A method and system for monitoring abnormal vibration of a cable intermediate joint

Abstract

The present application relates to a kind of cable intermediate joint abnormal vibration monitoring method and system, wherein the method comprises the following steps: obtaining the historical multi-source data of sample cable intermediate joint, and historical multi-source data includes vibration signal, environmental data and equipment signal;The historical multi-source data is preprocessed, including decoupling to vibration signal, obtain vibration signal decoupling feature, and vibration signal and equipment signal are carried out spectrum analysis, obtain vibration spectrum feature and equipment spectrum feature;Based on the historical multi-source data after pre-processing, training sample set is constructed;Machine learning model is constructed, and machine learning model is iteratively trained using training sample set, and the trained cable intermediate joint abnormal vibration research model is obtained;The multi-source data of target cable intermediate joint is acquired in real time, and after pre-processing, input into the trained cable intermediate joint abnormal vibration research model, and the abnormal vibration monitoring result of target cable intermediate joint is obtained.

Description

TECHNICAL FIELD

[0001] The present application relates to a kind of cable intermediate joint abnormal vibration monitoring method and system, belong to power equipment abnormal vibration monitoring technical field. BACKGROUND

[0002] Power cable gradually replaces overhead line and is used in urban distribution network due to its characteristics of being buried underground, not occupying space, safety and the like. However, cable faults have concealment, which not only causes a large amount of power loss, but also seriously affects the normal work of electrical equipment, and light line work current increases, the service life of electrical equipment is shortened, and heavy production, scientific research, medical surgery and other activities in progress are suddenly interrupted, and fires and electric shock accidents are caused, causing incalculable losses. According to literature reports, in addition to external damage caused faults, cable accessory quality and aging problems account for more than 63% of all cable faults, and cable intermediate joint is an important part of cable accessories and a weak point of power cable, and when cable fault occurs, it is often accompanied by abnormal mechanical vibration. In order to timely analyze cable faults, it is necessary to accurately and comprehensively measure the abnormal vibration of cable intermediate joint.

[0003] The existing cable intermediate joint monitoring methods are: 1. Cable temperature, partial discharge online monitoring, dielectric loss online monitoring and the like. Such monitoring systems often focus on a certain single fault, and cannot realize comprehensive fault analysis of abnormal vibration. 2. Artificial power-off maintenance is carried out, and insulation test is carried out by cable fault tester. This method consumes a large amount of human resources. 3. In addition, the prior art lacks intelligent comprehensive analysis and judgment of abnormal vibration of cable intermediate joint, cannot accurately analyze the differences between abnormal conditions such as external damage risk and electromagnetic vibration and normal conditions such as road driving, and leads to low fault discrimination accuracy. SUMMARY

[0004] In order to solve the problems in the prior art, the present application provides a kind of cable intermediate joint abnormal vibration monitoring method and system.

[0005] The technical scheme of the present application is as follows:

[0006] On the one hand, the present application provides a kind of cable intermediate joint abnormal vibration monitoring method, comprising the following steps:

[0007] Obtain historical multi-source data of sample cable intermediate joint, and the historical multi-source data includes vibration signal, environmental data and device signal;

[0008] Preprocess the historical multi-source data, including decoupling the vibration signal to obtain vibration signal decoupling features, and performing frequency spectrum analysis on the vibration signal and the device signal to obtain vibration frequency spectrum features and device frequency spectrum features;

[0009] Based on the pretreated historical multi-source data, a training sample set is constructed, each sample in the sample set containing vibration signal decoupling features, vibration spectrum features, device spectrum features, environmental data and abnormal vibration labels of the sample cable intermediate joint;

[0010] A machine learning model is constructed, and the training sample set is used to iteratively train the machine learning model to obtain a trained cable intermediate joint abnormal vibration judgment model;

[0011] Real-time multi-source data of the target cable intermediate joint are acquired, pretreated and input into the trained cable intermediate joint abnormal vibration judgment model to obtain the abnormal vibration monitoring result of the target cable intermediate joint.

[0012] As a preferred embodiment, the step of decoupling the vibration signal to obtain the vibration signal decoupling features specifically comprises:

[0013] For the time series of the vibration signal, the sequence decomposition method is used to decompose it into a trend component, a periodic component and a random component, and the specific formula is as follows:

[0014] ;

[0015] Wherein, X 0 is the time series of the vibration signal, X 1 is the trend component, X 2 is the periodic component, X 3 is the random component;

[0016] Wherein, the trend component is extracted by the moving average method, and the specific formula is as follows:

[0017] ;

[0018] Wherein, f is the periodic frequency of the time series of the vibration signal, and t represents the time;

[0019] The periodic component is calculated by the average value under the same frequency, and the specific formula is as follows:

[0020] ;

[0021] ;

[0022] Wherein, represents the value of the i th periodic vibration signal minus the trend component, is the total number of periods, and , is the length of the time series of the vibration signal;

[0023] The random component is obtained by subtracting the trend component and the periodic component from the time series of the vibration signal.

[0024] As a preferred embodiment, the method of performing spectral analysis on the vibration signal and the equipment signal specifically comprises:

[0025] The vibration signal and the equipment signal are obtained, and the spectral curve of the vibration signal and the spectral curve of the equipment signal are obtained through Fourier decomposition or fast Fourier decomposition.

[0026] The spectral curve of the vibration signal and the spectral curve of the equipment signal are subjected to feature extraction, and vibration spectral features and equipment spectral features are obtained.

[0027] As a preferred embodiment, the machine learning model adopts a linear regression model, a decision tree model, a support vector machine model, or an LSTM neural network model.

[0028] In another aspect, the present application also provides a cable intermediate joint abnormal vibration monitoring system, comprising: a host computer and an environment monitoring module, a cable intermediate joint monitoring module, a multi-source data aggregation module and a research and judgment analysis module arranged in the host computer;

[0029] The environment monitoring module and the cable intermediate joint monitoring module are used to obtain historical multi-source data of a sample cable intermediate joint, and the historical multi-source data comprises a vibration signal, environmental data and an equipment signal.

[0030] The multi-source data aggregation module is used to pre-process the historical multi-source data, including decoupling the vibration signal to obtain vibration signal decoupling features, and performing spectral analysis on the vibration signal and the equipment signal to obtain vibration spectral features and equipment spectral features.

[0031] The research and judgment analysis module is based on the pre-processed historical multi-source data to construct a training sample set, each sample in the sample set contains vibration signal decoupling features, vibration spectral features, equipment spectral features, environmental data and an abnormal vibration label of a sample cable intermediate joint, and a machine learning model is constructed, the training sample set is used to iteratively train the machine learning model, and a trained cable intermediate joint abnormal vibration research and judgment model is obtained.

[0032] The environment monitoring module and the cable intermediate joint monitoring module obtain multi-source data of a target cable intermediate joint in real time, input the multi-source data to the multi-source data aggregation module for pre-processing, and then input the pre-processed multi-source data to the trained cable intermediate joint abnormal vibration research and judgment model in the research and judgment analysis module to obtain an abnormal vibration monitoring result of the target cable intermediate joint.

[0033] As a preferred embodiment, the step of decoupling the vibration signal to obtain vibration signal decoupling features specifically comprises:

[0034] For the time series of the vibration signal, the sequence decomposition method is used to decompose into a trend component, a periodic component and a random component, and the specific formula is as follows:

[0035]

[0036] wherein, X 0 is the time series of the vibration signal, X 1 is the trend component, X 2 is the periodic component, X 3 is the random component;

[0037] The trend component is extracted by using the moving average method, and the specific formula is as follows:

[0038]

[0039] wherein, f is the periodic frequency of the time series of the vibration signal, and t represents the time;

[0040] The periodic component is calculated by using the average value under the same frequency, and the specific formula is as follows:

[0041]

[0042]

[0043] wherein, represents the value of the i th periodic vibration signal minus the trend component, is the total number of periods, and , is the length of the time series of the vibration signal;

[0044] The random component is obtained by subtracting the trend component and the periodic component from the time series of the vibration signal.

[0045] As a preferred embodiment, the method for performing spectrum analysis on the vibration signal and the device signal by the multi-source data aggregation module is specifically as follows:

[0046] The vibration signal and the device signal are obtained, and the frequency spectrum curve of the vibration signal and the frequency spectrum curve of the device signal are obtained by Fourier decomposition or fast Fourier decomposition;

[0047] The frequency spectrum curve of the vibration signal and the frequency spectrum curve of the device signal are subjected to feature extraction, and vibration spectrum features and device spectrum features are obtained.

[0048] As a preferred embodiment, in the research and judgment analysis module, the machine learning model adopts a linear regression model, a decision tree model, a support vector machine model or an LSTM neural network model.

[0049] ​​​​In still another aspect, the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor implements the method for monitoring abnormal vibration of cable intermediate joint as described in any of the embodiments of the present application when executing the program.

[0050] In still another aspect, the present application provides a computer readable storage medium, having stored thereon a computer program, which, when executed by a processor, implements the method for monitoring abnormal vibration of cable intermediate joint as described in any of the embodiments of the present application.

[0051] Additional aspects and advantages of the present application will be set forth in the description that follows, and in part will be obvious from the description, or can be learned by practice of the present application. Moreover, the advantages of the present application can be realized and obtained by means of the instrumentalities and combinations particularly pointed out in the appended claims. BRIEF DESCRIPTION OF DRAWINGS

[0052] Figure 1 a flow chart of the method according to the first embodiment of the present application;

[0053] Figure 2 a schematic diagram of the system according to the second embodiment of the present application;

[0054] Figure 3 a schematic diagram of the internal circuit module of the host according to the second embodiment of the present application.

[0055] In the drawings:

[0056] 11, central processing unit; 12, A / D conversion circuit; 13, communication module; 14, battery; 15, power taking module; 151, I / V conversion module; 152, power taking coil; 21, high frequency sensor; 22, water level sensor; 23, temperature sensor; 31, temperature and humidity sensor; 32, vibration sensor; 33, gas sensor. DETAILED DESCRIPTION

[0057] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the present application.

[0058] It should be understood that the step numbers used herein are only for the convenience of description, and are not limited to the execution sequence of the steps.

[0059] It is to be understood that the terminology used in the description herein is for the purpose of describing the particular embodiments only and is not intended to be limiting. As used in the description and the appended claims, the singular forms "a," "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.

[0060] The terms "including" and "comprising" as used herein specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0061] The term "and / or" means any combination of one or more of the associated listed items and all possible combinations thereof, and includes these combinations.

[0062] Embodiment One:

[0063] Referring to Figure 1 The embodiment provides a cable intermediate joint abnormal vibration monitoring method, and specifically comprises the following steps:

[0064] S100 obtains historical multi-source data of a sample cable intermediate joint, the historical multi-source data comprising vibration signals, environmental data and device signals; wherein the vibration signal acquisition function mainly uses an XC-3810P vibration sensor (or other sensors with equivalent performance), uses a piezoelectric quartz crystal or a piezoelectric ceramic (such as zirconium titanate lead) as a sensitive element, and has the characteristic of generating an electric charge when subjected to mechanical stress. When the sensor is subjected to force caused by acceleration, the piezoelectric material generates an electric charge change, which is converted into a voltage signal through a circuit, the frequency response amplitude measurement error value is ≤0.2 Hz, and the acquisition distance is ≥6 m; the environmental data acquisition function mainly collects environmental temperature and humidity, gas content, water level around the sensor and the like collected by independent sensors; and the device signal acquisition function is mainly responsible for integrating signals of sensors wound on the surface of the cable, and specifically comprises cable surface temperature signals, high-frequency electromagnetic signals and other device-related signals, in addition, the Beidou or GPS positioning is combined to mark the specific geographic location of the cable.

[0065] S200, pre-process the historical multi-source data, including decoupling the vibration signal, the vibration signal decoupling is mainly aimed at the vibration signal time series data, by decoupling the trend, periodicity and occasional factors of the data sequence, to disassemble the vibration under normal level, so as to obtain the vibration signal decoupling feature. And the vibration signal and the device signal are analyzed by spectrum, the vibration spectrum feature and the device spectrum feature are obtained, the spectrum analysis is mainly aimed at the high frequency electromagnetic signal and the vibration signal, and the spectrum curve of the corresponding signal is obtained by the general method such as Fourier decomposition and fast Fourier decomposition. In the daily normal operation of the device, the spectrum information in the normal operation will be formed, when the device abnormally vibrates or fails, the spectrum will change accordingly, therefore, the state of the cable intermediate joint can be better reflected by the spectrum feature.

[0066] S300, based on the pre-processed historical multi-source data, a training sample set is constructed, each sample in the sample set contains vibration signal decoupling feature, vibration spectrum feature, device spectrum feature, environment data and abnormal vibration label of sample cable intermediate joint (i.e. label of whether sample cable intermediate joint occurs abnormal vibration at corresponding historical time). The feature data in each sample is constructed to form a structured table, each row of the table is a data sample of a specific time and a specific position, and each column is a specific data value, such as vibration trend slope, cable surface temperature, etc.

[0067] S400, a machine learning model is constructed, and the machine learning model is iteratively trained by using the training sample set, to obtain a trained cable intermediate joint abnormal vibration judgment model;

[0068] S500, real-time acquisition of multi-source data of target cable intermediate joint, pre-processing and inputting into the trained cable intermediate joint abnormal vibration judgment model, to obtain abnormal vibration monitoring result of the target cable intermediate joint. The abnormal vibration monitoring result is sent to the staff through system alarm, short message reminder and other ways, the system alarm is pushed through mobile phone APP, computer website and other ways in real time, the location information of the fault point is pushed, the short message reminder is the network short message function, and the notification short message is sent in time, so that the cable operation and maintenance personnel can obtain the fault condition in time and respond quickly.

[0069] As a preferred embodiment of the embodiment, in step S200, the step of decoupling the vibration signal to obtain the vibration signal decoupling feature specifically includes:

[0070] For vibration signal time series{ x t}= X 0, the sequence decomposition method is used to decompose into trend component, periodic component and random component, and the specific formula is as follows:

[0071] ;

[0072] wherein, X 0 is a time series of the vibration signal, X 1 is a trend component, X 2 is a periodic component, X 3 is a random component;

[0073] wherein, the trend component is extracted by using the method of moving average, and the specific formula is as follows:

[0074] ;

[0075] wherein, f is a periodic frequency of the time series of the vibration signal, and the vibration signal is a power frequency of 50 Hz, and t represents a time;

[0076] The periodic component is calculated by using the average value under the same frequency, and the specific formula is as follows:

[0077] ;

[0078] ;

[0079] wherein, represents a value of the i th periodic vibration signal minus the trend component, is a total number of periods, and is obtained by performing division operation on the periodic frequency by the actual length of the time series and then rounding, that is, , is a length of the time series of the vibration signal;

[0080] The random component is obtained by subtracting the trend component and the periodic component from the time series of the vibration signal, that is:

[0081] .

[0082] After obtaining the trend component X 1, the trend change slope can be obtained by curve fitting in a linear regression manner; for the periodic component X 2, the trend change slope and the peak-valley difference of the equipment cycle can be analyzed, and the trend change slope and the peak-valley difference of the equipment cycle are taken as the decoupling features of the vibration signal.

[0083] In another embodiment, the trend change slope and the peak-valley difference of the equipment cycle can also be directly used for equipment fault analysis. When the trend change slope is greater than a specific threshold, it can be determined that there is a risk of chronic equipment failure, so as to issue an alarm signal; when the peak-valley difference of the equipment cycle continuously increases and is greater than a specific threshold, it can be determined that there is a risk of acute equipment failure, so as to issue an alarm signal.

[0084] As a preferred embodiment of the present embodiment, the method of performing spectrum analysis on the vibration signal and the equipment signal is specifically:

[0085] The vibration signal and the equipment signal are obtained, and the spectrum curve of the vibration signal and the spectrum curve of the equipment signal are obtained through Fourier decomposition or fast Fourier decomposition;

[0086] The spectrum curve of the vibration signal and the spectrum curve of the equipment signal are subjected to feature extraction, and vibration spectrum features and equipment spectrum features are obtained.

[0087] As a preferred embodiment of the present embodiment, the machine learning model adopts a linear regression model, a decision tree model, a support vector machine model or an LSTM neural network model. The data required for input of the machine learning model includes four parameters, i.e., hyperparameters, training samples, test samples and samples to be predicted. The hyperparameters are parameters that need to be set artificially in model training, for example, the hyperparameters of a tree model include the number of trees, the depth of the tree, the minimum number of samples for dividing nodes, the minimum number of samples for result nodes, etc.; the hyperparameters of an LSTM neural network model include the number of neurons, the depth of the network, the learning rate, etc. The training of the model is divided into a grid search part, a model training part and a test part. The working modes among the three parts are as follows: first, the grid search part performs grid search and cross-validation through given small initial hyperparameters and training samples to obtain the optimal hyperparameter combination; second, the model training part takes the optimal hyperparameter combination as input, combines extreme random tree algorithms, etc. to train the training samples into the cable intermediate joint abnormal vibration judgment model, and then verifies the accuracy of the model by inputting test samples. If the model does not reach the required accuracy, the process of the first two parts is repeated, and the operation is repeated until the model reaches the required accuracy.

[0088] The embodiment is based on vibration signal monitoring, integrates environmental data monitoring and equipment signal monitoring, realizes global perception, and provides strong support for comprehensive research and judgment of vibration abnormalities. The device covers key information of cable intermediate joints, realizes comprehensive monitoring of various potential fault factors of cable joints, improves the accuracy and comprehensiveness of vibration abnormality fault warning, and effectively reduces the fault risk caused by single factor, accidental factor or insufficient monitoring. Based on machine learning research and judgment, the multi-dimensional data is comprehensively analyzed, the abnormal vibration fault point is accurately found, and the data is transmitted in real time; when the temperature or external damage state of the cable intermediate joint exceeds the set threshold, the system can accurately locate the position of the fault intermediate joint within 32s, timely control the abnormal vibration defect, quickly convey the information to the operation and maintenance personnel through the graphic alarm, and make them respond to the fault in time. At present, 86 cases of over-temperature fault and 6 cases of line damage warning have been successfully found, such as quickly positioning and processing the cable intermediate head interphase insulation breakdown short circuit fault of a power supply company on May 20, 2024, reducing power loss, timely warning of cable over-temperature of a cable with a length of 8 kilometers in a bustling commercial street, avoiding nearly ten million losses to the local economy due to sudden power failure caused by cable breakdown.

[0089] Embodiment two:

[0090] The embodiment provides a cable intermediate joint abnormal vibration monitoring system, specifically referring to Figure 2 , comprising a host and an environment monitoring module, a cable intermediate joint monitoring module, a multi-source data aggregation module and a research and judgment analysis module arranged in the host; the host is fixed through a fixed support, the environment monitoring module comprises a water level sensor 22, a gas sensor 33, a temperature and humidity sensor 31 and the like; the cable intermediate joint monitoring module comprises a temperature sensor 23, a high-frequency sensor 21 and a vibration sensor 32 arranged on a cable intermediate joint clamping ring, and the data collected by the sensors is transmitted to the host through a data connection line.

[0091] The circuit module in the host is as follows Figure 3As shown, a central processor 11 is provided, as well as an A / D conversion circuit 12, a communication module 13 and a battery 14 connected thereto, the battery 14 is connected with a power taking module 15, the input end of the A / D conversion circuit 12 is directly connected with the environment monitoring part and the cable joint monitoring part, receives analog signals and converts them into digital signals and then outputs to the central processor 11, the central processor 11 processes the digital signals and then uploads to the remote monitoring device through the communication module 13, the battery 14 provides power for the central processor 11 and charges through the power taking module 15. The communication module 13 is an NB-IOT communication module, NB-IOT is built on a cellular network, only consumes about 180KHz of bandwidth, can be directly deployed in GSM network, UMTS network or LTE network; supports long standby time, efficient connection of devices with high requirements for network connection, the NB-IOT communication module has low power consumption, can improve the service life of the battery, and can also provide very comprehensive indoor cellular data connection coverage. The shell of the host is provided with a fixing device, which is fixed on the cable trench wall through the fixing piece and the fixing hole. The input end of the A / D conversion circuit 12 receives analog signals collected by the temperature sensor 23, the vibration sensor 32 and the like in the cable joint monitoring part, converts them into digital signals and then outputs to the processor, the processor analyzes and judges the digital signals and then uploads to the remote monitoring background through the communication module, the communication adopts an NB-IOT low-power Internet of Things communication module. The average power consumption is 2uA, which can be used stably for 10 years and is maintenance-free. The power taking module 15 includes an I / V conversion module 151 and a power taking coil 152.

[0092] The embodiment applies NB-IoT Internet of Things communication mode, signal is stable, and data concentrator is not needed, so that collected data is directly transmitted to the background. This communication mode has the advantages of low power consumption, wide coverage, and large connection, overcomes the problems of weak communication signal and high signal attenuation rate in the underwater cable trench, realizes wide-area coverage of the alarm signal, ensures the timeliness and stability of data transmission, and effectively solves the problems of unstable data transmission and limited transmission distance of the traditional communication mode in the cable monitoring scene. The low-power-consumption design of the non-rechargeable battery power supply is adopted, the average power consumption is only 2uA, the non-rechargeable battery power supply is adopted without external power supply. The rated voltage is DC 3.6V, the capacity is 4000mAH, and the static power consumption is less than or equal to 30uA. This low-power-consumption design greatly prolongs the service life of the device, reduces the maintenance cost, and also improves the adaptability of the device in complex environments, such as stable operation in places such as cable trenches where external power supply is not easy to obtain. The device has been applied in a certain urban area for three months, and the proportion of tripping power failure accidents caused by abnormal vibration of the cable intermediate joint has been reduced to 22.22% (target value ≤ 30%), effectively reducing the fault tripping accident rate. The device is small in size, easy to install, does not occupy the cable corridor space, supports on-site installation without power failure, can be installed by a single person under voltage, can be connected to the background through the WeChat applet, does not need to be powered off, and the average power consumption is 2uA. The battery does not need to be replaced for 10 years. This makes it possible to complete the installation and deployment of the device without affecting the normal operation of the cable, reduces the influence on the power supply, improves the installation efficiency, and reduces the installation cost. The cumulative sales in a certain province exceed 4 million yuan, saving a large amount of 10kV line manual maintenance cost and equipment loss, reducing operation and maintenance, repair cost and labor intensity, and improving the management level of the power grid.

[0093] In order to facilitate installation, the hardware circuit board of the host is produced automatically by SMT welding process, small and modular production and assembly process is adopted to improve production efficiency, and through multi-layer buffer packaging process, the device can work normally and stably in multiple environmental conditions such as severe cold, high temperature and water immersion. In terms of external structure, the host shell is made of PC flame-retardant material, the flame-retardant fire resistance level reaches A0 level, the shell is packaged by pressing plate type + embedded silica gel gasket, and the protection level reaches IP67.

[0094] The environmental monitoring module and the cable intermediate joint monitoring module are used to obtain historical multi-source data of the sample cable intermediate joint, and the historical multi-source data includes vibration signals, environmental data and device signals. The module is used to realize the function of step S100 in embodiment one, and details are not repeated here.

[0095] The multi-source data aggregation module is used for pre-processing the historical multi-source data, including decoupling the vibration signals to obtain vibration signal decoupling features, and performing frequency spectrum analysis on the vibration signals and the device signals to obtain vibration frequency spectrum features and device frequency spectrum features. The module is used to realize the function of step S200 in embodiment one, and details are not repeated here.

[0096] The research and analysis module is based on the pre-processed historical multi-source data, constructs a training sample set, each sample in the sample set contains the vibration signal decoupling feature, vibration spectrum feature, device spectrum feature, environment data and abnormal vibration label of the sample cable intermediate joint, and constructs a machine learning model, iteratively trains the machine learning model using the training sample set, and obtains the trained cable intermediate joint abnormal vibration research and judgment model; this module is used to realize the functions of steps S300 and S400 in embodiment one, and will not be described here;

[0097] The environment monitoring module and the cable intermediate joint monitoring module acquire the multi-source data of the target cable intermediate joint in real time, input them to the multi-source data aggregation module for pre-processing, and then input them to the trained cable intermediate joint abnormal vibration research and judgment model in the research and analysis module to obtain the abnormal vibration monitoring result of the target cable intermediate joint; this module is used to realize the function of step S500 in embodiment one, and will not be described here.

[0098] It also includes an alarm and positioning module, which includes system alarm, SMS reminder and other functions. The system alarm pushes the fault information and the location information of the fault point in real time through mobile phone APP, computer website, etc., and the SMS reminder calls the network SMS function to send notification SMS in time, so that the cable operation and maintenance personnel can obtain the fault condition in time and respond quickly.

[0099] The cable intermediate joint abnormal vibration monitoring system provided by the embodiment has comprehensive functions and realizes global perception of cable intermediate joint information. The invention uses the vibration module to locate the region of the breakdown damage fault of the cable line, combines the temperature monitoring module to monitor the real-time temperature of the cable intermediate joint, simultaneously uses the differential operation of double temperature data to realize the over-temperature alarm, and realizes the early warning of external damage of the cable line through the superposition algorithm of device signal monitoring + vibration signal monitoring. A set of device can comprehensively cover the key information related to the abnormal vibration of the cable intermediate joint.

[0100] It is widely applicable and suitable for various harsh operating environments. It is suitable for 10kV-1000kV cable intermediate joints, covering multiple voltage level ranges from power distribution to power transmission. The device body realizes IP67 level waterproof and dustproof, and simultaneously realizes A0 level flame-retardant and fire-resistant performance. It can ensure the normal and stable operation of the device in multiple environmental conditions such as severe cold, high temperature and water immersion. In terms of communication, it uses unique NB-IoT narrowband Internet of Things communication technology, does not require a wireless terminal, is not affected by the operating environment, and can be used normally in places where the cable communication signal is weak and GPRS signal cannot collect data transmission. Even in a water depth of 1 meter underground cable trench, it can directly send data transmission to the background.

[0101] Convenient to use, low cost of installation, operation and maintenance. The product is easy to install, and single live installation can be realized. The background access can be realized by using WeChat applet. Compared with similar products, there is no need to stop power installation, and the installation is simple and fast. The low-power design is adopted, the average power consumption is 2uA, and the product can be used stably for 10 years without replacing the battery, which greatly reduces the maintenance cost. The perfect software system can make the potential fault information sent to the operation and maintenance personnel in time, eliminate the fault, reduce the loss caused by power failure to life and production, and make the false alarm rate of the software below 0.1%.

[0102] As a preferred embodiment of the present embodiment, the multi-source data aggregation module decouples the vibration signal, and the step of obtaining the vibration signal decoupling feature specifically includes:

[0103] For the time series of the vibration signal, the sequence decomposition method is used for decomposition into a trend component, a periodic component and a random component, and the specific formula is as follows:

[0104] ;

[0105] Wherein, X 0 is the time series of the vibration signal, X 1 is the trend component, X 2 is the periodic component, X 3 is the random component;

[0106] Wherein, the trend component is extracted by using the moving average method, and the specific formula is as follows:

[0107] ;

[0108] Wherein, f is the period frequency of the time series of the vibration signal, and t represents the time;

[0109] The periodic component is calculated by using the average value under the same frequency, and the specific formula is as follows:

[0110] ;

[0111] ;

[0112] Wherein, represents the value of the i th periodic vibration signal minus the trend component, is the total number of periods, and , is the length of the time series of the vibration signal;

[0113] The random component is obtained by subtracting the trend component and the periodic component from the time series of the vibration signal.

[0114] As a preferred embodiment of the present embodiment, the method for performing spectrum analysis on the vibration signal and the equipment signal by the multi-source data aggregation module is specifically as follows:

[0115] Obtaining the vibration signal and the equipment signal, and obtaining the spectrum curve of the vibration signal and the spectrum curve of the equipment signal through Fourier decomposition or fast Fourier decomposition;

[0116] Extracting features from the spectrum curve of the vibration signal and the spectrum curve of the equipment signal to obtain vibration spectrum features and equipment spectrum features.

[0117] As a preferred embodiment of the present embodiment, the machine learning model in the research and analysis module adopts a linear regression model, a decision tree model, a support vector machine model or an LSTM neural network model.

[0118] Embodiment three:

[0119] The present embodiment provides an electronic device, which includes a memory, a processor and a computer program stored in the memory and executable on the processor, and the processor implements the method according to any one of the embodiments of the present application when executing the program.

[0120] Embodiment four:

[0121] The present embodiment provides a computer readable storage medium, which stores a computer program, and the program is executable on a processor to implement the method according to any one of the embodiments of the present application.

[0122] In the embodiments of the present application, "at least one" means one or more, and "multiple" means two or more. "And / or" describes the association relationship of the associated objects, which means that there can be three kinds of relationships, for example, A and / or B, which means that A exists alone, A and B exist together, and B exists alone. Wherein A and B can be singular or plural. The character " / " generally represents an "or" relationship between the associated objects before and after it. "At least one of the following" and the like means any combination of these items, including any combination of single or multiple items. For example, at least one of a, b and c can mean: a, b, c, a and b, a and c, b and c, or a and b and c, wherein a, b and c can be single or multiple.

[0123] Those of ordinary skill in the art can realize that the units and algorithm steps described in the embodiments disclosed herein can be realized by electronic hardware, computer software and a combination of electronic hardware and computer software. Whether the functions are realized in hardware or software mode depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0124] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the system, device and unit described above can refer to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0125] In several embodiments provided in the present application, any function realized in the form of a software function unit and sold or used as an independent product can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the parts of the technical solutions that make contributions to the prior art or the parts of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the embodiments of the present application. The foregoing storage medium includes a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0126] The above is only an embodiment of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent flow transformation obtained by using the content of the specification and drawings, or direct or indirect application in other related technical fields, is also included in the patent protection scope of the present application.

Claims

1. A method for monitoring abnormal vibration of a cable joint, characterized in that, Includes the following steps: Historical multi-source data of the sample cable intermediate joint was obtained. The historical multi-source data included vibration signals, environmental data, and equipment signals. Among them, the environmental data included ambient temperature and humidity, gas content, and water level around the sensor, and the equipment signals included cable surface temperature signal and high-frequency electromagnetic signal. Historical multi-source data is preprocessed, including decoupling vibration signals to obtain vibration signal decoupling characteristics, and performing spectral analysis on vibration signals and equipment signals to obtain vibration spectral characteristics and equipment spectral characteristics. Based on preprocessed historical multi-source data, a training sample set is constructed. Each sample in the sample set contains the vibration signal decoupling characteristics, vibration spectrum characteristics, equipment spectrum characteristics, environmental data, and abnormal vibration labels of the sample cable intermediate joint. A machine learning model is constructed, and the machine learning model is iteratively trained using a training sample set to obtain a well-trained abnormal vibration judgment model for cable intermediate joints. The multi-source data of the target cable intermediate joint is acquired in real time, preprocessed, and then input into the trained abnormal vibration judgment model of the cable intermediate joint to obtain the abnormal vibration monitoring results of the target cable intermediate joint. The step of decoupling the vibration signal and obtaining its decoupling characteristics specifically includes: For a time series of vibration signals, the sequence decomposition method is used to decompose it into trend components, periodic components, and random components. The specific formula is as follows: ； in, X 0 represents the time series of the vibration signal. X 1 represents the trend component. X 2 is a periodic component. X 3 represents a random component; The trend component is extracted using a moving average method, with the specific formula as follows: ； in, f Let be the periodic frequency of the time series of the vibration signal, and t represent time. The periodic components are calculated using the average value at the same frequency, as shown in the following formula: ； ； in, This represents the value of the i-th period vibration signal minus the trend component. The total number of periods, and , The time series length of the vibration signal; The random component is obtained by subtracting the trend component and the periodic component from the time series of the vibration signal; The vibration signal decoupling characteristics include the trend change slope obtained by curve fitting of the trend component through linear regression, and the peak-valley difference of the equipment cycle obtained by analyzing the peak-valley difference of the periodic component before and after the periodic component.

2. The method for monitoring abnormal vibration of a cable joint as described in claim 1, characterized in that, The method for performing spectral analysis on vibration signals and equipment signals is as follows: Acquire vibration signals and equipment signals, and obtain the frequency spectrum curves of the vibration signals and equipment signals through Fourier decomposition or fast Fourier decomposition. Feature extraction is performed on the spectral curves of vibration signals and equipment signals to obtain vibration spectral features and equipment spectral features.

3. The method for monitoring abnormal vibration of a cable joint as described in claim 1, characterized in that, The machine learning model used is a linear regression model, a decision tree model, a support vector machine model, or an LSTM neural network model.

4. A cable joint abnormal vibration monitoring system, characterized in that, include: The host computer, along with an environmental monitoring module, a cable joint monitoring module, a multi-source data aggregation module, and an analysis module, all housed within the host computer. The environmental monitoring module and the cable joint monitoring module are used to acquire historical multi-source data of the sample cable joints. The historical multi-source data includes vibration signals, environmental data and equipment signals. Among them, the environmental data includes ambient temperature and humidity, gas content and water level around the sensor, and the equipment signals include cable surface temperature signal and high-frequency electromagnetic signal. The multi-source data aggregation module is used to preprocess historical multi-source data, including decoupling vibration signals to obtain vibration signal decoupling characteristics, and performing spectral analysis on vibration signals and equipment signals to obtain vibration spectral characteristics and equipment spectral characteristics. The analysis module constructs a training sample set based on preprocessed historical multi-source data. Each sample in the sample set includes the vibration signal decoupling characteristics, vibration spectrum characteristics, equipment spectrum characteristics, environmental data, and abnormal vibration labels of the sample cable joint. A machine learning model is then constructed, and the training sample set is used to iteratively train the machine learning model to obtain a trained abnormal vibration analysis model for the cable joint. The environmental monitoring module and the cable intermediate joint monitoring module acquire multi-source data of the target cable intermediate joint in real time. After being preprocessed by the multi-source data aggregation module, the data is then input into the trained abnormal vibration judgment model of the cable intermediate joint in the judgment and analysis module to obtain the abnormal vibration monitoring results of the target cable intermediate joint. The step of the multi-source data aggregation module decoupling the vibration signal and obtaining the decoupling characteristics of the vibration signal specifically includes: For a time series of vibration signals, the sequence decomposition method is used to decompose it into trend components, periodic components, and random components. The specific formula is as follows: ； in, X 0 represents the time series of the vibration signal. X 1 represents the trend component. X 2 is a periodic component. X 3 represents a random component; The trend component is extracted using a moving average method, with the specific formula as follows: ； in, f Let be the periodic frequency of the time series of the vibration signal, and t represent time. The periodic components are calculated using the average value at the same frequency, as shown in the following formula: ； ； in, This represents the value of the i-th period vibration signal minus the trend component. The total number of periods, and , The time series length of the vibration signal; The random component is obtained by subtracting the trend component and the periodic component from the time series of the vibration signal; The vibration signal decoupling characteristics include the trend change slope obtained by curve fitting of the trend component through linear regression, and the peak-valley difference of the equipment cycle obtained by analyzing the peak-valley difference of the periodic component before and after the periodic component.

5. The abnormal vibration monitoring system for cable joints according to claim 4, characterized in that, The method by which the multi-source data aggregation module performs spectral analysis on vibration signals and equipment signals is as follows: Acquire vibration signals and equipment signals, and obtain the frequency spectrum curves of the vibration signals and equipment signals through Fourier decomposition or fast Fourier decomposition. Feature extraction is performed on the spectral curves of vibration signals and equipment signals to obtain vibration spectral features and equipment spectral features.

6. The abnormal vibration monitoring system for cable joints according to claim 4, characterized in that, In the analysis module, the machine learning model adopts a linear regression model, a decision tree model, a support vector machine model, or an LSTM neural network model.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the abnormal vibration monitoring method for cable intermediate joints as described in any one of claims 1 to 3.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the abnormal vibration monitoring method for cable intermediate joints as described in any one of claims 1 to 3.

Images