A power distribution network intelligent terminal sub-health early warning method

Abstract

This invention discloses a method for early warning of sub-health conditions in distribution network smart terminals. It involves acquiring first recorded data, extracting features from the first recorded data, and performing data modeling to obtain a target model. This target model is then used to detect data collected from the terminal under test to determine the terminal's sub-health condition. Simultaneously, the target model is modified based on the actual condition of the distribution smart terminal, effectively ensuring its applicability. Embodiments of this invention facilitate the full utilization of existing data from the distribution master station, enabling online detection and early warning of sub-health conditions in distribution network smart terminals during use. Through correlation analysis of detection reports, terminal failure patterns can be obtained, allowing for convenient early intervention in operation and maintenance, effectively reducing the damage rate of smart terminals, extending their service life, saving operation and maintenance resources, and achieving lean operation and maintenance of distribution network smart terminals. This invention can be widely applied in the field of distribution network operation and maintenance technology.

Description

Technical Field

[0001] This invention relates to the field of power distribution network operation and maintenance technology, and in particular to a method for early warning of sub-health in intelligent terminals of power distribution networks. Background Technology

[0002] With the continuous growth in the number of power devices and the ongoing development of IoT power distribution terminal applications, the scale of IoT power distribution smart terminals is constantly expanding. Consequently, the number of defective or damaged IoT power distribution smart terminals is also increasing, leading to a sharp rise in maintenance workload and costs. The demand for maintenance technicians is difficult to fill. People are exploring ways to solve this problem.

[0003] Current technology mainly relies on the terminal's own alarm system and adopts a fault-based repair mode, which makes it difficult to proactively detect sub-healthy terminals. When a terminal is damaged, the manpower and material resources required for repair are large, and the terminal damage rate is high. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a practical and efficient method for early warning of sub-health in smart terminals of power distribution networks.

[0005] This invention provides a method for early warning of sub-health in distribution network smart terminals, comprising: acquiring recorded data from the database of the distribution master station as first recorded data, with each smart terminal as a unit; wherein the first recorded data includes at least one of the following: historical record data, real-time recorded data, and time-series recorded data; extracting features from the first recorded data to obtain a model feature curve set and a model feature data set; performing data modeling based on the model feature curve set and the model feature data set to obtain a target model; correcting the target model based on the actual condition of the smart terminal to be detected by performing longitudinal and horizontal comparative analysis on the smart terminal to be detected; performing sub-health detection on the data collected by the smart terminal to be detected using the target model, and generating a detection report according to type; and performing correlation analysis through a terminal ledger based on the detection reports of smart terminals of the same type to obtain the failure patterns of the smart terminals of the same type.

[0006] Optionally, the step of acquiring recorded data from the database of the power distribution master station as the first recorded data, on a unit basis using intelligent terminals, includes at least one of the following: acquiring environmental data; wherein the environmental data includes at least one of the following: time, terminal temperature, terminal humidity, terminal latitude and longitude, terminal height above ground, and installation location; acquiring communication data; wherein the communication data includes at least one of the following: wireless module signal and information flow; acquiring service channel data; wherein the service channel data includes at least one of the following: service online / offline records and telemetry and tele-signaling anomaly information; acquiring management channel data; wherein the management channel data includes at least one of the following: device CPU, memory, and hard disk usage, management online / offline records, and management proactive alarm information.

[0007] Optionally, the step of extracting features from the first recorded data to obtain a model feature curve group and a model feature data group, and performing data modeling based on the model feature curve group and the model feature data group to obtain a target model, includes: filtering abnormal smart terminals and marking the remaining smart terminals as first terminals; dividing the first recorded data into type dimensions and grouping the first terminals according to the type dimensions; wherein the type dimensions include at least one of the following: geographical range, installation location, antenna type, number of on / off cycles, online rate, specified value range, alarm intensity; aggregating and averaging the first recorded data of the first terminals in the same group according to time order and / or the type dimensions to obtain a basic curve group and a basic data group of the type dimensions; configuring a deviation threshold and type weight, and integrating the basic curve groups and basic data groups of each type dimension according to the group to obtain the target model.

[0008] Optionally, the step of detecting the data collected by the smart terminal to be detected using the target model and generating a detection report by type includes: determining the detection dimension; determining the grouping of the smart terminal to be detected according to the dimension in which it is located; comparing the data collected by the smart terminal to be detected with the target model to obtain the deviation degree of each data between the smart terminal to be detected and the target model; weighting, aggregating, and summarizing the deviation degrees to obtain the overall deviation degree of the smart terminal to be detected; comparing the overall deviation degree with the deviation threshold of the target model to obtain the detection result; and generating a detection report based on the deviation degrees, the overall deviation degree, and the detection result.

[0009] Optionally, it also includes: pushing information about detected potential sub-health terminals to a visualization interface for display in order of importance.

[0010] Optionally, it also includes collecting feedback results from on-site inspections, the feedback results being used to correct the target model.

[0011] Embodiments of the present invention also provide a system for early warning of sub-health of intelligent terminals in a power distribution network, comprising: a first module, the first module being used to acquire recorded data from the database of the power distribution master station as first recorded data on a unit basis for intelligent terminals; wherein, the first recorded data includes at least one of the following: historical record data, real-time recorded data, and time-series recorded data; a second module, the second module being used to extract features from the first recorded data to obtain a model feature curve set and a model feature data set, and to perform data modeling based on the model feature curve set and the model feature data set to obtain a target model; a third module, the third module being used to correct the target model based on the actual condition of the intelligent terminal to be detected by performing longitudinal and horizontal comparative analysis on the intelligent terminal to be detected; a fourth module, the fourth module being used to perform sub-health detection on the data collected by the intelligent terminal to be detected through the target model, and to generate a detection report according to the type; and a fifth module, the fifth module being used to perform correlation analysis through a terminal ledger based on the detection reports of intelligent terminals of the same type to obtain the failure patterns of the intelligent terminals of the same type.

[0012] Embodiments of the present invention also provide an electronic device, including a processor and a memory; the memory is used to store a program; the processor executes the program to implement the method described above.

[0013] Embodiments of the present invention also provide a computer-readable storage medium storing a program that is executed by a processor to implement the method described above.

[0014] Embodiments of the present invention also provide a computer program product, including a computer program that, when executed by a processor, implements the method described above.

[0015] The embodiments of the present invention have the following beneficial effects: By obtaining first record data from the database of the distribution master station, performing feature extraction and data modeling on the first record data to obtain a target model, and then using the target model to detect data collected from the terminal to be tested, potential sub-healthy terminals can be detected. This facilitates the full utilization of existing data from the distribution master station, enabling online detection of sub-healthy terminals in the distribution network during use. Timely detection before substantial damage to the smart terminal allows for early intervention in maintenance, effectively reducing the damage rate of smart terminals, extending their service life, saving maintenance resources, and achieving lean maintenance of distribution network smart terminals. Through correlation analysis of the detection reports, the failure patterns of terminals of similar equipment can be derived, which is beneficial for early warning of sub-health conditions of smart terminals and eliminating sub-health conditions at their inception. Simultaneously, the embodiments of the present invention modify the target model according to the actual condition of the distribution smart terminal, effectively ensuring the applicability of the target model. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart of the method steps provided in the embodiments of the present invention;

[0018] Figure 2 This is a basic terminal temperature curve provided in an embodiment of the present invention;

[0019] Figure 3 This is a terminal failure pattern curve provided in an embodiment of the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0021] To address the current limitations of technology in proactively detecting sub-healthy terminals, the high resource consumption for maintenance, and the high terminal damage rate, this invention provides a method for early warning of sub-health in distribution network smart terminals, referring to... Figure 1 , Figure 1 This is a flowchart of the method steps provided in an embodiment of the present invention. The method includes steps S100 to S500:

[0022] S100. Using smart terminals as units, record data is obtained from the database of the power distribution master station as the first record data; wherein, the first record data includes at least one of the following: historical record data, real-time record data, and time-series record data.

[0023] Specifically, taking a smart terminal as a unit, the system retrieves at least one of the following recorded data from the distribution master station's database: historical data, real-time recorded data, and time-series recorded data. Based on the concept of "smart terminal as a unit," the first recorded data retrieved from the distribution master station's database comprises the historical data, real-time recorded data, and time-series recorded data of one or more smart terminals within the distribution network. The retrieval of the first recorded data includes at least one of the following steps: S110 to S140.

[0024] S110. Obtain environmental data. The dimensions of the environmental data to be obtained include at least one of the following: time, terminal temperature, terminal humidity, terminal latitude and longitude, terminal height above the ground, and installation location.

[0025] S120. Acquire communication data. The dimensions of the acquired communication data include at least one of the following: wireless module signal and SIM card traffic.

[0026] S130. Obtain business channel data. The dimensions of the business channel data to be obtained include business online / offline records and at least one of the following: telemetry and teleindication anomaly information.

[0027] S140. Obtain management channel data. The dimensions of the obtained management channel data include at least one of the following: device CPU, memory, hard disk usage, management online / offline records, and management proactive alarm information.

[0028] By acquiring different types of data from various smart terminals, it is beneficial to analyze the terminal's operation from multiple angles and in all aspects, and to effectively establish an accurate target model.

[0029] S200. Extract features from the first recorded data to obtain a set of model feature curves and a set of model feature data. Perform data modeling based on the set of model feature curves and the set of model feature data to obtain the target model.

[0030] Specifically, the steps S210 to S240 are included:

[0031] S210: Filter out abnormal smart terminals and mark the remaining smart terminals as the first terminal.

[0032] Specifically, based on the marking of intelligent terminals by the power distribution master station, abnormal intelligent terminals can be identified, the identified abnormal terminals can be filtered, and the remaining intelligent terminals can be marked as the first terminal. Among them, abnormal intelligent terminals include terminals that have been determined to be damaged, intelligent terminals with obviously abnormal data in each dimension of the first recorded data, etc. After filtering out the abnormal intelligent terminals, the remaining ones are all normal intelligent terminals, which can correctly construct the target model.

[0033] S220. Divide the first recorded data into type dimensions, and group the first terminals according to the type dimensions. The type dimensions include at least one of the following: geographical range, installation location, antenna type, number of on / off cycles, online rate, specified value range, and alarm intensity.

[0034] Specifically, the first terminals are grouped according to the dimensions in the first recorded data. In embodiments of the present invention, the groups can be formed, for example, according to the geographical range of the smart terminals, or according to other dimensions of the smart terminals, such as the online rate of the smart terminals. By grouping the first terminals, the target model constructed from the first recorded data of the smart terminals under the same environmental or equipment conditions can be more suitable for the detection of sub-healthy terminals under such conditions, thus ensuring the accuracy of sub-healthy terminal detection.

[0035] S230. Aggregate and average the first record data of the first terminal in the same group according to the time order and / or type dimension to obtain the basic curve group and the basic data group of the type dimension.

[0036] Specifically, the first record data of the first terminal within the same group is aggregated and averaged according to time order and / or dimension to obtain the basic curve and basic data of the dimension. Specifically, environmental data and communication data are aggregated and averaged according to time order and dimension to obtain the basic curve set for environmental and communication data in that dimension; the basic curve set for environmental data is automatically updated during seasonal changes. Business channel data and management channel data are aggregated and averaged according to dimension to obtain the basic data set for business channel and management channel data in that dimension.

[0037] For specific examples, please refer to... Figure 2 , Figure 2This is a terminal temperature baseline curve provided in an embodiment of the present invention. For example, if the dimension is configured as a geographical range, terminal temperature data of first terminals within the same geographical range at the same time can be selected from the first recorded data of all first terminals. The average value is calculated to obtain the average terminal temperature within that dimension. A terminal temperature curve is generated based on the average terminal temperature over 24 hours. This terminal temperature curve is the baseline curve for the terminal temperature dimension within that group. It should be noted that in the embodiments of the present invention, multiple groups can be generated according to the division of dimensions. Each group has a baseline curve group and a baseline data group, used for sub-health detection of smart terminals within the dimension.

[0038] S240, configure the deviation threshold and type weight, integrate the basic curve groups and basic data groups of each dimension according to the grouping, and obtain the target model.

[0039] Specifically, in each group, a deviation threshold is configured for the basic curves within the basic curve group and the basic data within the basic data group, and a type weight is configured for each group dimension. The basic curve groups and basic data groups of each dimension are then integrated to obtain the target model. The target model includes at least one of the following: an environment model, a communication model, a business channel model, and a management channel model.

[0040] This concludes the detailed description of step S200. Step S200 enables the acquisition of a comprehensive target model, which is beneficial for detecting sub-healthy terminals.

[0041] S300: Based on the actual condition of the smart terminal to be tested, the target model is corrected by performing longitudinal and horizontal comparative analysis on the smart terminal to be tested.

[0042] Specifically, based on the actual condition of the smart terminals to be tested, longitudinal and horizontal comparative analyses are performed. The deviation threshold and type weights are dynamically adjusted based on the analysis results to correct the target model, making its identification of sub-healthy terminals more accurate. The longitudinal comparative analysis includes historical and current status analysis, while the horizontal comparative analysis analyzes the operational status of similar devices. Taking the horizontal comparative analysis as an example, based on feedback from on-site inspections, the overall deviation of correctly identified sub-healthy terminals is compared with that of normal terminals to obtain the type data with the largest deviation, thereby adjusting the weight of this type.

[0043] S400: Perform sub-health detection on the data collected from the smart terminal to be tested using the target model, and generate a test report according to the type.

[0044] Specifically, step S400 includes steps S410 to S450:

[0045] S410. Determine the detection dimension. Based on the dimension in which the smart terminal to be detected is located, determine the grouping of the smart terminals to be detected.

[0046] Specifically, the detection dimension is determined. Based on the dimension in which the smart terminal to be detected is located, the smart terminal to be detected is grouped. For example, the detection dimension can be determined as a geographical range, and the smart terminal to be detected is grouped according to the geographical range in which it is located.

[0047] S420. Compare the data collected by the smart terminal to be tested with the target model to obtain the deviation of each data between the smart terminal to be tested and the target model.

[0048] Specifically, the data collected by the smart terminal under test is comprehensively compared with the basic curve set and basic data set in the target model to obtain the data deviation degree between the smart terminal data under test and each basic curve and basic data in the target model. It can be understood that the above-mentioned data deviation degrees are the results of comparing different types of data collected by the smart terminal under test with the corresponding data types in the target model. For example, within the geographical area where the terminal under test is located, the terminal temperature collected by the terminal under test is compared with the terminal temperature of the target model within that geographical area; if there is a certain deviation, the terminal temperature deviation degree is obtained. Similarly, the management active alarm information deviation degree is obtained by comparing the management active alarm information of the terminal under test with that of the target model. All deviation degrees obtained through comparisons of different types of data constitute the various data deviation degrees.

[0049] S430. Weighted aggregate and categorize the deviations of each data point to obtain the overall deviation of the smart terminal to be tested.

[0050] Specifically, weights are assigned to all data types. In this embodiment, the total weight value can be 100. In other embodiments, the total weight value can be any other value that can represent the sum of weights. Each data type is then graded and scored according to its data deviation. In this embodiment, the total score can be 100. In other embodiments, the total score can be any other value. Based on the weight assignment and scoring of each data type, the overall deviation of the terminal is calculated using the following formula:

[0051] d = sum((w / 100) × e)

[0052] In the formula, d is the overall deviation of the smart terminal to be detected, w is the weight of each data type, and e is the score assigned to the data type corresponding to the weight.

[0053] It should be noted that, in the implementation of this embodiment of the invention, the weight allocation and the numerical values ​​of the graded scoring can be dynamically adjusted according to the on-site situation data fed back by the user or the actual warning needs.

[0054] S440. Compare the overall deviation with the deviation threshold of the target model to obtain the detection result.

[0055] Specifically, the deviation between the smart terminal and the target model is compared with the configured deviation threshold. If the deviation exceeds the threshold, the smart terminal to be tested is found to be in a sub-healthy state. If the deviation does not exceed the threshold, the smart terminal to be tested is found to be a normal terminal.

[0056] S450. Generate a test report based on the deviation of each data point, the overall deviation, and the test results.

[0057] Specifically, a test report is generated based on the deviation of each data point, the overall deviation, and the test results. The test report includes the above information and may also include at least one of the following: the basic information of the terminal and the original information collected by the terminal.

[0058] By comprehensively comparing the data collected by the smart terminal under test with the corresponding data in the target model, the deviation of each data is obtained. The deviations of each data are aggregated and classified into an overall deviation. The overall deviation is compared with the deviation threshold to obtain the test result and generate a test report. This method can accurately identify sub-healthy smart terminals in the power distribution network and intervene in operation and maintenance in advance before the smart terminal is substantially damaged, thereby reducing the damage rate of the terminal.

[0059] S500: Based on the test reports of similar smart terminals, perform correlation analysis through the terminal ledger to obtain the failure patterns of similar smart terminals.

[0060] Specifically, based on the test reports of similar smart terminals, correlation analysis is performed through the terminal ledger to derive the failure patterns of similar smart terminals, and then... Figure 3 , Figure 3 This invention provides a failure pattern curve for smart terminals. Based on the detection results of similar smart terminals, and through correlation with terminal ledgers, the failure data initiation points, failure formation points, and failure occurrence points of each type of smart terminal are summarized and analyzed to derive the terminal failure pattern for that type of smart terminal. For example, during the operation of a smart terminal, a failure may begin to emerge at point A, a potential failure may form at point B, and a failure may actually occur at point C. Obtaining the failure pattern of smart terminals allows for early warning of sub-health conditions for that type of smart terminal, facilitating immediate handling of the terminal when a failure occurs, reducing the sub-health rate of the terminal, and thus reducing the failure and damage rate.

[0061] Embodiments of the present invention further include steps S600 to S700:

[0062] S600 pushes information about potential sub-healthy terminals detected to a visual interface for display in order of importance.

[0063] Specifically, the information of the detected potential sub-healthy terminals is pushed to the visualization interface for display in order of importance. The information of the potential sub-healthy terminals is the information in the detection report, and the criteria for determining importance can be: designated lines, power supply users, and important lines.

[0064] S700: Collect feedback results from on-site inspections; the feedback results are used to revise the target model.

[0065] Specifically, in the embodiments of the present invention, collecting feedback results from on-site inspections can be achieved by issuing an early warning through a visual interface when a sub-healthy smart terminal is detected, reminding maintenance personnel to carry out maintenance, and collecting the maintenance results through a feedback module as feedback results; alternatively, maintenance personnel can discover abnormalities during routine maintenance, collect the maintenance results through a feedback module as feedback results, and use the feedback results to correct the target model.

[0066] Embodiments of the present invention also provide a sub-health early warning system for intelligent terminals in a power distribution network, comprising: a first module, configured to acquire recorded data from the database of the power distribution master station as first recorded data on a unit basis; wherein the first recorded data includes at least one of the following: historical record data, real-time recorded data, and time-series recorded data; a second module, configured to extract features from the first recorded data to obtain a model feature curve set and a model feature data set, and to perform data modeling based on the model feature curve set and the model feature data set to obtain a target model; a third module, configured to modify the target model based on the actual condition of the intelligent terminal to be detected by performing longitudinal and horizontal comparative analysis on the intelligent terminal to be detected; a fourth module, configured to perform sub-health detection on the data collected from the intelligent terminal to be detected using the target model, and generate a detection report according to the type; and a fifth module, configured to perform correlation analysis based on the detection reports of intelligent terminals of the same type through a terminal ledger to obtain the failure patterns of intelligent terminals of the same type.

[0067] Embodiments of the present invention also provide an electronic device, including a processor and a memory; the memory is used to store a program; the processor executes the program to implement the method described above.

[0068] Embodiments of the present invention also provide a computer-readable storage medium storing a program that is executed by a processor to implement the method described above.

[0069] Embodiments of the present invention also provide a computer program product, including a computer program that, when executed by a processor, implements the method described above.

[0070] The embodiments of the present invention have the following beneficial effects:

[0071] 1. By using data modeling and target model detection, we can detect distribution smart terminals with sub-health conditions in the distribution network. This can help us to identify problematic smart terminals in a timely manner, issue early warnings, and intervene in operation and maintenance in advance before the smart terminals are damaged, thereby reducing the damage rate of smart terminals and effectively extending the service life of the terminals.

[0072] 2. Generate test reports from the test results and push them to the visualization interface in order of importance, making the test results visible and facilitating maintenance arrangements;

[0073] 3. By obtaining the terminal failure patterns of similar smart terminals based on the test report, it is possible to provide early warning of sub-health conditions of smart terminals, which is conducive to immediate handling of terminals when faults occur, reducing the sub-health rate of terminals, and thus reducing the failure and damage rate.

[0074] 4. The target model is modified to make the target model of the present invention more applicable.

[0075] The following is an application scenario provided by an embodiment of the present invention:

[0076] Using smart terminals as units, recorded data is obtained from the database of the power distribution station as the first recorded data; wherein, the first recorded data includes at least one of the following: historical record data, real-time recorded data, and time-series recorded data; features are extracted from the first recorded data to obtain a model feature curve set and a model feature data set; data modeling is performed based on the model feature curve set and the model feature data set to obtain the target model; sub-health detection is performed on the data collected by the smart terminals to be tested using the target model, and a detection report is generated according to the type; based on the detection reports of smart terminals of the same type, correlation analysis is performed through the terminal ledger to obtain the failure patterns of smart terminals of the same type; based on the actual condition of the smart terminals to be tested, the target model is corrected by performing longitudinal and horizontal comparative analysis of the smart terminals to be tested.

[0077] In some alternative embodiments, the functions / operations mentioned in the block diagrams may not occur in the order shown in the operation diagrams. For example, depending on the functions / operations involved, two consecutively shown blocks may actually be executed substantially simultaneously, or the blocks may sometimes be executed in reverse order. Furthermore, the embodiments presented and described in the flowcharts of this invention are provided by way of example to provide a more comprehensive understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of various operations is changed and sub-operations described as part of a larger operation are executed independently.

[0078] Furthermore, although the invention has been described in the context of functional modules, it should be understood that, unless otherwise stated, one or more of the functions and / or features may be integrated into a single physical device and / or software module, or one or more functions and / or features may be implemented in a separate physical device or software module. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding the invention. Rather, given the properties, functions, and internal relationships of the various functional modules in the apparatus disclosed herein, the actual implementation of the module will be understood within the scope of conventional skill of an engineer. Therefore, those skilled in the art can implement the invention as set forth in the claims using ordinary techniques without excessive experimentation. It is also understood that the disclosed concepts are merely illustrative and are not intended to limit the scope of the invention, which is determined by the full scope of the appended claims and their equivalents.

[0079] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0080] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0081] More specific examples (a non-exhaustive list) of computer-readable media include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0082] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0083] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0084] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

[0085] The above is a detailed description of the preferred embodiments of the present invention. However, the present invention is not limited to the embodiments described. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of the present invention.

Claims

1. A method for early warning of sub-health in intelligent terminals of power distribution networks, characterized in that, include: Using smart terminals as units, record data is obtained from the database of the power distribution master station as the first record data; wherein, the first record data includes at least one of the following: historical record data, real-time record data, and time-series record data; Feature extraction is performed on the first recorded data to obtain a model feature curve group and a model feature data group. Data modeling is performed based on the model feature curve group and the model feature data group to obtain the target model. The step of extracting features from the first recorded data to obtain a set of model feature curves and a set of model feature data, and then performing data modeling based on the set of model feature curves and the set of model feature data to obtain a target model includes: Filter out abnormal smart terminals and mark the remaining smart terminals as the first terminal; The first recorded data is divided into type dimensions, and the first terminal is grouped according to the type dimensions; wherein, the type dimensions include at least one of the following: geographical range, installation location, antenna type, number of on / off cycles, online rate, specified value range, alarm intensity; The first recorded data of the first terminal in the same group are aggregated and averaged according to time order and / or the type dimension to obtain the basic curve group of the type dimension and the basic data group of the type dimension; Configure the deviation threshold and type weight, and integrate the basic curve group and the basic data group of each type dimension according to the group to obtain the target model; Based on the actual condition of the smart terminal to be tested, the target model is modified by performing longitudinal and horizontal comparative analyses on the smart terminal to be tested. Based on the actual condition of the smart terminal to be tested, longitudinal and horizontal comparative analyses are performed on the smart terminal to be tested. The deviation threshold and type weight are dynamically adjusted according to the analysis results to correct the target model. The longitudinal comparative analysis includes historical status analysis and current status analysis, while the horizontal comparative analysis is the operation status analysis of similar devices. The target model is used to perform sub-health detection on the data collected by the smart terminal to be detected, and a detection report is generated according to the type. Based on the test reports of similar smart terminals, a correlation analysis was conducted through the terminal ledger to obtain the failure patterns of the similar smart terminals.

2. The method for early warning of sub-health in a smart terminal of a power distribution network according to claim 1, characterized in that, The first set of recorded data, obtained from the database of the power distribution master station on a unit basis using smart terminals, includes at least one of the following: Acquire environmental data; wherein the environmental data includes at least one of the following: time, terminal temperature, terminal humidity, terminal latitude and longitude, terminal height above ground, and installation location; Acquire communication data; wherein the communication data includes at least one of the following: wireless module signal, information traffic; Acquire business channel data; wherein, the business channel data includes at least one of the following: business online / offline records, telemetry and teleindication anomaly information; Acquire management channel data; wherein, the management channel data includes at least one of the following: device CPU, memory, hard disk usage, management online / offline records, and management proactive alarm information.

3. The method for early warning of sub-health in a smart terminal of a power distribution network according to claim 1, characterized in that, The step of detecting the data collected by the smart terminal to be detected using the target model and generating a detection report according to the type includes: Determine the detection dimension; based on the dimension in which the smart terminal to be detected is located, determine the grouping of the smart terminals to be detected. The data collected by the smart terminal to be tested is compared with the target model to obtain the deviation degree of each data between the smart terminal to be tested and the target model; The weighted aggregation and classification of the various data deviations are used to obtain the overall deviation of the smart terminal to be tested. The overall deviation is compared with the deviation threshold of the target model to obtain the detection result; A test report is generated based on the deviation of each data point, the overall deviation, and the test results.

4. The method for early warning of sub-health in a smart terminal of a power distribution network according to claim 1, characterized in that, Also includes: Information on potential sub-health conditions detected will be pushed to a visual interface for display in order of importance.

5. The method for early warning of sub-health in a smart terminal of a power distribution network according to claim 1, characterized in that, Also includes: The feedback results from the on-site inspection are collected and used to revise the target model.

6. A pre-warning system for sub-health status of intelligent terminals in power distribution networks, characterized in that, include: The first module is used to retrieve recorded data from the database of the power distribution master station as first recorded data, with the smart terminal as the unit; wherein the first recorded data includes at least one of the following: historical record data, real-time recorded data, and time-series recorded data; The second module is used to extract features from the first recorded data to obtain a set of model feature curves and a set of model feature data, and to perform data modeling based on the set of model feature curves and the set of model feature data to obtain a target model. The step of extracting features from the first recorded data to obtain a set of model feature curves and a set of model feature data, and then performing data modeling based on the set of model feature curves and the set of model feature data to obtain a target model includes: Filter out abnormal smart terminals and mark the remaining smart terminals as the first terminal; The first recorded data is divided into type dimensions, and the first terminal is grouped according to the type dimensions; wherein, the type dimensions include at least one of the following: geographical range, installation location, antenna type, number of on / off cycles, online rate, specified value range, alarm intensity; The first recorded data of the first terminal in the same group are aggregated and averaged according to time order and / or the type dimension to obtain the basic curve group of the type dimension and the basic data group of the type dimension; Configure the deviation threshold and type weight, and integrate the basic curve group and the basic data group of each type dimension according to the group to obtain the target model; The third module is used to modify the target model by performing longitudinal and horizontal comparative analyses on the smart terminal to be tested based on the actual condition of the smart terminal to be tested. Based on the actual condition of the smart terminal to be tested, longitudinal and horizontal comparative analyses are performed on the smart terminal to be tested. The deviation threshold and type weight are dynamically adjusted according to the analysis results to correct the target model. The longitudinal comparative analysis includes historical status analysis and current status analysis, while the horizontal comparative analysis is the operation status analysis of similar devices. The fourth module is used to perform sub-health detection on the data collected by the smart terminal to be detected through the target model, and generate a detection report according to the type. The fifth module is used to perform correlation analysis on the detection reports of similar smart terminals through the terminal ledger to obtain the failure patterns of the similar smart terminals.

7. An electronic device, characterized in that, Including the processor and memory; The memory is used to store programs; The processor executes the program to implement the method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The storage medium stores a program that is executed by a processor to implement the method as described in any one of claims 1 to 5.

9. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method as described in any one of claims 1 to 5.

Images