtain feature guidance vector parameters; The support vector machine classification model in the trained settlement discrimination model is used to classify the feature-oriented vector parameters to obtain the settlement information of each target facility.

2. The electric power facility monitoring method according to claim 1, wherein: Before classifying the power grid location information using the trained settlement discrimination model, the monitoring method further includes: Utilizing the autoencoding neural network model in the settlement discrimination model to calculate the power grid positioning training set, to obtain settlement training parameters; Calculating the settlement training parameters using the support vector machine classification model in the settlement discrimination model to obtain settlement state parameters; When the settlement state parameters match the preset training conditions, it is determined that the settlement discrimination model training is completed.

3. The electric power facility monitoring method according to claim 1, wherein: The step of encoding and decoding the positioning steering vector parameters using the trained autoencoder neural network model to obtain feature steering vector parameters includes: Encoding the positioning and steering vector parameters using an encoder in the autoencoding neural network model to obtain encoding feature parameters; The encoding feature parameters are extracted and used as the feature-guided vector parameters.

4. The electric power facility monitoring method according to claim 1, wherein: The step of obtaining grid location information includes: Acquiring the target positioning information of each target facility; Extracting facility coding parameters and longitude and latitude coordinate parameters from the target positioning information; Calculate the number of facilities based on all the facility coding parameters; Arranging the longitude and latitude coordinate parameters according to the corresponding relationship between the number of facilities and the facility coding parameters to obtain a signal source array; Obtaining distribution direction parameters of each of the target facilities based on the latitude and longitude coordinate parameters; The signal source array and the distribution direction parameters are calculated using a multi-source information fusion positioning matrix model to obtain the power grid positioning information.

5. The electric power facility monitoring method according to claim 4, wherein: Before classifying the power grid location information using the trained settlement discrimination model, the method further includes: Constructing identification space parameters according to the signal source array in the power grid positioning information; the identification space parameters include the positioning steering vector parameters; Generate characteristic vector parameters and use the characteristic vector parameters to project within the identification space parameters to obtain projection vector parameters; The spatial coefficient parameters are obtained by processing the eigenvector parameters, the projection vector parameters and the positioning guidance vector parameters; Determining whether the spatial coefficient parameter of each of the positioning and steering vector parameters matches a preset discrimination value; If so, the corresponding positioning steering vector parameter is determined to be noise and is excluded; if not, the corresponding positioning steering vector parameter is retained to denoise and update the power grid positioning information.

6. The electric power facility monitoring method according to claim 5, wherein: The step of obtaining the spatial coefficient parameters by processing the eigenvector parameters, the projection vector parameters and the positioning guidance vector parameters comprises: When the inner product of the projection vector parameter and any of the positioning steering vector parameters is equal to the inner product of the eigenvector parameter and the corresponding positioning steering vector parameter, the spatial coefficient parameter corresponding to the positioning steering vector parameter is calculated using the corresponding positioning steering vector parameter and the eigenvector parameter.

7. A power facility monitoring system, characterized in that: It includes an acquisition module for acquiring power grid positioning information; the power grid positioning information includes target positioning information of at least one target facility; A settlement module is used to classify the power grid location information using the trained settlement discrimination model to obtain the settlement information of the target facility; The settlement discrimination model is trained using a power grid positioning training set; a judgment module, configured to generate a settlement status corresponding to the target facility if it is determined that the settlement information matches a preset settlement condition, and use the settlement status as a monitoring result; The settlement module is specifically used to: obtain the positioning steering vector parameters in the power grid positioning information; each of the target positioning information corresponds to the positioning steering vector parameters; use the autoencoding neural network model in the trained settlement discrimination model to encode and decode the positioning steering vector parameters to obtain feature steering vector parameters; use the support vector machine classification model in the trained settlement discrimination model to classify the feature steering vector parameters to obtain the settlement information of each target facility.

8. A power facility monitoring device, comprising a memory and a processor, characterized in that: The memory stores a power facility monitoring method, and the processor adopts the method according to any one of claims 1 to 6 when executing the power facility monitoring method.

9. A storage medium, characterized in that: The device stores a computer program that can be loaded by a processor and execute the method according to any one of claims 1 to 6.

Images