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

Abstract

The application discloses a kind of power cable state evaluation method and system based on multi-source data cooperation, belong to state evaluation technical field.It includes: based on the time granularity of pre-set, the electrical parameter set and cable surface temperature of power cable are obtained;Based on information gain, the feature selection is carried out to the electrical parameter set, and the state evaluation characteristic quantity is determined;Through the equivalent thermal circuit model and attribute parameter of pre-set power cable, based on cable surface temperature, the core temperature of power cable is obtained;State characteristic quantity is fused with core temperature, and input trained state evaluation model is handled, and the state evaluation result of power cable is obtained;Among them, the state evaluation model includes convolution layer, pooling layer, multiple groups of LSTM network and CRF module connected in turn.Can improve the accuracy and real-time performance of power cable operation monitoring, solve the problem that the real-time performance and accuracy of existing power cable state evaluation are poor.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of state evaluation, in particular to a power cable state evaluation method and system based on multi-source data collaboration. BACKGROUND

[0002] The statements in this section merely refer to the background of the present application and do not necessarily constitute the prior art.

[0003] Power cable operation and maintenance mainly refers to the daily maintenance, state evaluation, fault handling and preventive management of cable lines used in the power system to ensure the safety, stability and efficiency of power supply, which is crucial for the reliable operation of the power system.

[0004] State evaluation is an important part of power cable operation and maintenance, aiming to evaluate and predict the health status of power cables, and further design operation and maintenance schemes according to the health status of power cables to improve operation and maintenance efficiency and reduce operation and maintenance cost. The existing state evaluation of power cables mostly uses Internet of Things (IOT) technology, integrates various monitoring sensors and data analysis platforms to realize remote monitoring, data analysis, state evaluation and early warning functions of cable systems, and improves operation and maintenance efficiency and response speed to ensure long-term stable operation of power cables. However, the existing power method still has the following shortcomings:

[0005] 1. Power cables are mostly three-core distribution cable structures, and the temperature of cable cores is an important factor for judging the health status of cables, but since the data is difficult to collect through sensors, the surface temperature of power cables is mostly used for state evaluation, which cannot reflect the true state of power cables, resulting in insufficient accuracy of state evaluation.

[0006] 2. In the existing technology, most of the health status evaluation of power cables is directly based on the operation data of power cables (such as current, voltage and other electrical parameters), and the operation data types are numerous and the data volume is large, the information provided by them has different effects on the health status evaluation of power cables, increasing the complexity of health status evaluation and affecting the real-time performance of data processing.

[0007] 3. The existing power cable state evaluation model has limited feature extraction capability and cannot effectively utilize the time sequence features of data, affecting the effectiveness of power cable state evaluation. SUMMARY

[0008] In order to solve the problems of the prior art, the present application provides a power cable state evaluation method and system based on multi-source data collaboration, which selects multiple parameters that can comprehensively reflect the true operation state of power cables to collaboratively evaluate the state of power cables, improving the accuracy and real-time performance of power cable state evaluation.

[0009] In a first aspect, the present application provides a power cable state evaluation method based on multi-source data collaboration;

[0010] A power cable state evaluation method based on multi-source data collaboration comprises:

[0011] Based on a preset time granularity, an electrical parameter set and a cable surface temperature of the power cable are obtained;

[0012] Based on information gain, feature selection is performed on the electrical parameter set to determine state evaluation feature quantities; based on the cable surface temperature, a core temperature of the power cable is obtained through a preset equivalent thermal circuit model of the power cable and attribute parameters;

[0013] The state feature quantities and the core temperature are fused and input into a trained state evaluation model for processing to obtain a state evaluation result of the power cable;

[0014] The state evaluation model comprises convolution layers, pooling layers, multiple groups of LSTM networks and a CRF module connected in sequence.

[0015] In some embodiments, the feature selection based on information gain on the electrical parameters to determine the state evaluation feature quantities specifically comprises:

[0016] 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 are calculated;

[0017] 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, and the state evaluation feature quantities are screened.

[0018] In some embodiments, the core temperature of the power cable is represented as:

[0019] ;

[0020] wherein, is the core temperature, is the current, is the resistance, is the thermal resistance of the insulation layer, is the thermal resistance of the inner lining layer, is the thermal resistance of the skin layer, is the thermal resistance of the environmental medium, is the loss coefficient of the armor layer, is the cable surface temperature.

[0021] In some embodiments, the fusion of the state feature quantities and the core temperature specifically comprises:

[0022] The state characteristic quantity and the core temperature are fused by a gray correlation degree method to analyze the correlation between the state characteristic quantity, the core temperature and the power cable state, and to determine the correlation degree between the state characteristic quantity, the core temperature and the power cable state.

[0023] According to the correlation degree, the state characteristic quantity and the core temperature are weighted and fused.

[0024] In some embodiments, inputting the fused state characteristic quantity and core temperature into the trained state evaluation model for processing specifically comprises:

[0025] The fused state characteristic quantity and core temperature are input into a convolution layer for convolution calculation to obtain a local feature matrix.

[0026] The local feature matrix is subjected to a pooling operation by a pooling layer to obtain a plurality of feature mapping matrices and compression, generate a feature matrix and input a plurality of groups of LSTM networks for processing to obtain a plurality of groups of prediction results.

[0027] The plurality of groups of prediction results are input into a CRF layer for processing to determine the final state evaluation result.

[0028] In some embodiments, the plurality of 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 taken as the final state evaluation result.

[0029] In some embodiments, the activation function of the convolution layer is a tanh function, and the activation function of the hidden layer in the LSTM network is a Relu activation function.

[0030] In a second aspect, the present application provides a power cable state evaluation system based on multi-source data collaboration.

[0031] A power cable state evaluation system based on multi-source data collaboration comprises:

[0032] An acquisition module is configured to acquire an electrical parameter set and a cable surface temperature of a power cable based on a preset time granularity.

[0033] A feature selection module is configured to perform feature selection on the electrical parameter set based on information gain to determine a state evaluation characteristic quantity, and acquire a core temperature of the power cable based on the cable surface temperature through a preset equivalent thermal circuit model and attribute parameters of the power cable.

[0034] A state evaluation module is configured to fuse the state evaluation characteristic quantity and the core temperature, and input the fused state evaluation characteristic quantity and core temperature into a trained state evaluation model for processing to obtain a state evaluation result of the power cable.

[0035] The state evaluation model comprises a convolution layer, a pooling layer, a plurality of groups of LSTM networks and a CRF module connected in sequence.

[0036] In a third aspect, the present application provides an electronic device;

[0037] An electronic device comprising a memory and a processor and computer instructions stored on the memory and running on the processor, when the computer instructions are run by the processor, the steps of the power cable state evaluation method based on multi-source data collaboration described above are completed.

[0038] In a fourth aspect, the present application provides a computer readable storage medium;

[0039] A computer readable storage medium for storing computer instructions, when the computer instructions are executed by a processor, the steps of the power cable state evaluation method based on multi-source data collaboration described above are completed.

[0040] Compared with the prior art, the present application has the beneficial effects that:

[0041] 1、The technical scheme provided by the present application uses the core temperature of the power cable and other features that are strongly related to the health status of the power cable to collaboratively evaluate the state of the power cable, which can comprehensively and truly reflect the health status of the power cable, and reduces the complexity of data analysis and the amount of data processing when evaluating the state, thereby improving the real-time performance and robustness of the state evaluation.

[0042] 2、The technical scheme provided by the present application combines the CNN network, the LSTM network and the CRF layer to run the monitoring, which can improve the model training speed and suppress overfitting, and when evaluating, the attention mechanism is used to give greater weight to important parameters, thereby improving the accuracy of the state evaluation. DETAILED DESCRIPTION

[0043] It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the present application. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0044] It should be noted that the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, the singular form is intended to include the plural form unless the context clearly indicates otherwise, and it should be understood that the terms "comprise" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device comprising a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to the process, method, product or device.

[0045] In the case of no conflict, the embodiments in the present application and the features in the embodiments can be combined with each other.

[0046] Embodiment one

[0047] The prior art cable operation monitoring uses a large amount of data and cannot fully reflect the real operation state of the power cable, and therefore the present application provides a power cable state evaluation method based on multi-source data cooperation.

[0048] Next, a power cable state evaluation method based on multi-source data cooperation disclosed in the present embodiment will be described in detail. The power cable state evaluation method based on multi-source data cooperation comprises the following steps:

[0049] S1, based on a preset time granularity, obtaining an electrical parameter set and a cable surface temperature of the power cable.

[0050] Temperature is one of the most intuitive manifestations of the cable state. Once the cable operation is abnormal, overheating phenomenon will occur, and the cable surface temperature is the most easily collected temperature information. Therefore, in the present embodiment, the cable surface temperature is used to calculate the subsequent battery temperature to pre-detect the abnormality of the cable state. The electrical parameters are the most accurate manifestation of the power cable state. Based on the above data, the power cable state is evaluated to improve the authenticity of the method described in the present embodiment.

[0051] In the present embodiment, the electrical parameters include voltage, current, partial discharge signal, power and other time series data.

[0052] S2, performing feature selection on the electrical parameter set based on information gain to determine state evaluation feature quantities;

[0053] During the operation of the power cable, a large amount of operation data is generated, which causes the data processing to take too long. Moreover, the operation data types are various, and the influence on the power cable health state evaluation is different, which is easy to interfere with the state evaluation and affect its robustness. Therefore, in the present embodiment, the information gain is used to select the electrical parameters in the electrical parameter set to filter out the electrical parameters that do not provide important information.

[0054] As an implementation manner, S2 specifically comprises:

[0055] S201, 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 electrical parameter set.

[0056] The overall information entropy of the electrical parameter set is expressed as follows:

[0057] ;

[0058] In the formula, denotes a set of electrical parameters, denotes a proportion of the i-th electrical parameter in the set of electrical parameters, is a total number of electrical parameter types.

[0059] The information entropy of the electrical parameters is represented as follows:

[0060] ;

[0061] In the formula, denotes a number of electrical parameters with a value of denotes a number of electrical parameters with a value of denotes a number of electrical parameters belonging to the k-th category, denotes an electrical parameter.

[0062] S202, determining the information gain of each electrical parameter according to the overall information entropy and the information entropy of each electrical parameter.

[0063] wherein the information gain is represented as follows:

[0064] .

[0065] S203, arranging the information gain in ascending order, and taking the first 30% of the electrical parameters as the state evaluation characteristic quantity.

[0066] S3, obtaining the attribute parameters and the cable surface temperature of the power cable, and determining the core temperature of the power cable according to the preset equivalent thermal circuit model of the power cable, the attribute parameters and the cable surface temperature.

[0067] The internal temperature of the power cable is monitored by a sensor, which requires modification of the structure of the power cable and cannot be applied to power cables that have been put into use. In addition, it is easily affected by the environment. Therefore, in the embodiment, the equivalent thermal circuit model of the power cable is used to calculate the core temperature, which has a small amount of data processing and is easy to implement.

[0068] As an implementation manner, S3 specifically includes:

[0069] S301, constructing an equivalent thermal circuit model of the power cable according to the structure of the power cable.

[0070] In the 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 represented as:

[0071] ;

[0072] wherein, is the core temperature, is the current, is the resistance, is the insulation layer thermal resistance, is the inner liner layer thermal resistance, is the skin layer thermal resistance, is the environmental medium thermal resistance, is the armor layer loss coefficient, is the cable surface temperature.

[0073] S302, according to the insulation layer thermal resistance coefficient, the insulation layer thickness and the insulation layer diameter, the insulation layer thermal resistance is determined; according to the inner liner layer thermal resistance coefficient, the inner liner layer thickness and the inner liner layer diameter, the inner liner layer thermal resistance is determined; according to the skin layer thermal resistance coefficient, the skin layer thickness and the skin layer diameter, the skin layer thermal resistance is determined; according to the soil thermal resistance coefficient, the cable buried depth and the outer diameter of the cable, the environmental medium thermal resistance is determined.

[0074] The insulation layer thermal resistance is represented as:

[0075] ;

[0076] In the formula, is the insulation layer thermal resistance, is the insulation layer thermal resistance coefficient, is the insulation layer thickness, is the insulation layer diameter.

[0077] The inner liner layer thermal resistance is represented as:

[0078] ;

[0079] In the formula, is the inner liner layer thermal resistance, is the inner liner layer thermal resistance coefficient, is the inner liner layer thickness, is the inner liner layer diameter.

[0080] The skin layer thermal resistance is represented as:

[0081] ;

[0082] In the formula, is the skin layer thermal resistance, is the skin layer thermal resistance coefficient, is the skin layer thickness, is the skin layer diameter.

[0083] The environmental medium thermal resistance is represented as:

[0084] ;

[0085] In the formula, is the soil thermal resistance coefficient, is the cable buried depth, is the outer diameter of the cable.

[0086] S303, the insulation layer thermal resistance, the inner liner thermal resistance, the skin layer thermal resistance and the environmental medium thermal resistance are brought into the equivalent thermal circuit model of the power cable, and the core temperature is determined, and is expressed as follows:

[0087] ;

[0088] wherein, is the core temperature, is the current, is the resistance, is the insulation layer thermal resistance, is the inner liner thermal resistance, is the skin layer thermal resistance, is the environmental medium thermal resistance, is the armor layer loss coefficient, is the cable surface temperature.

[0089] In this embodiment, the resistance and the armor layer loss coefficient are parameters provided by the power cable manufacturer, 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 is monitored to reflect the real state of the power cable, thereby providing accurate basic data for the health state assessment of the power cable.

[0090] S4, the state characteristic quantity and the core temperature are fused and input into the trained state assessment model for processing to obtain the state assessment result of the power cable.

[0091] In order to quantify the influence degree of different characteristics on the state of the power cable, the state characteristic quantity and the core temperature are fused before the state assessment. In order to improve the feature ability of the time series data, the state assessment model is improved, and the improved state assessment model includes convolution layers, pooling layers, multiple groups of LSTM networks and CRF modules connected in turn.

[0092] As an implementation manner, S4 specifically includes:

[0093] S401, the state characteristic quantity and the core temperature are analyzed for correlation with the state of the power cable by the grey correlation degree method, and the correlation degree thereof with the state of the power cable is determined; and the state characteristic quantity and the core temperature are weighted and fused according to the correlation degree. Wherein, the correlation degree is expressed as follows:

[0094] ;

[0095] In the formula, is the correlation coefficient, which is expressed as:

[0096] ;

[0097] In the formula, is a power cable state sequence, is a resolution coefficient, generally taking a value of 0.5, is a state feature quantity sequence or a conductor temperature sequence, represents a position in the sequence.

[0098] Finally, after normalization processing of the correlation degrees corresponding to the state feature quantity and the conductor temperature, the state feature quantity and the conductor temperature are weighted and fused as weights to obtain a feature data matrix.

[0099] S402, input the weighted and fused state feature quantity and conductor temperature into a trained state evaluation model for processing to obtain a state evaluation result of the power cable. Specifically, the method comprises the following steps.

[0100] S4021, input the feature data matrix into a convolution layer for convolution calculation to obtain a local feature matrix.

[0101] Specifically, the local feature matrix is represented as:

[0102] ;

[0103] In the formula, is a weight, is a convolution vector matrix obtained by convolution calculation on the fused state feature quantity and conductor temperature, is a bias matrix, is a rectified linear unit (ReLU) activation function.

[0104] S4022, perform a pooling operation on the convolution vector matrix through a pooling layer to obtain a plurality of feature mapping matrices and compression to generate a feature matrix, and the dimension of the feature matrix is the same as the dimension of the input feature data matrix.

[0105] In the formula,

[0106] ;

[0107] In the formula, is a feature mapping matrix, is the number of feature matrices.

[0108] S4023, input the feature matrix into a plurality of groups of LSTM networks respectively for processing to obtain a plurality of groups of prediction results.

[0109] In the embodiment, the LSTM network adopts an existing LSTM network, including a forgetting gate, an input gate, an output gate and a memory cell, and no improvement is made here, which will not be described herein.

[0110] wherein the prediction result is represented as

[0111] ;

[0112] ;

[0113] ;

[0114] ;

[0115] ;

[0116] wherein, is the output at time t, is the initial output obtained through a sigmoid layer, and Tanh is a hyperbolic tangent activation function, is the information update at time t, is the output obtained through a sigmoid layer, is the information update at time t-1, is a vector obtained by activating the elements in the vector i that need to be saved through a sigmoid function, is a new candidate value generated using a Tanh layer, is a sigmoid activation function, is the input at time t, is the output at time t-1, , , , , , denotes a weight matrix, , denotes a bias vector.

[0117] In the embodiment, the result output by each group of LSTM networks at time point t is a health condition score of the power cable at time point t. The weighted fused state characteristic quantity and the conductor temperature are obtained by processing existing data when the LSTM network is trained, and the health condition score is added to construct a training set, and the LSTM network is trained through the training set.

[0118] S4024, the post-input prediction results of multiple groups are input into a CRF layer for processing to obtain the correct probability of each group of prediction results, and the prediction result with the largest probability is taken as the final state evaluation result.

[0119] Exemplarily, the prediction result is represented as , that is, the input of the CRF layer, and the given label is represented as , then for the given input Y, the probability of the correct prediction result of each group is:

[0120] ;

[0121] ;

[0122] In the formula, denotes the probability of correct prediction, denotes the prediction result with the maximum probability, denotes the numerical maximum function.

[0123] Embodiment two

[0124] The embodiment discloses a power cable state evaluation system based on multi-source data collaboration, comprising:

[0125] An acquisition module is configured to acquire an electrical parameter set and a cable surface temperature of a power cable based on a preset time granularity.

[0126] A 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, and acquire a core temperature of the power cable based on the cable surface temperature through a preset equivalent thermal circuit model of the power cable and attribute parameters.

[0127] A state evaluation module is configured to fuse the state feature quantity and the core temperature and input a trained state evaluation model for processing to acquire a state evaluation result of the power cable.

[0128] The state evaluation model comprises a convolution layer, a pooling layer, a plurality of LSTM networks and a CRF module connected in sequence.

[0129] It should be noted that the above acquisition module, feature selection module and state evaluation module correspond to the steps in Embodiment One, and the above modules have the same examples and application scenarios as the corresponding steps, but are not limited to the content disclosed in Embodiment One. It should be noted that the above modules as part of the system can be executed in a computer system such as a set of computer executable instructions.

[0130] Embodiment three

[0131] The embodiment three of the present application provides an electronic device, comprising a memory and a processor, and computer instructions stored in the memory and running on the processor, when the computer instructions are run by the processor, the steps of the above power cable state evaluation method based on multi-source data collaboration are completed.

[0132] Embodiment four

[0133] Embodiment four of the present application provides a computer readable storage medium for storing computer instructions, which when executed by a processor, complete the steps of the power cable state evaluation method based on multi-source data collaboration described above.

[0134] The present application is described in reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device that implements the functions specified in one or more flows in the flowcharts and / or one or more blocks in the block diagrams.

[0135] These computer program instructions can also be stored in a computer readable memory that can direct the 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 product including instruction devices that implement the functions specified in one or more flows in the flowcharts and / or one or more blocks in the block diagrams.

[0136] These computer program instructions can also be loaded into a computer or other programmable data processing device to perform a series of operation steps 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 flows in the flowcharts and / or one or more blocks in the block diagrams.

[0137] The description of each embodiment in the above embodiments has its own emphasis, and the parts not described in detail in a certain embodiment can refer to the related description of other embodiments.

[0138] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A power cable state evaluation method based on multi-source data collaboration, characterized in that, The method comprises the following steps: Based on the preset time granularity, the electrical parameter set and the cable surface temperature of the power cable are obtained; Based on information gain, the electrical parameter set is selected for feature selection to determine the state evaluation feature quantity; Based on the cable surface temperature, the core temperature of the power cable is obtained through the preset equivalent thermal circuit model and attribute parameters of the power cable; The state feature quantity and the core temperature are fused and input into the trained state evaluation model for processing to obtain the state evaluation result of the power cable; The state evaluation model comprises a convolution layer, a pooling layer, a plurality of LSTM networks and a CRF module connected in sequence, the plurality of prediction results output by the plurality of LSTM networks are input into the CRF module for processing, the correct probability of each prediction result is obtained, and the prediction result with the maximum probability is taken as the final state evaluation result; the correct probability of each prediction result is: ； ； In the formula, denotes the probability of correct prediction, denotes the prediction result with the maximum probability, denotes a numerical maximization function; The feature selection based on information gain for the electrical parameters to determine the state evaluation feature quantity specifically comprises: 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 are calculated; 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, and the state evaluation feature quantity is screened; The core temperature of the power cable is represented as: ； wherein, is the conductor temperature, is the current, is the resistance, is the insulation layer thermal resistance, is the inner sheath thermal resistance, is the skin layer thermal resistance, is the ambient medium thermal resistance, is the armour layer loss coefficient, is the cable surface temperature; The thermal resistance of the insulation layer is represented as: ； wherein Rth is the thermal resistance of the insulation layer, Rth is the thermal resistance of the insulation layer, Rth is the thermal resistance of the insulation layer, Rth is the thermal resistance of the insulation layer, The thermal resistance of the inner lining layer is represented as: ； wherein Ri is the thermal resistance of the inner liner, Ri is the thermal resistance of the inner liner, Ri is the thermal resistance of the inner liner, Ri is the thermal resistance of the inner liner; The thermal resistance of the skin layer is represented as: ； wherein Rskin is the skin layer thermal resistance, Rskin is the skin layer thermal resistance coefficient, Rskin is the skin layer thickness, Rskin is the skin layer diameter; The thermal resistance of the environmental medium is represented as: ; wherein is the thermal resistance coefficient of the soil, is the depth of the cable, is the outer diameter of the cable; The fusion of the state feature quantity and the core temperature specifically comprises: The correlation between the state feature quantity, the core temperature and the state of the power cable is analyzed by the grey correlation degree method to determine the correlation degree of the state feature quantity and the core temperature with the state of the power cable; the state feature quantity and the core temperature are weighted and fused according to the correlation degree; the correlation degree is represented as follows: The correlation degree is represented as follows: ； wherein is a correlation coefficient and is expressed as: ； wherein is a power cable state sequence, is a resolution coefficient, and is set to 0.5, is a state feature quantity sequence or a conductor temperature sequence, denotes a position in the sequence; The input of the fused state feature quantity and the core temperature into the trained state evaluation model for processing specifically comprises: The fused state feature quantity and the core temperature are input into the convolution layer for convolution calculation to obtain a local feature matrix; The local feature matrix is subjected to a pooling operation by the pooling layer to obtain a plurality of feature mapping matrices which are compressed to generate a feature matrix and input into a plurality of LSTM networks for processing to obtain a plurality of prediction results; The plurality of prediction results are input into the CRF layer for processing to determine the final state evaluation result. The plurality of prediction results are input into the CRF layer for processing to determine the final state evaluation result.

2. The power cable condition assessment method based on multi-source data collaboration of claim 1, wherein, The activation function of the convolution layer is a tanh function, and the activation function of the hidden layer in the LSTM network is a Relu activation function.

3. A power cable condition assessment system based on multi-source data collaboration, adopting the power cable condition assessment method based on multi-source data collaboration according to any one of claims 1-2, characterized in that, The method comprises the following steps: The acquisition module is configured to obtain the electrical parameter set and the cable surface temperature of the power cable based on the preset time granularity; The feature selection module is configured to select the electrical parameter set for feature selection based on information gain to determine the state evaluation feature quantity; Based on the cable surface temperature, the core temperature of the power cable is obtained through the preset equivalent thermal circuit model and attribute parameters of the power cable; The state evaluation module is configured to fuse the state feature quantity and the conductor temperature, input the trained state evaluation model for processing, and obtain a state evaluation result of the power cable. The state evaluation model comprises convolution layers, pooling layers, multiple sets of LSTM networks, and a CRF module connected in sequence; the multiple sets of prediction results output by the multiple sets of LSTM networks are input into the CRF module for processing to obtain a correct probability of each set of prediction results, and the prediction result with the maximum probability is taken as the final state evaluation result; the correct probability of each set of prediction results is: ； ； In the formula, denotes probability of correct prediction, denotes the prediction result with the maximum probability, denotes a numerical maximization function; The feature selection of the electrical parameters based on the information gain comprises the following steps: According to the set of electrical parameters, the overall information entropy of the set of electrical parameters and the information entropy of each electrical parameter in the set of electrical parameters are calculated. 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, and the state evaluation feature quantity is screened. The conductor temperature of the power cable is represented as: ； wherein, is the core temperature, is the current, is the resistance, is the insulation layer thermal resistance, is the inner sheath thermal resistance, is the skin layer thermal resistance, is the ambient medium thermal resistance, is the armour layer loss coefficient, is the cable surface temperature; The thermal resistance of the insulation layer is represented as: ； wherein Rth is the thermal resistance of the insulation layer, Rth is the thermal resistance of the insulation layer, Rth is the thermal resistance of the insulation layer, Rth is the thermal resistance of the insulation layer, The thermal resistance of the inner lining layer is represented as: ； wherein Ri is the thermal resistance of the inner liner, Ri is the thermal resistance of the inner liner, Ri is the thermal resistance of the inner liner, Ri is the thermal resistance of the inner liner; The thermal resistance of the skin layer is represented as: ； wherein Rskin is the skin layer thermal resistance, Rskin is the skin layer thermal resistance coefficient, Rskin is the skin layer thickness, Rskin is the skin layer diameter; The thermal resistance of the environmental medium is represented as: ; wherein is the thermal resistance coefficient of the soil, is the cable burial depth, is the cable outer diameter; The fusion of the state feature quantity and the conductor temperature comprises the following steps: The correlation between the state feature quantity, the conductor temperature, and the state of the power cable is analyzed by the grey correlation degree method to determine the correlation degree of the state feature quantity and the conductor temperature with the state of the power cable; the state feature quantity and the conductor temperature are weighted and fused according to the correlation degree; the correlation degree is represented as follows: The correlation degree is represented as follows: ； In the formula, is a correlation coefficient and is expressed as: ； In the formula, is a power cable state sequence, is a resolution coefficient, and is 0.5, is a state feature quantity sequence or a conductor temperature sequence, denotes a position in the sequence; The input of the fused state feature quantity and the conductor temperature into the trained state evaluation model for processing comprises the following steps: The fused state feature quantity and the conductor temperature are input into the convolution layer for convolution calculation to obtain a local feature matrix; The local feature matrix is subjected to a pooling operation by the pooling layer to obtain multiple feature mapping matrices, which are compressed to generate a feature matrix and input into the multiple sets of LSTM networks for processing to obtain multiple sets of prediction results; The multiple sets of prediction results are input into the CRF layer for processing to determine the final state evaluation result. The input of the multiple sets of prediction results into the CRF layer for processing comprises the following steps: the multiple sets of prediction results are input into the CRF layer for processing to obtain a correct probability of each set of prediction results, and the prediction result with the maximum probability is taken as the final state evaluation result.

4. An electronic device, comprising: The computer instructions are stored in the memory and run on the processor, and when the computer instructions are run by the processor, the power cable state evaluation method based on multi-source data collaboration according to any one of claims 1-2 is completed.

5. A computer-readable storage medium, characterized in that, The computer instructions are stored in the memory and run on the processor, and when the computer instructions are run by the processor, the power cable state evaluation method based on multi-source data collaboration according to any one of claims 1-2 is completed.

Images