Power cable state evaluation method and system based on multi-source data collaboration

Abstract

The invention discloses a power cable state evaluation method and system based on multi-source data collaboration, and belongs to the technical field of state evaluation. Comprising the steps of obtaining an electrical parameter set and a cable surface temperature of a power cable based on a preset time granularity; performing feature selection on the electrical parameter set based on the information gain, and determining a state evaluation feature quantity; acquiring the core temperature of the power cable based on the cable surface temperature through a preset power cable equivalent thermal circuit model and attribute parameters; fusing the state characteristic quantity and the core temperature, and inputting the fused state characteristic quantity and the core temperature into a trained state evaluation model for processing to obtain a state evaluation result of the power cable; wherein the state evaluation model comprises a convolutional layer, a pooling layer, a plurality of groups of LSTM networks and a CRF module which are connected in sequence. The accuracy and the real-time performance of power cable operation monitoring can be improved, and the problem that the real-time performance and the accuracy of existing power cable state evaluation are poor is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of status assessment, and in particular to a method and system for power cable status assessment based on multi-source data collaboration. Background Art

[0002] The statements in this section merely mention background art related to the present invention and do not necessarily constitute prior art.

[0003] Power cable operation and maintenance mainly refers to the daily maintenance, status assessment, fault handling and preventive management of cable lines used in power systems to ensure safe, stable and efficient power supply. This work is crucial to the reliable operation of power systems.

[0004] Condition assessment is an important part of power cable operation and maintenance. It aims to evaluate and predict the health status of power cables, and further design operation and maintenance plans based on the health status of power cables to improve operation and maintenance efficiency and reduce operation and maintenance costs. Existing power cable condition assessments mostly use Internet of Things (IOT) technology to integrate a variety of monitoring sensors and data analysis platforms to achieve remote monitoring, data analysis, condition assessment and early warning functions of cable systems, improve operation and maintenance efficiency and response speed, and ensure the long-term stable operation of power cables. However, existing power methods still have the following shortcomings: 1. Most power cables are three-core distribution cable structures. The temperature of the cable core is an important factor in judging the health status of the cable. However, since its data is difficult to collect through sensors, the surface temperature of the power cable is often used for status assessment, which cannot reflect the true status inside the power cable, making the accuracy of the status assessment insufficient.

[0005] 2. In the existing technology, most of the health status of power cables are evaluated directly through the operating data of the power cables (such as electrical parameters such as current and voltage). There are many types of operating data and the data volume is huge. The information provided has different effects on the health status evaluation of the power cables, which increases the complexity of the health status evaluation and affects the real-time performance of data processing.

[0006] 3. The existing power cable status assessment model has limited feature extraction capabilities and cannot effectively utilize the time series characteristics of the data, which affects the effectiveness of the power cable status assessment. Summary of the invention

[0007] In order to address the deficiencies in the prior art, the present invention provides a method, system, electronic device and computer-readable storage medium for power cable status assessment based on multi-source data collaboration, which selects multiple parameters that can comprehensively reflect the actual operating status of the power cable to collaboratively conduct power cable status assessment, thereby improving the accuracy and real-time performance of power cable status assessment.

[0008] In a first aspect, the present invention provides a method for evaluating the state of a power cable based on multi-source data collaboration; A power cable condition assessment method based on multi-source data collaboration, comprising: Based on a preset time granularity, obtain the electrical parameter set and cable surface temperature of the power cable; Based on information gain, feature selection is performed on the electrical parameter set to determine the state assessment feature quantity; through the preset power cable equivalent thermal circuit model and attribute parameters, the core temperature of the power cable is obtained based on the cable surface temperature; The state feature quantity is integrated with the core temperature and input into the trained state assessment model for processing to obtain the state assessment result of the power cable; The state assessment model includes a convolutional layer, a pooling layer, multiple groups of LSTM networks and a CRF module connected in sequence.

[0009] In some implementations, the performing feature selection on the electrical parameters based on information gain to determine the state evaluation feature quantity specifically includes: According to the electrical parameter set, calculating the overall information entropy of the electrical parameter set and the information entropy of each electrical parameter in the electrical parameter set; According to the overall information entropy and the information entropy of each electrical parameter, the information gain of each electrical parameter is determined and arranged in ascending order to filter the state evaluation feature quantity.

[0010] In some embodiments, the core temperature of the power cable is expressed as: ; in, is the core temperature, is the current, is the resistance, is the thermal resistance of the insulation layer, is the thermal resistance of the lining layer, is the thermal resistance of the epidermis, is the ambient medium thermal resistance, is the armor layer loss coefficient, is the cable surface temperature.

[0011] In some implementations, the fusing of the state characteristic quantity with the wire core temperature specifically includes: The grey correlation method is used to analyze the correlation between the state characteristic quantity and the core temperature and the power cable state, and determine the correlation between them and the power cable state. According to the correlation degree, the state feature quantity and the core temperature are weighted and fused.

[0012] In some embodiments, the fused state feature quantity and the core temperature are input into a trained state assessment model for processing, specifically including: The fused state feature quantity and core temperature are input into the convolution layer for convolution calculation to obtain the local feature matrix; The local feature matrix is ​​pooled through the pooling layer to obtain multiple feature mapping matrices and compress them, generate feature matrices and input them into multiple groups of LSTM networks for processing to obtain multiple groups of prediction results; Multiple sets of prediction results are input into the CRF layer for processing to determine the final state evaluation results.

[0013] In some implementations, multiple groups of prediction results are input into the CRF layer for processing to obtain the correct probability of each group of prediction results, and the prediction result with the highest probability is used as the final state evaluation result.

[0014] In some implementations, the activation function of the convolutional layer is a tanh function, and the activation function of the hidden layer in the LSTM network is a Relu activation function.

[0015] In a second aspect, the present invention provides a power cable status assessment system based on multi-source data collaboration; A power cable status assessment system based on multi-source data collaboration, comprising: The acquisition module is configured to: acquire an electrical parameter set and a cable surface temperature of the power cable based on a preset time granularity; The feature selection module is configured to: perform feature selection on the electrical parameter set based on information gain to determine the state evaluation feature quantity; obtain the core temperature of the power cable based on the cable surface temperature through a preset power cable equivalent thermal circuit model and attribute parameters; The state assessment module is configured to: fuse the state feature quantity with the core temperature, and input the trained state assessment model for processing to obtain the state assessment result of the power cable; The state assessment model includes a convolutional layer, a pooling layer, multiple groups of LSTM networks and a CRF module connected in sequence.

[0016] In a third aspect, the present invention provides an electronic device; An electronic device comprises a memory and a processor and computer instructions stored in the memory and executed on the processor. When the computer instructions are executed by the processor, the steps of the power cable status assessment method based on multi-source data collaboration are completed.

[0017] In a fourth aspect, the present invention provides a computer-readable storage medium; A computer-readable storage medium is used to store computer instructions. When the computer instructions are executed by a processor, the steps of the above-mentioned power cable status assessment method based on multi-source data collaboration are completed.

[0018] Compared with the prior art, the present invention has the following beneficial effects: 1. The technical solution provided by the present invention utilizes the core temperature of the power cable and other characteristics that are strongly related to the health status of the power cable to collaboratively evaluate the status of the power cable. On the one hand, it can comprehensively and truly reflect the health status of the power cable. On the other hand, it reduces the complexity of data analysis and the amount of data processing during status evaluation, and improves the real-time and robustness of status evaluation.

[0019] 2. The technical solution provided by the present invention combines the CNN network, LSTM network and CRF layer for operation monitoring, which can improve the model training speed while suppressing overfitting. When performing evaluation, the attention mechanism is used to give greater weights to important parameters to improve the accuracy of state evaluation. DETAILED DESCRIPTION

[0020] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used in the present invention have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0021] It should be noted that the terms used herein are only for describing specific embodiments, and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that the terms "include" and "have" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0022] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0023] Embodiment 1 The amount of data used in cable operation monitoring in the prior art is large and cannot fully reflect the actual operating status of the power cable. Therefore, the present invention provides a power cable status assessment method based on multi-source data collaboration.

[0024] Next, a method for evaluating the state of a power cable based on multi-source data collaboration disclosed in this embodiment is described in detail. The method for evaluating the state of a power cable based on multi-source data collaboration comprises the following steps: S1. Based on a preset time granularity, obtain an electrical parameter set and a cable surface temperature of the power cable.

[0025] Temperature is one of the most intuitive manifestations of cable status. Once the cable runs abnormally, overheating will occur. The cable surface temperature is the easiest temperature information to collect. Therefore, in this embodiment, the subsequent calculation of the battery core temperature is based on the cable surface temperature to pre-detect abnormalities in the cable status. Electrical parameters are the most accurate manifestation of the power cable status. The power cable status is evaluated based on the above data to improve the authenticity of the method described in this embodiment.

[0026] In this embodiment, the electrical parameters include time series data such as voltage, current, partial discharge signal, and power.

[0027] S2. Select features of the electrical parameter set based on information gain to determine the characteristic quantity of state evaluation; Since a huge amount of operation data is generated during the operation of the power cable, data processing takes too long; and the types of operation data are diverse, and the impact on the health status assessment of the power cable is also different, which is easy to interfere with the status assessment and affect its robustness. Therefore, in this embodiment, the information gain is used to select the electrical parameters in the electrical parameter set to screen out the electrical parameters with the most distinguishing ability and remove the electrical parameters that do not provide important information.

[0028] As an implementation mode, S2 specifically includes: S201. Calculate, according to the electrical parameter set, the overall information entropy of the electrical parameter set and the information entropy of each electrical parameter in the electrical parameter set.

[0029] Among them, the overall information entropy of the electrical parameter set is expressed as follows: ; In the formula, Represents the electrical parameter set, Indicates the proportion of the i-th type of electrical parameters in the electrical parameter set, is the total number of electrical parameter types.

[0030] The information entropy of electrical parameters is expressed as follows: ; In the formula, Indicates that the value is The number of electrical parameters, represents the number of electrical parameters belonging to category k, Indicates electrical parameters.

[0031] S202: Determine the information gain of each electrical parameter according to the overall information entropy and the information entropy of each electrical parameter.

[0032] Among them, the information gain is expressed as follows: .

[0033] S203 , sort the information gains in ascending order, and use the first 30% of the electrical parameters as state evaluation feature quantities.

[0034] S3. Obtain the property parameters of the power cable and the cable surface temperature, and determine the core temperature of the power cable according to a preset power cable equivalent thermal circuit model, property parameters and cable surface temperature.

[0035] Monitoring the internal temperature of the power cable by means of a sensor requires modification of the structure of the power cable, which is not applicable to power cables that have already been put into use and is easily affected by the environment. Therefore, in this embodiment, the core temperature is calculated using an equivalent thermal circuit model of the power cable, which requires little data processing and is easy to implement.

[0036] As an implementation method, S3 specifically includes: S301. Construct an equivalent thermal circuit model of the power cable according to the structure of the power cable.

[0037] In this embodiment, the power cable is a three-core distribution cable structure, which includes an insulation layer, an inner lining layer and a skin layer. The equivalent thermal circuit model of the power cable is expressed as: ; in, is the core temperature, is the current, is the resistance, is the thermal resistance of the insulation layer, is the thermal resistance of the lining layer, is the thermal resistance of the epidermis, is the ambient medium thermal resistance, is the armor layer loss coefficient, is the cable surface temperature.

[0038] S302. Determine the thermal resistance of the insulation layer according to the thermal resistance coefficient, thickness and diameter of the insulation layer; determine the thermal resistance of the inner lining layer according to the thermal resistance coefficient, thickness and diameter of the inner lining layer; determine the thermal resistance of the epidermis layer according to the thermal resistance coefficient, thickness and diameter of the epidermis layer; determine the thermal resistance of the ambient medium according to the thermal resistance coefficient of the soil, the cable burial depth and the outer diameter of the cable.

[0039] Among them, the thermal resistance of the insulation layer is expressed as: ; In the formula, is the thermal resistance of the insulation layer, is the thermal resistance coefficient of the insulation layer, is the thickness of the insulation layer, is the insulation diameter.

[0040] The thermal resistance of the lining layer is expressed as: ; In the formula, is the thermal resistance of the lining layer, is the thermal resistance coefficient of the lining layer, is the thickness of the lining layer, is the inner lining diameter.

[0041] The thermal resistance of the skin layer is expressed as: ; In the formula, is the thermal resistance of the epidermis, is the thermal resistance coefficient of the skin layer, is the thickness of the epidermis, is the diameter of the epidermis.

[0042] The thermal resistance of the ambient medium is expressed as: ; In the formula, is the soil thermal resistance coefficient, is the cable burial depth, is the outer diameter of the cable.

[0043] S303, the thermal resistance of the insulation layer, the thermal resistance of the lining layer, the thermal resistance of the skin layer and the thermal resistance of the ambient medium are brought into the equivalent thermal circuit model of the power cable to determine the core temperature, which is expressed as follows: ; in, is the core temperature, is the current, is the resistance, is the thermal resistance of the insulation layer, is the thermal resistance of the lining layer, is the thermal resistance of the epidermis, is the ambient medium thermal resistance, is the armor layer loss coefficient, is the cable surface temperature.

[0044] In this embodiment, the resistance and armor loss coefficient are factory-provided parameters of the power cable, and the current can be measured by a current sensor. Through this step, the real-time core temperature of the power cable can be obtained, and the temperature inside the power cable can be monitored to reflect the real state inside the power cable, providing accurate basic data for the health status assessment of the power cable.

[0045] S4. The state feature quantity is integrated with the core temperature and input into the trained state assessment model for processing to obtain the state assessment result of the power cable.

[0046] In order to quantify the influence of different features on the power cable status, the status feature quantity is fused with the core temperature before status assessment. In order to improve the feature capability of time series data, the status assessment model is improved. The improved status assessment model includes sequentially connected convolutional layers, pooling layers, multiple groups of LSTM networks and CRF modules.

[0047] As an implementation mode, S4 specifically includes: S401. The state characteristic quantity, the core temperature and the power cable state are analyzed by the grey correlation method to determine the correlation between the state characteristic quantity and the core temperature and the power cable state; and the state characteristic quantity and the core temperature are weighted and fused according to the correlation. The correlation is expressed as follows: ; In the formula, is the correlation coefficient, expressed as: ; In the formula, is the power cable state sequence, is the resolution coefficient, which is generally taken as 0.5. is the state characteristic sequence or the core temperature sequence, Indicates the position in the sequence.

[0048] Finally, after normalizing the correlation between the state feature quantity and the core temperature, the state feature quantity and the core temperature are weighted and fused as weights to obtain a feature data matrix.

[0049] S402: Input the weighted fused state feature quantity and the core temperature into the trained state assessment model for processing to obtain the state assessment result of the power cable. Specifically including: S4021. Input the feature data matrix into the convolution layer for convolution calculation to obtain a local feature matrix.

[0050] Specifically, the local feature matrix is ​​expressed as: ; In the formula, represents the weight, Represents the convolution vector matrix obtained by convolution calculation of the fused state feature quantity and the core temperature. represents the bias matrix, It is the Rectified Linear Unit (ReLU) activation function.

[0051] S4022. Perform a pooling operation on the convolution vector matrix through a pooling layer, obtain multiple feature mapping matrices and compress them to generate a feature matrix, and the dimension of the feature matrix is ​​the same as the dimension of the input feature data matrix.

[0052] Among them, the feature matrix is ​​expressed as: ; In the formula, is the feature mapping matrix, is the number of feature matrices.

[0053] S4023. Input the feature matrix into multiple groups of LSTM networks for processing to obtain multiple groups of prediction results.

[0054] In this embodiment, the LSTM network adopts the existing LSTM network, including a forget gate, an input gate, an output gate and a memory unit, which are not improved here and will not be described in detail.

[0055] Among them, the prediction result is expressed as: ; ; ; ; ; In the formula, is the output at time t, is the initial output obtained through the sigmoid layer, Tanh is the hyperbolic tangent activation function, is the information update at time t, is the output obtained through the sigmoid layer, is the information update at time t-1, is the vector obtained by activating the elements in the vector i that needs to be saved by the sigmoid function. For new candidate values ​​generated using the Tanh layer, is the sigmoid activation function, is the input at time t, is the output at time t-1, , , , , , represents the weight matrix, , Represents the bias vector.

[0056] In this embodiment, the result output by each LSTM network at time point t is the health status score of the power cable at time point t. When training the LSTM network, the weighted fused state feature quantity and core temperature are obtained by processing the existing data and adding the health status score to construct a training set, and the LSTM network is trained by the training set.

[0057] S4024. Input the multiple groups of prediction results into the CRF layer for processing, obtain the correct probability of each group of prediction results, and use the prediction result with the largest probability as the final state evaluation result.

[0058] For example, the prediction result is expressed as , which is the input of the CRF layer, the given label is represented as , then for a given input Y, the probability that each set of predictions is correct is: ; ; In the formula, express The probability of predicting correctly, represents the prediction result with the highest probability, Represents a numerical maximization function.

[0059] Embodiment 2 This embodiment discloses a power cable status assessment system based on multi-source data collaboration, including: The acquisition module is configured to: acquire an electrical parameter set and a cable surface temperature of the power cable based on a preset time granularity; The feature selection module is configured to: perform feature selection on the electrical parameter set based on information gain to determine the state evaluation feature quantity; obtain the core temperature of the power cable based on the cable surface temperature through a preset power cable equivalent thermal circuit model and attribute parameters; The state assessment module is configured to: fuse the state feature quantity with the core temperature, and input the trained state assessment model for processing to obtain the state assessment result of the power cable; The state assessment model includes a convolutional layer, a pooling layer, multiple groups of LSTM networks and a CRF module connected in sequence.

[0060] It should be noted that the acquisition module, feature selection module and state evaluation module correspond to the steps in Embodiment 1, and the examples and application scenarios implemented by the modules and the corresponding steps are the same, but are not limited to the contents disclosed in Embodiment 1. It should be noted that the modules as part of the system can be executed in a computer system such as a set of computer executable instructions.

[0061] Embodiment 3 Embodiment 3 of the present invention provides an electronic device, including a memory and a processor, and computer instructions stored in the memory and executed on the processor. When the computer instructions are executed by the processor, the steps of the above-mentioned power cable status assessment method based on multi-source data collaboration are completed.

[0062] Embodiment 4 Embodiment 4 of the present invention provides a computer-readable storage medium for storing computer instructions. When the computer instructions are executed by a processor, the steps of the above-mentioned power cable status assessment method based on multi-source data collaboration are completed.

[0063] The present invention is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate a device for implementing the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0064] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device that implements the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0065] These computer program instructions may also be loaded onto a computer or other programmable data processing device, and a series of operating steps may be executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0066] The description of each embodiment in the above embodiments has different emphases. For parts not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0067] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for evaluating the state of a power cable based on multi-source data collaboration, characterized in that: include: Based on a preset time granularity, obtain the electrical parameter set and cable surface temperature of the power cable; Perform feature selection on the electrical parameter set based on information gain to determine the characteristic quantity of state evaluation; The core temperature of the power cable is obtained based on the cable surface temperature through the preset power cable equivalent thermal circuit model and property parameters; The state feature quantity is integrated with the core temperature and input into the trained state assessment model for processing to obtain the state assessment result of the power cable; The state assessment model includes a convolutional layer, a pooling layer, multiple groups of LSTM networks and a CRF module connected in sequence.

2. The power cable status assessment method based on multi-source data collaboration according to claim 1 is characterized in that: The feature selection of electrical parameters based on information gain to determine the state evaluation feature quantity specifically includes: According to the electrical parameter set, calculating the overall information entropy of the electrical parameter set and the information entropy of each electrical parameter in the electrical parameter set; According to the overall information entropy and the information entropy of each electrical parameter, the information gain of each electrical parameter is determined and arranged in ascending order to filter the state evaluation feature quantity.

3. The power cable status assessment method based on multi-source data collaboration according to claim 1 is characterized in that: The core temperature of the power cable is expressed as: ； in, is the core temperature, is the current, is the resistance, is the thermal resistance of the insulation layer, is the thermal resistance of the lining layer, is the thermal resistance of the epidermis, is the ambient medium thermal resistance, is the armor layer loss coefficient, is the cable surface temperature.

4. The method for evaluating power cable status based on multi-source data collaboration according to claim 1, characterized in that: The fusion of the state characteristic quantity and the core temperature specifically includes: The grey correlation method is used to analyze the correlation between the state characteristic quantity and the core temperature and the power cable state, and determine the correlation between them and the power cable state. According to the correlation degree, the state feature quantity and the core temperature are weighted and fused.

5. The method for evaluating power cable status based on multi-source data collaboration according to claim 1, characterized in that: The fused state feature quantity and core temperature are input into the trained state assessment model for processing, which specifically includes: The fused state feature quantity and core temperature are input into the convolution layer for convolution calculation to obtain the local feature matrix; The local feature matrix is ​​pooled through the pooling layer to obtain multiple feature mapping matrices and compress them, generate feature matrices and input them into multiple groups of LSTM networks for processing to obtain multiple groups of prediction results; Multiple sets of prediction results are input into the CRF layer for processing to determine the final state evaluation results.

6. The method for evaluating power cable status based on multi-source data collaboration according to claim 5, characterized in that: The step of inputting multiple groups of prediction results into the CRF layer for processing is specifically as follows: inputting multiple groups of prediction results into the CRF layer for processing, obtaining the correct probability of each group of prediction results, and taking the prediction result with the largest probability as the final state evaluation result.

7. The method for evaluating power cable status based on multi-source data collaboration according to claim 1, characterized in that: The activation function of the convolutional layer is the tanh function, and the activation function of the hidden layer in the LSTM network is the Relu activation function.

8. A power cable status assessment system based on multi-source data collaboration, characterized in that: include: The acquisition module is configured to: acquire an electrical parameter set and a cable surface temperature of the power cable based on a preset time granularity; The feature selection module is configured to: perform feature selection on the electrical parameter set based on information gain to determine a state evaluation feature quantity; The core temperature of the power cable is obtained based on the cable surface temperature through the preset power cable equivalent thermal circuit model and property parameters; The state assessment module is configured to: fuse the state feature quantity with the core temperature, and input the trained state assessment model for processing to obtain the state assessment result of the power cable; The state assessment model includes a convolutional layer, a pooling layer, multiple groups of LSTM networks and a CRF module connected in sequence.

9. An electronic device, characterized in that: The invention comprises a memory and a processor and computer instructions stored in the memory and executed on the processor. When the computer instructions are executed by the processor, the method for evaluating the state of a power cable based on multi-source data collaboration as described in any one of claims 1 to 7 is completed.

10. A computer-readable storage medium, characterized in that: Used to store computer instructions, which, when executed by a processor, complete the power cable status assessment method based on multi-source data collaboration as described in any one of claims 1 to 7.

Images