Remote monitoring method of power equipment based on 5G short slice

Abstract

This application discloses a method for remote monitoring of power equipment based on 5G short slices. It configures 5G network short slices for power equipment based on network slicing technology to remotely monitor the operating status of power equipment. At the same time, it uses data processing technology based on deep learning to perform time series analysis on the collected current, voltage and temperature data, and comprehensively characterizes the working status of power equipment based on the time series joint information of current data and voltage data. Furthermore, by performing fine-grained interactive response analysis on the working status characteristics and equipment temperature characteristics of the power equipment, the abnormal correlation response pattern between the two is identified, thereby realizing real-time monitoring of the working status of the power equipment and abnormality identification. In this way, the accuracy and real-time performance of power equipment monitoring can be effectively improved, which helps to timely discover potential faults and abnormal conditions and reduce the risk of power supply interruption due to equipment failure.

Description

Technical Field

[0001] The present application relates to the field of intelligent monitoring, and more specifically, to a remote monitoring method for power equipment based on 5G short slicing. Background Art

[0002] With the rapid development of society and the continuous growth of electricity demand, the scale of power systems is expanding, and the number of power equipment is also increasing. To ensure the stable operation of power systems, timely and effective status monitoring and fault diagnosis of power equipment are particularly important. Traditional power equipment monitoring methods rely on manual inspections and regular maintenance. This method is not only inefficient but also difficult to achieve real-time monitoring of equipment status, failing to meet the safety and reliability requirements of modern power systems.

[0003] With the development of smart grids, remote monitoring of power equipment is becoming increasingly important. However, existing power equipment monitoring systems typically rely on wired communication networks or low-bandwidth wireless communication technologies. These technologies have numerous limitations in transmission rate, latency, and reliability, making them difficult to meet the real-time and high-reliability requirements of modern power systems. Furthermore, traditional monitoring systems often only provide simple data collection and parameter threshold-based alarm functions, lacking in-depth data analysis and intelligent decision-making capabilities. These systems struggle to adapt to the complex and ever-changing operating environments of power equipment and are prone to missed and false alarms, limiting monitoring accuracy.

[0004] In recent years, 5G technology, with its ultra-high data transmission rates, ultra-low latency, and large number of connections, has been widely adopted across various industries. In the power sector, the application of 5G technology offers new possibilities for remote monitoring of power equipment. It not only provides high-speed data transmission capabilities but also, through network slicing, offers customized network services for different application scenarios, meeting the needs of diverse businesses. Therefore, a method and system for remote monitoring of power equipment based on 5G short slicing is highly anticipated. Summary of the Invention

[0005] In order to solve the above technical problems, the present application is proposed. The embodiment of the present application provides a remote monitoring method for power equipment based on 5G short slices, which configures 5G network short slices for power equipment based on network slicing technology to remotely monitor the operating status of the power equipment, and at the same time uses data processing technology based on deep learning to perform time series analysis on the collected current, voltage and temperature data, and comprehensively characterizes the working status of the power equipment based on the time series joint information of the current data and voltage data. Furthermore, by performing fine-grained interactive response analysis on the working status characteristics and equipment temperature characteristics of the power equipment, the abnormal correlation response pattern between the two can be identified, thereby realizing real-time monitoring and abnormal identification of the working status of the power equipment. In this way, the accuracy and real-time performance of power equipment monitoring can be effectively improved, which helps to timely discover potential faults and abnormal conditions and reduce the risk of power supply interruption due to equipment failure.

[0006] According to one aspect of the present application, a method for remote monitoring of power equipment based on 5G short slicing is provided, which includes:

[0007] Connecting the monitored power equipment to the 5G network and configuring a 5G network short slice for the monitored power equipment;

[0008] A time queue of state parameters of the monitored power equipment is collected by a sensor group, wherein the state parameters include current value, voltage value and temperature value;

[0009] Transmitting the time queue of the status parameters of the monitored power equipment to a status remote monitoring center via the 5G network short slice;

[0010] At the state remote monitoring center, extracting the temperature characteristics and working state characteristics of the power equipment from the time queue of the state parameters of the monitored power equipment to obtain a temperature time series associated implicit feature vector and a power equipment working state time series representation vector;

[0011] At the state remote monitoring center, a fine-grained interactive response analysis is performed on the power equipment working state time series representation vector and the temperature time series associated implicit feature vector to obtain a power equipment working state-temperature presentation interactive coupling feature vector;

[0012] In the state remote monitoring center, it is determined whether the working state of the monitored power equipment is abnormal based on the interactive coupling characteristic vector presented by the working state-temperature of the power equipment.

[0013] Compared with the existing technology, the present application provides a method for remote monitoring of power equipment based on 5G short slices. It configures 5G network short slices for power equipment based on network slicing technology to remotely monitor the operating status of power equipment. At the same time, it uses data processing technology based on deep learning to perform time series analysis on the collected current, voltage and temperature data, and comprehensively characterizes the working status of power equipment based on the time series joint information of current data and voltage data. Furthermore, by performing fine-grained interactive response analysis on the working status characteristics and equipment temperature characteristics of the power equipment, the abnormal correlation response pattern between the two can be identified, thereby realizing real-time monitoring of the working status of the power equipment and abnormality identification. In this way, the accuracy and real-time performance of power equipment monitoring can be effectively improved, which helps to timely discover potential faults and abnormal conditions and reduce the risk of power supply interruption due to equipment failure. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The above and other purposes, features, and advantages of the present application will become more apparent through a more detailed description of the embodiments of the present application in conjunction with the accompanying drawings. The accompanying drawings are intended to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the drawings, the same reference numerals generally represent the same components or steps.

[0015] Figure 1 Flowchart of a method for remote monitoring of power equipment based on 5G short slicing according to an embodiment of the present application;

[0016] Figure 2 Schematic diagram of data flow of a remote monitoring method for power equipment based on 5G short slicing according to an embodiment of the present application;

[0017] Figure 3 This is a flowchart of sub-step S5 of the remote monitoring method for power equipment based on 5G short slicing according to an embodiment of the present application. DETAILED DESCRIPTION

[0018] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described herein.

[0019] As used in this application and the claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" are not intended to refer to the singular but may include the plural. Generally speaking, the terms "comprises" and "include" only indicate the inclusion of the steps and elements specifically identified, and these steps and elements do not constitute an exclusive list. A method or apparatus may also include other steps or elements.

[0020] Although the present application makes various references to certain modules in the system according to embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are illustrative only, and different aspects of the system and method can use different modules.

[0021] Flowcharts are used in this application to illustrate the operations performed by the systems according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the various steps may be processed in reverse order or simultaneously, as needed. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0022] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described herein.

[0023] In recent years, 5G technology, with its ultra-high data transmission rates, ultra-low latency, and large number of connections, has been widely adopted across various industries. In the power sector, the application of 5G technology offers new possibilities for remote monitoring of power equipment. It not only provides high-speed data transmission capabilities but also, through network slicing, offers customized network services for different application scenarios, meeting the needs of diverse businesses. Therefore, a method and system for remote monitoring of power equipment based on 5G short slicing is highly anticipated.

[0024] In the technical solution of this application, a remote monitoring method for power equipment based on 5G short slicing is proposed. Figure 1 This is a flowchart of a method for remote monitoring of power equipment based on 5G short slicing according to an embodiment of the present application. Figure 2 Schematic diagram of data flow of a remote monitoring method for power equipment based on 5G short slices according to an embodiment of the present application. Figure 1 and Figure 2As shown, according to an embodiment of the present application, a remote monitoring method for power equipment based on 5G short slices includes the following steps: S1, connecting the monitored power equipment to a 5G network, and configuring a 5G network short slice for the monitored power equipment; S2, collecting a time queue of state parameters of the monitored power equipment through a sensor group, wherein the state parameters include current value, voltage value and temperature value; S3, transmitting the time queue of the state parameters of the monitored power equipment to a state remote monitoring center through the 5G network short slice; S4, in the state remote monitoring center, extracting the temperature characteristics and the working state characteristics of the power equipment from the time queue of the state parameters of the monitored power equipment to obtain a temperature time series associated implicit feature vector and a power equipment working state time series representation vector; S5, in the state remote monitoring center, performing fine-grained interactive response analysis on the power equipment working state time series representation vector and the temperature time series associated implicit feature vector to obtain a power equipment working state-temperature presentation interactive coupling feature vector; S6, in the state remote monitoring center, determining whether there is an abnormality in the working state of the monitored power equipment based on the power equipment working state-temperature presentation interactive coupling feature vector.

[0025] In particular, S1 connects the monitored power equipment to the 5G network and configures a short 5G network slice for the monitored power equipment. Those skilled in the art will appreciate that network slicing refers to the division of multiple, isolated virtual networks on the same physical infrastructure network. Network slices are completely isolated from each other, and errors and failures in a particular slice do not affect other slices. Each virtual network is built based on different service requirements to flexibly address different network application scenarios. For example, in the field of autonomous driving, low-latency and high-reliability connections may be more emphasized, while in the application of large-scale IoT devices, the number of connections and energy efficiency may be more important.

[0026] Specifically, network slicing uses network virtualization technology to abstract various physical resources in the network into virtual resources. Based on specified network functions and specific access network technologies, the overall structure consists of infrastructure and slice instances running on top of the infrastructure, namely the service layer, providing an end-to-end virtual network. The service layer describes the system architecture at a logical level and consists of network functions and the connections between functions. These network functions are usually defined in the form of software packages, which provide templates that define deployment and operational requirements (connectivity, interfaces, KPI requirements, etc.). The infrastructure layer describes the network elements and resources required to maintain the operation of a network slice at a physical level, including computing resources (such as IT servers in data centers) and network resources (such as aggregation switches, edge routers, cables, etc.). After the infrastructure resources are virtualized, they are allocated to each slice according to the slice requirements, forming different virtual networks and providing specific services. In the technical solution of this application, by configuring 5G network short slices for power equipment, customized network services can be provided for the remote monitoring needs of power equipment. Each network slice can be optimized according to the specific monitoring task requirements (such as bandwidth, latency, security, etc.), ensuring the transmission quality and efficiency of monitoring data while reducing operating costs.

[0027] In particular, S2 collects a time queue of the status parameters of the monitored power equipment through a sensor group, and the status parameters include current value, voltage value, and temperature value. It should be understood that current, voltage, and temperature are key parameters in the process of monitoring the operating status of power equipment, and effectively reflect the load condition, performance, and health status of the power equipment. In the technical solution of the present application, real-time monitoring and trend analysis of the current, voltage, and temperature of the power equipment helps to identify potential problems and signs of failure of the equipment, so that preventive measures can be taken in advance to avoid the occurrence of equipment failure.

[0028] In particular, S3 transmits the time queue of the state parameters of the monitored power equipment to the remote status monitoring center via the 5G network short slice. That is, in the technical solution of this application, the time queue of the state parameters of the monitored power equipment is transmitted to the remote status monitoring center via the 5G network short slice to achieve efficient, reliable, and secure data transmission, and utilizes the powerful computing resources and data analysis capabilities of the remote status monitoring center to perform real-time and efficient data processing and analysis, thereby achieving real-time remote monitoring of the operating status of the power equipment.

[0029] In particular, the S4 extracts the temperature characteristics and working state characteristics of the power equipment from the time queue of the state parameters of the monitored power equipment at the state remote monitoring center to obtain the temperature time series associated implicit feature vector and the power equipment working state time series representation vector. In the technical solution of the present application, first, in order to perform separate time series feature extraction for each parameter to improve the accuracy of data analysis, the present application further queues the time queue of the state parameters of the monitored power equipment according to the parameter sample dimension to obtain the time queue of the current value, the time queue of the voltage value and the time queue of the temperature value, thereby making the data structure of the power equipment state parameters clearer and more orderly, so as to facilitate a more detailed analysis of the trend and pattern of each parameter changing over time, and provide richer data dimensions for subsequent deep learning analysis; then, the time queue of the current value, the time queue of the voltage value and the time queue of the temperature value are input into the time queue encoder based on the bidirectional LSTM model to obtain the current time series associated implicit feature vector, the voltage time series associated implicit feature vector and the temperature time series associated implicit feature vector. Since current, voltage and temperature are dynamic parameters that change over time, in order to fully learn the time series variation characteristics of each parameter so as to accurately understand the operating status of the power equipment, this application uses a time queue encoder based on a bidirectional LSTM model to process the time queue of the current value, the time queue of the voltage value and the time queue of the temperature value respectively. Those skilled in the art should know that the bidirectional LSTM model can effectively capture the long-term dependencies in the time series data through memory units and gating mechanisms, while considering the forward and backward time context information in the time series data, thereby achieving a comprehensive and in-depth understanding of the time series context information of the current, voltage and temperature data to generate the current time series associated implicit feature vector, the voltage time series associated implicit feature vector and the temperature time series associated implicit feature vector. Subsequently, the current time series associated implicit feature vector and the voltage time series associated implicit feature vector are input into the power equipment working state joint characterization network to obtain the power equipment working state time series representation vector. Considering that current and voltage are the two most critical parameters of power equipment, the two jointly reflect the electrical characteristics and working state of the power equipment. Therefore, in order to achieve a comprehensive characterization of the working state of the power equipment, the present application further uses a joint characterization network for the working state of the power equipment to jointly encode the current time series associated implicit feature vector and the voltage time series associated implicit feature vector, so as to integrate multi-source parameter information, improve the richness of feature expression, and generate a comprehensive feature representation of the working state of the power equipment. In a specific example of the present application, the joint characterization network for the working state of the power equipment achieves feature fusion by calculating the Hadamard product between the current time series associated implicit feature vector and the voltage time series associated implicit feature vector to obtain the time series representation vector of the working state of the power equipment.

[0030] In particular, the S5, in the state remote monitoring center, performs a fine-grained interactive response analysis on the time series representation vector of the working state of the power equipment and the implicit feature vector associated with the temperature time series to obtain the interactive coupling feature vector of the working state of the power equipment-temperature. It should be understood that when the power equipment is working, heat will be generated inside it, causing the temperature to rise. If the temperature continues to rise and exceeds the normal operating temperature range of the equipment, the performance of the equipment may decline, thereby affecting the electrical characteristics and working state of the equipment. In other words, there is a certain interactive effect between the temperature of the equipment and its working state, and this interactive pattern effectively reveals the health status and potential failure risks of the equipment. Based on this, the present application further performs an interactive response analysis on the time series representation vector of the working state of the power equipment and the implicit feature vector associated with the temperature time series to reveal the interactive response pattern between the temperature of the equipment and its working state, thereby achieving an in-depth understanding of the health status of the power equipment. In a specific example of the present application, such as Figure 3 As shown, the S5 includes: S51, performing principal component analysis on the power equipment working state time series representation vector and the temperature time series associated implicit feature vector respectively to obtain a set of power equipment working state time series principal component feature components and a set of temperature time series principal component feature components; S52, performing optimal pairing multi-scale interactive coupling on the set of power equipment working state time series principal component feature components and the set of temperature time series principal component feature components to obtain the power equipment working state-temperature presentation interactive coupling feature vector.

[0031] Specifically, in S51, principal component analysis is performed on the power equipment operating state time series representation vector and the temperature time series associated implicit feature vector to obtain a set of power equipment operating state time series principal component feature components and a set of temperature time series principal component feature components. That is, principal component analysis (PCA) technology is used to process the power equipment operating state time series representation vector and the temperature time series associated implicit feature vector to reduce the complexity of the data and extract the main features therein, so that the subsequent interactive response analysis process focuses on the main feature parts, thereby obtaining a set of power equipment operating state time series principal component feature components and a set of temperature time series principal component feature components. In a specific example, the specific steps of performing principal component analysis on the power equipment working state timing representation vector and the temperature timing associated implicit feature vector respectively include: calculating the covariance matrix of the power equipment working state timing representation vector to obtain the power equipment working state timing feature covariance matrix, and performing eigenvalue decomposition on the power equipment working state timing feature covariance matrix to obtain a set of power equipment working state timing principal component eigenvalues ​​and a corresponding set of power equipment working state timing principal component feature components.

[0032] Specifically, in S52, the set of the main component characteristic components of the power equipment working state time series and the set of the main component characteristic components of the temperature time series are optimally paired and multi-scale interactively coupled to obtain the power equipment working state-temperature interactively coupled characteristic vector. In the technical solution of the present application, first, each power equipment working state time series main component characteristic component in the set of the main component characteristic components of the power equipment working state time series is used as a query vector, and the set of the main component characteristic components of the temperature time series is used as a query library, and the main component characteristic components of the temperature time series that best match each query vector are matched from the query library to obtain a set of optimal matching pairs of {power equipment working state time series main component characteristic component; temperature time series main component characteristic component}; here, by adopting the optimal matching algorithm to perform feature query matching on the sets of the two main component characteristic components, the Mahalanobis distance is used as a metric to query the most relevant feature pairs between the two, and a one-to-one correspondence between the main component characteristic components of the power equipment working state time series and the main component characteristic components of the temperature time series is established, thereby achieving accurate matching of the interactive response pattern with strong correlation between the working state of the power equipment and the temperature of the equipment. Next, the set of the best matching pairs of {power equipment working state time series principal component feature components; temperature time series principal component feature components} is subjected to multi-scale interactive fusion to obtain the power equipment working state-temperature interactive coupling feature vector. Specifically, first, each best matching pair of {power equipment working state time series principal component feature components; temperature time series principal component feature components} in the set of the best matching pairs of {power equipment working state time series principal component feature components; temperature time series principal component feature components} is input into the multi-scale interactive response coupling module to obtain a set of multi-scale interactive coupling representation vectors of the best matching pairs of power equipment working state-temperature; that is, each best matching pair is input into the multi-scale interactive response coupling module for interactive response analysis of the best pairing features. The multi-scale interactive response coupling module performs multi-level feature interaction and deep learning on the power equipment working state time series principal component feature components and temperature time series principal component feature components in the best matching pairs, which can effectively capture the interaction pattern between the power equipment working state and the equipment temperature at different abstract levels, and automatically learn and generate the interactive response feature representation of each best matching pair. Furthermore, the set of multi-scale interaction-coupling representation vectors for the optimal power device operating state-temperature pair is cascaded to obtain the power device operating state-temperature interaction-coupling feature vector. Here, by performing a cascade operation on the interaction-response feature representations of all optimal matching pairs to synthesize global information, the power device operating state-temperature interaction-coupling feature vector is obtained. This vector comprehensively reflects the interaction-response pattern between the device operating state and the device temperature, thereby providing a more comprehensive and in-depth analysis basis for identifying abnormalities in the power device operating state.

[0033] Among them, the specific process of matching the temperature time series principal component characteristic component that best matches each of the query vectors from the query library includes: calculating the Mahalanobis distance between the query vector and each temperature time series principal component characteristic component in the query library to obtain a set of power equipment working state-temperature principal component matching difference coefficients; selecting the temperature time series principal component characteristic component corresponding to the minimum value in the set of power equipment working state-temperature principal component matching difference coefficients as the optimal matching result of the query vector to obtain the {power equipment working state time series principal component characteristic component; temperature time series principal component characteristic component} best matching pair.

[0034] More specifically, the specific process of inputting each {power equipment working state time series principal component characteristic component; temperature time series principal component characteristic component} best matching pair in the set of {power equipment working state time series principal component characteristic component; temperature time series principal component characteristic component} best matching pairs into the multi-scale interactive response coupling module includes: respectively calculating the position point addition, position point subtraction and position point multiplication between the power equipment working state time series principal component characteristic component and the temperature time series principal component characteristic component in the said {power equipment working state time series principal component characteristic component; temperature time series principal component characteristic component} best matching pair to obtain the first best matching interaction representation vector, the second best matching interaction representation vector and the third best matching interaction representation vector ; The first best matching interaction representation vector, the second best matching interaction representation vector and the third best matching interaction representation vector are cascaded and fused, and then input into a one-dimensional convolutional layer containing a maximum pooling layer to obtain a multi-scale interaction representation vector of the best matching of the working state of the power equipment and the temperature principal component; the maximum value of the second norm of the characteristic component of the principal component of the time series of the working state of the power equipment, the second norm of the characteristic component of the principal component of the temperature time series and the preset scale adjustment parameter is selected as a scale scaling factor, and each eigenvalue in the multi-scale interaction representation vector of the best matching of the working state of the power equipment and the temperature principal component is divided by the scale scaling factor to obtain the multi-scale interaction coupling representation vector of the best matching pair of the working state of the power equipment and the temperature.

[0035] In summary, a fine-grained interactive response analysis is performed on the power equipment working state time series representation vector and the temperature time series associated implicit feature vector, including: processing the power equipment working state time series representation vector and the temperature time series associated implicit feature vector using the following principal component matching interactive fusion formula to obtain the power equipment working state-temperature interactive coupling feature vector, wherein the principal component matching interactive fusion formula is:

[0036] C1=U1Λ1U1 T

[0037] C2=U2Λ2U2T

[0038] U1=[v 11 ,v 12 ,…,v 1m ]

[0039]

[0040] U2=[v 21 ,v 22 ,…,v 2m ]

[0041]

[0042] v f =[v p1 ;v p2 ;...v pm ]

[0043] Wherein, C1 and C2 represent the covariance matrix of the power equipment working state time series representation vector and the temperature time series associated implicit eigenvector, respectively; Λ1 is a diagonal matrix consisting of a set of the main component eigenvalues ​​of the power equipment working state time series obtained by eigenvalue decomposition of the power equipment working state time series feature covariance matrix; λ 11 and λ 1m They represent the first and mth main component eigenvalues ​​of the working state time series of the power equipment in the set of main component eigenvalues ​​of the working state time series of the power equipment, respectively, where m is the number of main component eigenvalues ​​of the working state time series of the power equipment, U1 is the matrix composed of the set of main component eigenvalues ​​of the working state time series of the power equipment, and v 11 、v 12 、v 1i and v 1m Respectively represent the first, second, i-th and m-th main component characteristic components of the time series of the working state of the power equipment in the set of main component characteristic components of the time series of the working state of the power equipment, (·) T represents the transpose of the matrix, Λ2 is the diagonal matrix composed of the set of principal component eigenvalues ​​of the temperature time series obtained by eigenvalue decomposition of the temperature time series feature covariance matrix, and λ 21 and λ 2m Respectively represent the first and mth temperature time series principal component eigenvalues ​​in the set of the temperature time series principal component eigenvalues, U2 is the matrix composed of the set of temperature time series principal component characteristic components, v 21 、v 22 、v 2j 、v 2k and v 2mRespectively represent the first, second, jth, kth and mth temperature time series principal component characteristic components in the set of the temperature time series principal component characteristic components, S -1 The inverse matrix of the covariance matrix between the principal component characteristic component of the i-th power equipment working state time series and the principal component characteristic component of the j-th temperature time series, represents the index corresponding to the minimum value, k represents the index of the temperature time series principal component feature component with the smallest semantic difference from the time series principal component feature component of the i-th power equipment working state, ⊙ represents the dot product, Indicates point addition, represents point subtraction, conv1D represents one-dimensional convolution operation, MaxPool represents maximum pooling operation, ‖·‖2 represents the bi-norm of feature vector, ε is the preset scale adjustment parameter, max(·) is the maximum value function, v p1 、v p2 、v pi and v pm denote the multi-scale interaction coupling representation vectors of the first, second, i-th, and m-th power equipment working state-temperature optimal matching pairs, [·,·,·] denotes the cascade operation, and v f It indicates that the working state and temperature of the power equipment present an interactive coupling characteristic vector.

[0044] In particular, the S6, at the state remote monitoring center, determines whether there is an abnormality in the working state of the monitored power equipment based on the interactive coupling feature vector of the working state of the power equipment and the temperature. In a specific example of the present application, the interactive coupling feature vector of the working state of the power equipment and the temperature is input into a state monitoring module based on a classifier to obtain a state monitoring result, and the state monitoring result is used to indicate whether there is an abnormality in the working state of the monitored power equipment. Specifically, the classification process of the state monitoring module based on the classifier includes: using multiple fully connected layers of the classifier to fully connect encode the interactive coupling feature vector of the working state of the power equipment and the temperature to obtain an encoded feature vector; and passing the encoded feature vector through the softmax classification function of the classifier to obtain the state monitoring result.

[0045] Preferably, inputting the power equipment working state-temperature interaction coupling feature vector into a classifier-based state monitoring module to obtain a state monitoring result includes:

[0046] Determine a probability value of the power equipment working state-temperature mutual coupling obtained by inputting the power equipment working state-temperature mutual coupling feature vector into a state monitoring module based on a classifier, wherein the power equipment working state-temperature mutual coupling probability value indicates a probability that an abnormality exists in the working state of the monitored power equipment;

[0047] Calculating the product of a first power device operating state-temperature interactive coupling characteristic value of the power device operating state-temperature interactive coupling characteristic vector and the power device operating state-temperature interactive coupling probability value to obtain a first power device operating state-temperature interactive coupling characteristic probability product;

[0048] Calculating the product of a second power device operating state-temperature mutual coupling characteristic value of the power device operating state-temperature mutual coupling characteristic vector and a difference between one and the power device operating state-temperature mutual coupling probability value to obtain a second power device operating state-temperature mutual coupling characteristic inverse probability product;

[0049] Calculating the quotient of the first electric power device operating state-temperature mutual coupling characteristic value divided by one minus the electric power device operating state-temperature mutual coupling probability value to obtain the first electric power device operating state-temperature mutual coupling characteristic inverse probability quotient;

[0050] Calculating the quotient of the second electric power device operating state-temperature interactive coupling characteristic value and the electric power device operating state-temperature interactive coupling probability value to obtain a second electric power device operating state-temperature interactive coupling characteristic probability quotient;

[0051] After adding the product of the probability of the first power device working state-temperature presenting interactive coupling characteristics and the product of the inverse probability of the second power device working state-temperature presenting interactive coupling characteristics, subtracting the difference between the inverse probability quotient of the first power device working state-temperature presenting interactive coupling characteristics and the probability quotient of the second power device working state-temperature presenting interactive coupling characteristics to obtain a corrected characteristic value of the working state-temperature presenting interactive coupling of the power device;

[0052] For each first power device operating state-temperature presentation interactive coupling eigenvalue and second power device operating state-temperature presentation interactive coupling eigenvalue of the power device operating state-temperature presentation interactive coupling eigenvalue, matrix multiplying a power device operating state-temperature presentation interactive coupling correction characteristic matrix composed of the power device operating state-temperature presentation interactive coupling correction eigenvalues ​​by the power device operating state-temperature presentation interactive coupling eigenvector to obtain a corrected power device operating state-temperature presentation interactive coupling eigenvector; and

[0053] The corrected power equipment working state-temperature interactive coupling feature vector is input into a classifier-based state monitoring module to obtain a state monitoring result.

[0054] The correction process of the interactive coupling characteristic vector V of the working state and temperature of the power equipment is expressed as follows:

[0055]

[0056] m i,j ∈M,v i ∈V,v j(j≠i) ∈V

[0057]

[0058] Where V is the interactive coupling characteristic vector of the working state and temperature of the power equipment, v i is the first power device working state-temperature interactive coupling eigenvalue of the power device working state-temperature interactive coupling eigenvector, v j is the second power equipment working state-temperature interactive coupling eigenvalue of the power equipment working state-temperature interactive coupling eigenvector, m i,j is the eigenvalue of each position of the interactive coupling correction characteristic matrix of the working state and temperature of the power equipment, p is the probability that the working state of the monitored power equipment is abnormal, is the matrix multiplication, and V' is the corrected power device operating state-temperature interaction coupling eigenvector.

[0059] That is, when the power equipment working state time series representation vector and the temperature time series association implicit feature vector respectively represent the time series association superposition features of the current value and voltage value of the monitored power equipment and the time series association features of the temperature value of the monitored power equipment, when performing interactive response coupling based on the optimal matching of the feature principal component granularity, the power equipment working state-temperature interactive coupling feature vector will also cause interactive response diversity based on the optimal matching due to the difference in the feature principal component granularity time series pattern, affecting the iterative consistency of the classification mapping and reducing the accuracy of the classification results.

[0060] Therefore, the applicant of this application aims at the region-boundary integral relationship of the high-dimensional feature manifold of the interactively coupled eigenvector of the working state of the power equipment - temperature, and defines the overall probability distribution of the feature set of the interactively coupled eigenvector of the working state of the power equipment - temperature as the manifold constraint boundary to approximate the simply connected region representation of the high-dimensional manifold of the feature set of the interactively coupled eigenvector of the working state of the power equipment - temperature, which is composed of eigenvalue pairs, so as to avoid the ambiguity in the mapping of the diversified feature representation of the interactively coupled eigenvector of the working state of the power equipment - temperature to the local manifold representation in the high-dimensional feature class convergence space, improve the execution iteration consistency of each local feature distribution in the mapping task, so as to improve the feature classification convergence effect and improve the accuracy of the status monitoring results obtained by the classifier-based status monitoring module.

[0061] In summary, according to the embodiment of the present application, the remote monitoring method of power equipment based on 5G short slices is explained, which configures 5G network short slices for power equipment based on network slicing technology to remotely monitor the operating status of the power equipment, and at the same time uses data processing technology based on deep learning to perform time series analysis on the collected current, voltage and temperature data, and comprehensively characterizes the working status of the power equipment based on the time series joint information of the current data and voltage data. Furthermore, by performing fine-grained interactive response analysis on the working status characteristics and equipment temperature characteristics of the power equipment, the abnormal correlation response pattern between the two can be identified, thereby realizing real-time monitoring of the working status of the power equipment and abnormality identification. In this way, the accuracy and real-time performance of power equipment monitoring can be effectively improved, which helps to timely discover potential faults and abnormal conditions and reduce the risk of power supply interruption due to equipment failure.

[0062] While various embodiments of the present disclosure have been described above, the above descriptions are illustrative, non-exhaustive, and not intended to be limiting of the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is selected to best explain the principles of the embodiments, their practical applications, or improvements to existing technologies, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A remote monitoring method for power equipment based on 5G short slices, characterized in that: include: Connecting the monitored power equipment to the 5G network and configuring a 5G network short slice for the monitored power equipment; A time queue of state parameters of the monitored power equipment is collected by a sensor group, wherein the state parameters include current value, voltage value and temperature value; Transmitting the time queue of the status parameters of the monitored power equipment to a status remote monitoring center via the 5G network short slice; At the state remote monitoring center, extracting the temperature characteristics and working state characteristics of the power equipment from the time queue of the state parameters of the monitored power equipment to obtain a temperature time series associated implicit feature vector and a power equipment working state time series representation vector; At the state remote monitoring center, a fine-grained interactive response analysis is performed on the power equipment working state time series representation vector and the temperature time series associated implicit feature vector to obtain a power equipment working state-temperature presentation interactive coupling feature vector; At the state remote monitoring center, determining whether the working state of the monitored power equipment is abnormal based on the interactive coupling characteristic vector presented by the working state-temperature of the power equipment; Among them, a fine-grained interactive response analysis is performed on the power equipment working state time series representation vector and the temperature time series associated implicit feature vector, including: performing principal component analysis on the power equipment working state time series representation vector and the temperature time series associated implicit feature vector respectively to obtain a set of power equipment working state time series principal component feature components and a set of temperature time series principal component feature components; performing optimal pairing multi-scale interactive coupling on the set of power equipment working state time series principal component feature components and the set of temperature time series principal component feature components to obtain the power equipment working state-temperature presentation interactive coupling feature vector.

2. The method for remote monitoring of power equipment based on 5G short slices according to claim 1 is characterized in that: Extracting the temperature characteristics and the working state characteristics of the power equipment from the time queue of the state parameters of the monitored power equipment to obtain the temperature time series associated implicit feature vector and the power equipment working state time series representation vector, including: Queuing the time queue of the state parameters of the monitored power equipment according to the parameter sample dimension to obtain a time queue of current value, a time queue of voltage value and a time queue of temperature value; Performing time series feature extraction on the time queue of the current value, the time queue of the voltage value, and the time queue of the temperature value to obtain a current time series associated implicit feature vector, a voltage time series associated implicit feature vector, and the temperature time series associated implicit feature vector; The current time series associated implicit feature vector and the voltage time series associated implicit feature vector are input into a joint characterization network of the working state of the power equipment to obtain the time series representation vector of the working state of the power equipment.

3. The method for remote monitoring of power equipment based on 5G short slices according to claim 2 is characterized in that: Performing time series feature extraction on the time queue of the current value, the time queue of the voltage value, and the time queue of the temperature value to obtain a current time series associated implicit feature vector, a voltage time series associated implicit feature vector, and a temperature time series associated implicit feature vector, including: The time queue of the current value, the time queue of the voltage value and the time queue of the temperature value are input into a time queue encoder based on a bidirectional LSTM model to obtain the current time series associated implicit feature vector, the voltage time series associated implicit feature vector and the temperature time series associated implicit feature vector.

4. The method for remote monitoring of power equipment based on 5G short slices according to claim 3 is characterized in that: Performing principal component analysis on the power equipment working state time series representation vector to obtain a set of principal component characteristic components of the power equipment working state time series, including: Calculate the covariance matrix of the power equipment working state time series representation vector to obtain the power equipment working state time series feature covariance matrix, and perform eigenvalue decomposition on the power equipment working state time series feature covariance matrix to obtain a set of power equipment working state time series principal component eigenvalues ​​and a corresponding set of power equipment working state time series principal component feature components.

5. The method for remote monitoring of power equipment based on 5G short slices according to claim 4 is characterized in that: Performing optimal pairing multi-scale interactive coupling on a set of principal component characteristic components of the power equipment working state time series and a set of principal component characteristic components of the temperature time series to obtain the power equipment working state-temperature interactive coupling characteristic vector, including: Using each power equipment working state time series principal component characteristic component in the set of power equipment working state time series principal component characteristic components as a query vector, and using the set of temperature time series principal component characteristic components as a query library, matching the temperature time series principal component characteristic component that best matches each query vector from the query library to obtain a set of best matching pairs of {power equipment working state time series principal component characteristic component; temperature time series principal component characteristic component}; Multi-scale interactive fusion is performed on the set of the best matching pairs of {power equipment working state time series principal component characteristic component; temperature time series principal component characteristic component} to obtain the power equipment working state-temperature interactive coupling characteristic vector.

6. The method for remote monitoring of power equipment based on 5G short slices according to claim 5, characterized in that: Matching the temperature time series principal component characteristic component that best matches each query vector from the query library to obtain {power equipment working state time series principal component characteristic component; The set of the best matching pairs of temperature time series principal component characteristic components includes: Calculating the Mahalanobis distance between the query vector and each temperature time series principal component characteristic component in the query library to obtain a set of power equipment working state-temperature principal component matching difference coefficients; The temperature time series principal component characteristic component corresponding to the minimum value in the set of the power equipment working state-temperature principal component matching difference coefficients is selected as the optimal matching result of the query vector to obtain the {power equipment working state time series principal component characteristic component; temperature time series principal component characteristic component} best matching pair.

7. The method for remote monitoring of power equipment based on 5G short slices according to claim 6 is characterized in that: The {power equipment working state time series principal component characteristic component; The set of the best matching pairs of the temperature time series principal component characteristic component is subjected to multi-scale interactive fusion to obtain the interactive coupling characteristic vector of the power equipment working state-temperature presentation, including: Input each best matching pair of {power equipment working state time series principal component characteristic component; temperature time series principal component characteristic component} in the set of best matching pairs of {power equipment working state time series principal component characteristic component; temperature time series principal component characteristic component} into the multi-scale interactive response coupling module to obtain a set of multi-scale interactive coupling representation vectors of the best matching pairs of power equipment working state and temperature; The set of multi-scale interaction coupling representation vectors of the best matching of the working state of the power device and the temperature is cascaded to obtain the interaction coupling feature vector of the working state of the power device and the temperature.

8. The method for remote monitoring of power equipment based on 5G short slices according to claim 7 is characterized in that: The {power equipment working state time series principal component characteristic component; Each {main component characteristic component of the time series of the temperature} best matching pair in the set of {main component characteristic component of the time series of the power equipment working state; The best matching pairs of the temperature time series principal component characteristic components are input into the multi-scale interactive response coupling module to obtain a set of multi-scale interactive coupling representation vectors of the power equipment working state-temperature best matching pairs, including: Calculate the {power equipment working state time series principal component characteristic components respectively; The temperature time series principal component characteristic component} the power equipment working state time series principal component characteristic component and the temperature time series principal component characteristic component in the best matching pair are added, subtracted and multiplied by position points to obtain a first best matching interaction representation vector, a second best matching interaction representation vector and a third best matching interaction representation vector; The first best matching interaction representation vector, the second best matching interaction representation vector, and the third best matching interaction representation vector are cascaded and fused, and then input into a one-dimensional convolutional layer including a maximum pooling layer to obtain a multi-scale interaction representation vector of the best matching of the working state of the power equipment and the temperature principal component; The second norm of the characteristic component of the principal component of the working state time series of the power equipment, the second norm of the characteristic component of the principal component of the temperature time series and the maximum value of the preset scale adjustment parameter are selected as the scale scaling factor, and each eigenvalue in the multi-scale interaction representation vector of the best match between the working state of the power equipment and the temperature principal component is divided by the scale scaling factor to obtain the multi-scale interaction coupling representation vector of the best match between the working state of the power equipment and the temperature.

9. The method for remote monitoring of power equipment based on 5G short slices according to claim 8, characterized in that: Determining whether the working state of the monitored power equipment is abnormal based on the interactive coupling characteristic vector of the working state and temperature of the power equipment includes: The power equipment working state-temperature interactive coupling feature vector is input into a classifier-based state monitoring module to obtain a state monitoring result, which is used to indicate whether the working state of the monitored power equipment is abnormal.

Images