阅读说明：本技术 一种基于边缘计算的电力电缆监测分析系统及其方法 (Power cable monitoring and analyzing system and method based on edge calculation ) 是由 任广振 姜文东 曹俊平 李乃一 黄勃 王成珠 夏雯 倪晓璐 陶倩敏 郭能俊 于 2021-07-13 设计创作，主要内容包括：本发明公开了一种基于边缘计算的电力电缆监测分析系统及其方法,属于电力技术领域,包括监测模块全面监测电力电缆运行数据,并将数据分别发送至前处理器；前处理器用于对监测模块监测的数据进行分类、清洗和标准化,然后将数据通过统一接口发送至中央处理器；机器学习模块用于根据前处理器处理后的数据以及云数据库的大数据建立电力电缆运行模型,并通过机器学习优化模型,得到电力电缆最有运行数据标准；分析模块用于将电力电缆实时运行数据与标准模型数据进行对比分析,判断电力电缆故障隐患并对其进行定位,便于及时查询,以及在确认故障后判断故障类型并发出警报。(The invention discloses a power cable monitoring and analyzing system and a method thereof based on edge calculation, belonging to the technical field of electric power, comprising a monitoring module, a pre-processor and a monitoring module, wherein the monitoring module comprehensively monitors power cable operation data and respectively sends the data to the pre-processor; the preprocessor is used for classifying, cleaning and standardizing the data monitored by the monitoring module and then sending the data to the central processor through the unified interface; the machine learning module is used for establishing a power cable operation model according to the data processed by the pre-processor and the big data of the cloud database, and obtaining the most available operation data standard of the power cable through a machine learning optimization model; the analysis module is used for comparing and analyzing the real-time operation data of the power cable with the standard model data, judging the hidden trouble of the power cable and positioning the hidden trouble of the power cable, facilitating the timely inquiry, judging the fault type after confirming the fault and giving an alarm.)

1. An edge-computing-based power cable monitoring and analysis system, comprising:

the monitoring module comprehensively monitors the power cable operation data and respectively sends the data to the pre-processor;

the preprocessor is used for classifying, cleaning and standardizing the data monitored by the monitoring module and then sending the data to the central processor through the unified interface;

the machine learning module is used for establishing a power cable operation model according to the data processed by the pre-processor and the big data of the cloud database, and obtaining the most available operation data standard of the power cable through a machine learning optimization model;

the analysis module is used for comparing and analyzing the real-time operation data of the power cable with the standard model data, judging the fault hidden danger of the power cable, positioning the fault hidden danger, facilitating the timely inquiry, judging the fault type after the fault is confirmed and giving an alarm;

the cloud database is used for collecting construction and operation data of the power cable on the Internet, and is convenient for establishing a cable operation model according to big data;

the client is used for displaying the system operation result, sending an analysis report and an alarm to a manager and inquiring and inputting system data by the manager;

the storage module is used for storing the running logs and data of the preprocessor and the central processing unit;

and the central processing unit is used for coordinating the modules to work, calling corresponding data in each database in the authority of the central processing unit by the data calling command, and sending the control commands to the corresponding modules.

2. An edge-computing-based power cable monitoring and analysis system according to claim 1, wherein: the monitoring module comprises an intelligent grounding box unit, an intelligent well lid unit, an optical fiber vibration measuring unit and a video monitoring unit, wherein,

the intelligent grounding box unit is used as a connecting device of a grounding wire of a high-voltage grounding cable, is used for burglar alarm of a power line and the high-voltage grounding box, environment monitoring and real-time monitoring of cable circulation data, is internally provided with a 0-500A single-mode accurate current transformer and an accurate grounding circulation acquisition module, monitors current and voltage parameters (three items of six paths of ABC) of a grounding point of each high-voltage cable metal sheath of a high-voltage cable of 110kV or above in real time, realizes accurate measurement and timing inspection measurement of 0-500A grounding circulation, is provided with a door opening alarm device, sends a real-time alarm signal when a box door is pried open, and monitors environmental parameters in the box through the installed temperature and smoke sensors;

the intelligent well lid unit establishes an intelligent network of well lid management distributed over the power cable through the installed locator and the unlocking monitoring unit, so that the active sensing and real-time monitoring of the state of the well lid are realized, and once the well lid is lost, alarm information is immediately sent to the pre-processor;

the optical fiber vibration measuring unit measures vibration by adopting distributed optical fibers, and achieves the effect of automatically positioning the vibration position through an external vibration signal while sensitively capturing external vibration, thereby achieving real-time monitoring of external damage faults;

the video monitoring unit utilizes the intelligent camera to realize remote monitoring of a protective area and sends monitoring data to the preprocessor, and the preprocessor identifies suspicious target objects in video streams through images to realize tracking monitoring.

3. A power cable monitoring and analysis system based on edge calculation as claimed in claim 2 wherein: the optical fiber vibration measuring unit and the video monitoring unit are cooperatively operated through the preprocessor, the optical fiber vibration measuring unit captures an effective external broken vibration signal and sends the effective external broken vibration signal to the preprocessor, and the preprocessor feeds back and controls the video monitoring unit, triggers a camera of the video monitoring unit to shoot on site and sends shooting information to the preprocessor, so that bidirectional monitoring is realized.

4. A power cable monitoring and analysis system based on edge calculation as claimed in claim 2 wherein: the intelligent grounding box unit comprises a box body (1), wherein two installation cavities (3) are arranged in the box body (1), one side of each installation cavity (3) is hinged to a box door (2), the other side of each installation cavity (3) is fixedly provided with an anti-theft lock (4), one installation cavity (3) is internally provided with a cable grounding device (5), the installation cavity (3) at the cable grounding device (5) is internally provided with a fireproof device (6), the fireproof device (6) is a plurality of temperature and smoke sensors and a dry powder extinguishing agent storage tank, the other installation cavity (3) is internally and fixedly provided with a power supply network module (7), a UPS (UPS) storage device (8), a grounding current voltage monitoring device (9), an outer-damage-prevention monitoring device (10) and a fault positioning device (11), and the cable state is monitored through the grounding current voltage monitoring device (9), the outer-damage-prevention monitoring device (10) and the fault positioning device (11), the grounding current and voltage monitoring device (9) is a 0-500A single-mode accurate current transformer and an accurate grounding circulating current acquisition module, the outer-damage-prevention monitoring device (9) is a signal receiver connected with a distributed optical fiber, and the fault positioning device (11) is a GPS module provided with a number and used for positioning the position of each grounding box.

5. An edge-computing-based power cable monitoring and analysis system according to claim 1, wherein: the preprocessor comprises a data interaction unit and a data preprocessing unit, the data interaction unit is used for carrying out data transmission with the plurality of monitoring modules and the central processing unit, and the preprocessing unit is used for preprocessing the data transmitted by the monitoring modules, so that the operation burden of the central processing unit is reduced, and the operation efficiency of the system is improved.

6. An edge-computing-based power cable monitoring and analysis system according to claim 5, wherein: the preprocessing unit comprises data cleaning and data standardization, the data cleaning is realized in a FIFO (first in first out) mode in advance, the abnormal data of the set part are processed locally, the data standardization is realized by preprocessing the data monitored by the intelligent grounding box unit, the intelligent well lid unit, the optical fiber vibration measuring unit and the video monitoring unit, and the data are uploaded in a unified structure of an array + linked list + red and black trees, so that the data can be received and processed by the central processing unit conveniently.

7. An edge-computing-based power cable monitoring and analysis system according to claim 1, wherein: the machine learning module comprises a modeling unit and a learning unit, the modeling unit is used for establishing a power cable operation model according to monitoring data and big data, and the learning unit is used for optimizing the model through a neural learning network.

8. An edge-computing-based power cable monitoring and analysis system according to claim 7, wherein: the modeling unit comprises model establishment and model auditing, wherein the model establishment establishes an examination recognition base according to power monitoring data, extracts signals and characteristic parameters, and adopts different kernel functions and characteristic parameters for modeling, the model auditing is used for testing various established models, comparing the advantages and disadvantages of the kernel functions and selecting an optimum function model;

the learning unit learning process includes: the method comprises the steps of collecting power cable operation fault data in a cloud database, analyzing the fault data to obtain cause factors causing line faults, screening key factors causing the power faults from the cause factors by adopting a standardized stepwise regression method by taking the cause factors as independent variables and times of faults as dependent variables to obtain data of the key factors, normalizing the data of the key factors to obtain normalized feature data, learning high-dimensional features by using a multilayer cyclic neural network, optimizing a power cable model, and determining parameter standards.

9. An edge-computing-based power cable monitoring and analysis system according to claim 1, wherein: the analysis unit comprises fault prediction, fault location and fault alarm, wherein the fault prediction is used for comparing data processed by the preprocessor with power cable model parameters, abnormal parameters are obtained through comparison, fault hidden dangers are judged, the fault location is used for rapidly determining the fault location according to position information contained in the abnormal parameters, maintenance personnel can rapidly go to confirm and process the fault location, and the fault alarm is used for classifying the fault information and sending an alarm according to the classification.

10. A power cable monitoring and analyzing method based on edge calculation is characterized by comprising the following steps:

s1, the monitoring module comprehensively monitors power cable operation data through the intelligent grounding box unit, the intelligent well lid unit, the optical fiber vibration measurement unit and the video monitoring unit and respectively sends the data to the pre-processor;

s2, the preprocessor data interaction unit is used for carrying out data transmission with the plurality of monitoring modules and the central processing unit, and the preprocessing unit preprocesses the data transmitted by the monitoring modules, so that the operation burden of the central processing unit is reduced, and the system operation efficiency is improved;

s3, the central processing unit receives the preprocessed monitoring data, a power cable operation model is established through the machine learning module according to the monitoring data and the big data, the model is optimized through the neural learning network, and power cable model parameters are determined;

s4, the analysis unit compares the data processed by the preprocessor with the power cable model parameters through fault prediction to obtain abnormal parameters, judges fault hidden danger, quickly determines fault positions according to position information contained in the abnormal parameters through fault positioning, classifies the fault information through fault alarms, sends alarms to clients according to categories, and is convenient for managers to quickly go to corresponding power cable positions to process.

Technical Field

The invention relates to the technical field of power systems, in particular to a power cable monitoring and analyzing system and a power cable monitoring and analyzing method based on edge calculation.

Background

The power cable is an important facility for ensuring city power supply, the setting range is wide, the power cable is often contained in power pipelines inside and outside a whole city, the operation management of the current power cable generally adopts a mode of manual periodic inspection, the occurrence of events is difficult to prevent, scientific means must be utilized, effective precautionary measures are taken, and some automatic monitoring facilities are set, the monitoring data of the automatic monitoring facilities are all sent to a control center, the automatic monitoring facilities are processed and analyzed by a control center processor, the data transmission quantity of the automatic monitoring facilities is huge, the machine swing is easy to occur at the working peak, the system operation is unstable, and the effective monitoring and fault early warning functions of the power cable cannot be realized.

Disclosure of Invention

In view of the above technical deficiencies, the present invention provides a power cable monitoring and analyzing system and method based on edge calculation.

In order to solve the technical problems, the invention adopts the following technical scheme: the invention provides a power cable monitoring and analyzing system based on edge calculation, which comprises:

the monitoring module comprehensively monitors the power cable operation data and respectively sends the data to the pre-processor;

the preprocessor is used for classifying, cleaning and standardizing the data monitored by the monitoring module and then sending the data to the central processor through the unified interface;

the machine learning module is used for establishing a power cable operation model according to the data processed by the pre-processor and the big data of the cloud database, and obtaining the most available operation data standard of the power cable through a machine learning optimization model;

the analysis module is used for comparing and analyzing the real-time operation data of the power cable with the standard model data, judging the fault hidden danger of the power cable, positioning the fault hidden danger, facilitating the timely inquiry, judging the fault type after the fault is confirmed and giving an alarm;

the cloud database is used for collecting construction and operation data of the power cable on the Internet, and is convenient for establishing a cable operation model according to big data;

the client is used for displaying the system operation result, sending an analysis report and an alarm to a manager and inquiring and inputting system data by the manager;

the storage module is used for storing the running logs and data of the preprocessor and the central processing unit;

and the central processing unit is used for coordinating the modules to work, calling corresponding data in each database in the authority of the central processing unit by the data calling command, and sending the control commands to the corresponding modules.

In a preferred embodiment, the monitoring module comprises an intelligent grounding box unit, an intelligent manhole cover unit, an optical fiber vibration measuring unit and a video monitoring unit, wherein,

the intelligent grounding box unit is used as a connecting device of a grounding wire of a high-voltage grounding cable, is used for burglar alarm of a power line and the high-voltage grounding box, environment monitoring and real-time monitoring of cable circulation data, is internally provided with a 0-500A single-mode accurate current transformer and an accurate grounding circulation acquisition module, monitors current and voltage parameters (three items of six paths of ABC) of a grounding point of each high-voltage cable metal sheath of a high-voltage cable of 110kV or above in real time, realizes accurate measurement and timing inspection measurement of 0-500A grounding circulation, is provided with a door opening alarm device, sends a real-time alarm signal when a box door is pried open, and monitors environmental parameters in the box through the installed temperature and smoke sensors;

the intelligent well lid unit establishes an intelligent network of well lid management distributed over the power cable through the installed locator and the unlocking monitoring unit, so that the active sensing and real-time monitoring of the state of the well lid are realized, and once the well lid is lost, alarm information is immediately sent to the pre-processor;

the optical fiber vibration measuring unit measures vibration by adopting distributed optical fibers, and achieves the effect of automatically positioning the vibration position through an external vibration signal while sensitively capturing external vibration, thereby achieving real-time monitoring of external damage faults;

the video monitoring unit utilizes the intelligent camera to realize remote monitoring of a protective area and sends monitoring data to the preprocessor, and the preprocessor identifies suspicious target objects in video streams through images to realize tracking monitoring.

In an embodiment, the optical fiber vibration measurement unit and the video monitoring unit cooperatively operate through the pre-processor, the optical fiber vibration measurement unit captures an effective external breaking vibration signal and sends the signal to the pre-processor, and the pre-processor controls the video monitoring unit in a feedback manner, triggers the camera of the video monitoring unit to shoot on site, and sends shooting information to the pre-processor, so that bidirectional monitoring is realized.

Preferably, the intelligent grounding box unit comprises a box body, two installation cavities are formed in the box body, one side of each installation cavity is hinged with a box door, the other side of each installation cavity is fixedly provided with an anti-theft lock, one installation cavity is internally provided with a cable grounding device, the installation cavity at the cable grounding device is internally provided with a fireproof device, the fireproof devices are a plurality of temperature and smoke sensors and dry powder extinguishing agent storage tanks, the other installation cavity is internally and fixedly provided with a power supply network module, a UPS (uninterrupted power supply) electricity storage device, a grounding current voltage monitoring device, an external damage prevention monitoring device and a fault positioning device, the states of the cables are monitored through the grounding current voltage monitoring device, the external damage prevention monitoring device and the fault positioning device, and the grounding current voltage monitoring device is a 0-500A single-mode accurate current transformer and an accurate grounding circulating current acquisition module, the anti-external-damage monitoring device is a signal receiver connected with the distributed optical fiber, and the fault positioning device is a GPS module provided with a number and used for positioning the positions of all the grounding boxes.

In an embodiment, the preprocessor includes a data interaction unit and a data preprocessing unit, the data interaction unit is used for data transmission with the plurality of monitoring modules and the central processing unit, and the preprocessing unit is used for preprocessing data transmitted by the monitoring modules, so that the operation burden of the central processing unit is reduced, and the operation efficiency of the system is improved.

The data cleaning is realized in a manner of presetting FIFO (first in first out) and locally processing abnormal data of a set part, and the data standardization is realized by preprocessing the data monitored by the intelligent grounding box unit, the intelligent well lid unit, the optical fiber vibration measuring unit and the video monitoring unit and uploading the data in a unified structure of an array + a linked list + a red-black tree, so that the data can be received and processed by a central processing unit conveniently.

In a preferred embodiment, the machine learning module includes a modeling unit and a learning unit, the modeling unit is configured to establish a power cable operation model according to the monitoring data and the big data, and the learning unit is configured to optimize the model through a neural learning network.

The modeling unit comprises model establishment and model auditing, wherein the model establishment establishes a diagnosis and identification library according to power monitoring data, extracts signals and characteristic parameters, and adopts different kernel functions and characteristic parameters for modeling, the model auditing is used for testing various established models, comparing the advantages and disadvantages of the kernel functions and selecting an optimum function model;

the learning unit learning process includes: the method comprises the steps of collecting power cable operation fault data in a cloud database, analyzing the fault data to obtain cause factors causing line faults, screening key factors causing the power faults from the cause factors by adopting a standardized stepwise regression method by taking the cause factors as independent variables and times of faults as dependent variables to obtain data of the key factors, normalizing the data of the key factors to obtain normalized feature data, learning high-dimensional features by using a multilayer cyclic neural network, optimizing a power cable model, and determining parameter standards.

Preferably, the analysis unit comprises fault prediction, fault location and fault alarm, wherein the fault prediction is used for comparing data processed by the preprocessor with power cable model parameters, abnormal parameters are obtained through comparison, fault hidden dangers are judged, the fault location is used for rapidly determining a fault location according to position information contained in the abnormal parameters, so that maintenance personnel can rapidly go to confirm and process the fault location, and the fault alarm is used for classifying the fault information and sending an alarm according to the classification.

A power cable monitoring and analyzing method based on edge calculation comprises the following steps:

s1, the monitoring module comprehensively monitors power cable operation data through the intelligent grounding box unit, the intelligent well lid unit, the optical fiber vibration measurement unit and the video monitoring unit and respectively sends the data to the pre-processor;

s2, the preprocessor data interaction unit is used for data transmission with the plurality of monitoring modules and the central processing unit, and the preprocessing unit is used for preprocessing the data transmitted by the monitoring modules, so that the operation burden of the central processing unit is reduced, and the system operation efficiency is improved;

s3, the central processing unit receives the preprocessed monitoring data, a power cable operation model is established through the machine learning module according to the monitoring data and the big data, the model is optimized through the neural learning network, and power cable model parameters are determined;

s4, the analysis unit compares the data processed by the preprocessor with the power cable model parameters through fault prediction to obtain abnormal parameters, judges fault hidden danger, quickly determines fault positions according to position information contained in the abnormal parameters through fault positioning, classifies the fault information through fault alarms, sends alarms to clients according to categories, and is convenient for managers to quickly go to corresponding power cable positions to process.

The invention has the beneficial effects that:

1. the running state of the power cable is monitored in all aspects through the intelligent grounding box unit, the intelligent well cover unit, the optical fiber vibration measuring unit and the video monitoring unit, the manual inspection difficulty is reduced, external damage conditions such as theft and the like are monitored in a targeted mode, and the power cable is prevented from being damaged;

2. the data interaction unit is used for transmitting data with the monitoring modules and the central processing unit, the preprocessing unit is used for erecting the data transmitted by the monitoring modules nearby based on edge calculation and preprocessing the data, so that the operation burden of the central processing unit is reduced, the operation efficiency of the system is improved, the operation safety of the system is ensured, and the phenomenon of machine swinging is avoided;

3. the method comprises the steps of establishing a power cable operation model according to monitoring data and big data through a machine learning module, optimizing the model through a neural learning network, determining power cable model parameters, comparing the data processed by a preprocessor with the power cable model parameters through fault prediction, obtaining abnormal parameters through comparison, judging fault hidden dangers, rapidly determining fault positions through fault positioning according to position information contained in the abnormal parameters, classifying the fault information through fault alarms, sending alarms to a client according to categories, facilitating managers to rapidly go to the corresponding power cable positions to process, and improving the safety of the power cable.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.

FIG. 1 is a schematic diagram of the system of the present invention.

FIG. 2 is a schematic diagram of a detection module according to the present invention.

FIG. 3 is a diagram of a pre-processor according to the present invention.

FIG. 4 is a diagram of a machine learning module according to the present invention.

FIG. 5 is a schematic diagram of an analysis module according to the present invention.

Fig. 6 is a schematic structural diagram of the intelligent grounding box unit of the present invention.

Description of reference numerals: 1. a box body; 2. a box door; 3. a mounting cavity; 4. an anti-theft lock; 5. a cable grounding device; 6. a fire protection device; 7. a power supply network module; 8. a UPS power storage device; 9. a ground current voltage monitoring device; 10. an external damage prevention monitoring device; 11. and a fault positioning device.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Example (b): as shown in fig. 1 to 6, the present invention provides an edge calculation-based power cable monitoring and analyzing system, including:

the monitoring module is used for comprehensively monitoring the power cable operation data and respectively sending the data to the pre-processor;

the preprocessor is used for classifying, cleaning and standardizing the data monitored by the monitoring module and then sending the data to the central processor through the unified interface;

the machine learning module is used for establishing a power cable operation model according to the data processed by the pre-processor and the big data of the cloud database, and obtaining the most available operation data standard of the power cable through the machine learning optimization model;

the analysis module is used for comparing and analyzing the real-time operation data of the power cable with the standard model data, judging the fault hidden danger of the power cable, positioning the fault hidden danger, facilitating the timely inquiry, judging the fault type after the fault is confirmed and giving an alarm;

the cloud database is used for collecting construction and operation data of the power cable on the Internet, and is convenient for establishing a cable operation model according to big data;

the client is used for displaying the system operation result, sending an analysis report and an alarm to a manager and inquiring and inputting system data by the manager;

the storage module is used for storing the running logs and data of the preprocessor and the central processing unit;

and the central processing unit is used for coordinating the modules to work, calling corresponding data in each database in the authority of the data calling command, and sending the control commands to the corresponding modules.

Further, the monitoring module comprises an intelligent grounding box unit, an intelligent well lid unit, an optical fiber vibration measuring unit and a video monitoring unit, wherein,

the intelligent grounding box unit is used as a connecting device of a grounding wire of a high-voltage grounding cable, is used for burglar alarm of a power line and the high-voltage grounding box, environment monitoring and real-time monitoring of cable circulation data, is internally provided with a 0-500A single-mode accurate current transformer and an accurate grounding circulation acquisition module, monitors current and voltage parameters (three ABC paths) of a grounding point of each metal sheath of the high-voltage cable of 110kV or above in real time, realizes accurate measurement and timing inspection measurement of the 0-500A grounding circulation, is provided with a door opening alarm device, sends a real-time alarm signal when a box door is pried open, and monitors environmental parameters in the box through the installed temperature and smoke sensors;

the intelligent well lid unit establishes an intelligent network of well lid management distributed over the power cable through the installed locator and the unlocking monitoring unit, realizes active sensing and real-time monitoring of the state of the well lid, and immediately sends alarm information to a pre-processor once the well lid is lost;

the optical fiber vibration measuring unit measures vibration by adopting distributed optical fibers, and achieves the effect of automatically positioning the vibration position through an external vibration signal while sensitively capturing external vibration, thereby achieving real-time monitoring of external damage faults;

the video monitoring unit utilizes the intelligent camera to realize remote monitoring of a protective area and sends monitoring data to the preprocessor, and the preprocessor identifies suspicious target objects in video streams through images to realize tracking monitoring.

Furthermore, the optical fiber vibration measuring unit and the video monitoring unit cooperatively operate through the pre-processor, the optical fiber vibration measuring unit captures an effective external broken vibration signal and sends the effective external broken vibration signal to the pre-processor, and the pre-processor feeds back and controls the video monitoring unit, triggers a camera of the video monitoring unit to shoot on site and sends shooting information to the pre-processor, so that bidirectional monitoring is realized.

Furthermore, the intelligent grounding box unit comprises a box body 1, two installation cavities 3 are arranged in the box body 1, a box door 2 is hinged on one side of each installation cavity 3, an anti-theft lock 4 is fixedly installed on the other side of each installation cavity 3, a cable grounding device 5 is installed in one installation cavity 3, a fireproof device 6 is installed in the installation cavity 3 at the cable grounding device 5, the fireproof devices 6 are a plurality of temperature and smoke sensors and dry powder extinguishing agent storage tanks, a power supply network module 7, a UPS electricity storage device 8, a grounding current voltage monitoring device 9, an external damage prevention monitoring device 10 and a fault positioning device 11 are fixedly installed in the other installation cavity 3, the states of the cables are monitored through the grounding current voltage monitoring device 9, the external damage prevention monitoring device 10 and the fault positioning device 11, the grounding current voltage monitoring device 9 is a 0-500A single-mode accurate current transformer and an accurate grounding circulation collection module, the external damage prevention monitoring device 9 is a signal receiver connected with a distributed optical fiber, and the fault positioning device 11 is a GPS module provided with a number and used for positioning the positions of all the grounding boxes.

Furthermore, the preprocessor comprises a data interaction unit and a data preprocessing unit, the data interaction unit is used for carrying out data transmission with the plurality of monitoring modules and the central processing unit, and the preprocessing unit is used for preprocessing the data transmitted by the monitoring modules, so that the operation burden of the central processing unit is reduced, and the operation efficiency of the system is improved.

Furthermore, the preprocessing unit comprises data cleaning and data standardization, the data cleaning is realized in a pre-set FIFO mode, the abnormal data of the set part are processed locally, the data standardization is realized by preprocessing the data monitored by the intelligent grounding box unit, the intelligent well lid unit, the optical fiber vibration measurement unit and the video monitoring unit, and the data are uploaded in a unified structure of an array + linked list + red-black tree, so that the central processing unit can receive and process the data conveniently.

Further, the machine learning module comprises a modeling unit and a learning unit, the modeling unit is used for establishing a power cable operation model according to the monitoring data and the big data, and the learning unit is used for optimizing the model through a neural learning network.

Further, the modeling unit comprises model establishment and model auditing, wherein the model establishment is used for establishing an examination recognition base according to the power monitoring data, extracting signals and characteristic parameters, and modeling by adopting different kernel functions and characteristic parameters, the model auditing is used for testing various established models, comparing the advantages and disadvantages of the kernel functions and selecting an optimum function model;

the learning unit learning process includes: the method comprises the steps of collecting power cable operation fault data in a cloud database, analyzing the fault data to obtain cause factors causing line faults, screening key factors causing the power faults from the cause factors by adopting a standardized stepwise regression method by taking the cause factors as independent variables and times of faults as dependent variables to obtain data of the key factors, normalizing the data of the key factors to obtain normalized feature data, learning high-dimensional features by using a multilayer cyclic neural network, optimizing a power cable model, and determining parameter standards.

Furthermore, the analysis unit comprises fault prediction, fault location and fault alarm, wherein the fault prediction is used for comparing data processed by the preprocessor with power cable model parameters, abnormal parameters are obtained through comparison, fault hidden dangers are judged, the fault location is used for rapidly determining the fault location according to position information contained in the abnormal parameters, convenience in maintenance personnel rapidly go to confirm and process, and the fault alarm is used for classifying the fault information and giving an alarm according to the category.

A power cable monitoring and analyzing method based on edge calculation comprises the following steps:

s1, the monitoring module comprehensively monitors power cable operation data through the intelligent grounding box unit, the intelligent well lid unit, the optical fiber vibration measurement unit and the video monitoring unit and respectively sends the data to the pre-processor;

s2, the preprocessor data interaction unit is used for carrying out data transmission with the plurality of monitoring modules and the central processing unit, and the preprocessing unit is used for preprocessing the data transmitted by the monitoring modules, so that the operation burden of the central processing unit is reduced, and the system operation efficiency is improved;

s3, the central processing unit receives the preprocessed monitoring data, a power cable operation model is established through the machine learning module according to the monitoring data and the big data, the model is optimized through the neural learning network, and power cable model parameters are determined;

s4, the analysis unit compares the data processed by the preprocessor with the power cable model parameters through fault prediction to obtain abnormal parameters, judges fault hidden danger, quickly determines fault positions according to position information contained in the abnormal parameters through fault positioning, classifies the fault information through fault alarms, sends alarms to clients according to categories, and is convenient for managers to quickly go to corresponding power cable positions to process.

When the intelligent grounding box is used, the monitoring module comprehensively monitors the power cable operation data through the intelligent grounding box unit, the intelligent well lid unit, the optical fiber vibration measurement unit and the video monitoring unit, and respectively sends the data to the pre-processor; the preprocessing unit is used for carrying out data transmission with the monitoring modules and the central processing unit, and is used for erecting the data transmitted by the monitoring modules nearby and preprocessing the data based on edge calculation, so that the operation burden of the central processing unit is reduced, and the operation efficiency of the system is improved; the central processing unit receives the preprocessed monitoring data, a power cable operation model is established through the machine learning module according to the monitoring data and the big data, the model is optimized through the neural learning network, and power cable model parameters are determined; the analysis unit compares the data processed by the preprocessor with the power cable model parameters through fault prediction to obtain abnormal parameters, judges fault hidden dangers, quickly determines fault positions according to position information contained in the abnormal parameters through fault positioning, classifies the fault information through fault alarms, sends alarms to the client according to categories, and is convenient for managers to quickly go to the corresponding power cable positions to process.

It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.