Hydropower station management method and system based on data analysis, equipment and medium

Abstract

The invention relates to the technical field, in particular to a hydropower station management method and system based on data analysis, equipment and a medium, and the method mainly comprises the steps that a hydropower station fault diagnosis model is established, diagnosis is carried out through the hydropower station fault diagnosis model according to basic data, and a hydropower station fault diagnosis result is obtained. The robot control end obtains all fault devices corresponding to the data set based on the received fault data set, scores based on fault levels and distance information, sorts scores, outputs a checking sequence of the robot for the fault devices according to the sorting, and generates a robot walking path through the checking sequence; and when the robot arrives at the fault equipment, checking through the neural network image recognition model, and if the checking is correct, continuing to obtain the state data of the equipment. According to the method, research and innovation in the aspects of big data, cloud computing, Internet of Things, MR technology and the like are adopted, and reference is provided for intelligent management of the hydropower station industry.

Description

Technical Field

[0001] The present invention relates to the technical field, and more specifically, to a hydropower station management method, system, equipment, and medium based on data analysis. Background Art

[0002] Driven by the wave of intelligent manufacturing, as a core component in the energy field, the intelligent management of production equipment in hydropower stations has become a top priority for improving operation efficiency and ensuring safe production. However, there are still many challenges in the whole life cycle data asset management of current hydropower station production equipment. The existence of data islands hinders the effective circulation of information. When there are faults in the hydropower station, manual on-site inspections are still required, and the low operation and maintenance efficiency has become an urgent problem to be solved. In the field of hydropower stations, especially the in-depth integration and practical application in on-site equipment operations are still in the exploration stage. Summary of the Invention

[0003] The purpose of the present invention is to provide a hydropower station management method, system, equipment, and medium based on data analysis to solve the above problems in the prior art.

[0004] The present invention is achieved through the following technical solutions:

[0005] In a first aspect, a hydropower station management method based on data analysis includes:

[0006] Obtain the basic data of the current hydropower station, preprocess the basic data, establish a hydropower station fault diagnosis model, and diagnose according to the basic data through the hydropower station fault diagnosis model to output whether there is a fault in the current hydropower station;

[0007] If there is no fault, no processing is performed. If there is a fault, the data group with the fault is encrypted and sent to the remote transceiver. After receiving the encrypted data group with the fault, the remote transceiver decrypts and sends it to the inspection end, and the inspection end includes a robot control end and an MR control end;

[0008] The MR control end receives the data group with the fault and displays it, establishes a hydropower station database, and the hydropower station database includes a three-dimensional data model of the hydropower station, equipment appearance diagrams, and standard data of corresponding equipment, and maps all equipment to the three-dimensional data model of the hydropower station;

[0009] The robot control end obtains all the corresponding faulty equipment of the data group based on the received data group with the fault, obtains the current positioning information of the robot and maps it to the three-dimensional data model of the hydropower station, and obtains the distance information between the current robot and the faulty equipment through the three-dimensional data model of the hydropower station;

[0010] Set the fault level of the faulty device, and respectively judge the fault level of the current faulty device. Score based on the fault level and distance information, sort the score values, and output the viewing order of the robot for the faulty device according to the sorting. Generate the robot walking path through the viewing order;

[0011] When the robot arrives at the faulty device, verify through the neural network image recognition model. If the verification is correct, continue to obtain the status data of the device and transmit the status data of the current device back to the MR control terminal. If the verification is incorrect, report an error.

[0012] Preferably, the preprocessing of the basic data includes:

[0013] Set the first classification process and the second classification process. The first classification process is used to send a control signal for grouping the basic data sent by the same device into one group, and the second classification process is used to send a control signal for grouping the same type of basic data into one group;

[0014] According to the received external control instruction, select different classification processes to classify the data.

[0015] Preferably, it further includes:

[0016] Obtain the data group after classifying the basic data, and perform label marking on each data group with different classification processes;

[0017] Obtain the basic data in the data group after marking the labels, judge whether there are missing values in the data group. If there are, save the positions of the missing values, obtain the specific values of the data at different times in the basic data, and select the average value of the values to fill the missing positions.

[0018] Preferably, the establishment of the hydropower station fault diagnosis model includes:

[0019] Obtain any data group classified by the first classification process as the target data group, obtain the historical operation database of all devices, and obtain the label of the target data group to identify the device to which the current data group belongs;

[0020] Obtain the historical data of the device corresponding to the target data group from the historical operation database, compare the historical data with the data in the target data group, set an error threshold. If there is data in the target data group with a difference value from the historical data greater than the error threshold, mark the data as faulty and record the data value of the data group at that moment, and output it as the data group with a fault. If there is no data in the target data group with a difference value from the historical data greater than the error threshold, judge the next data group.

[0021] Preferably, setting the fault levels of the faulty devices and respectively determining the fault levels of the current faulty devices includes:

[0022] The fault levels include minor faults, general faults, and serious faults, and the severity increases in turn;

[0023] Obtain the number of data marked as faulty in the data groups with faults. If the number is less than 1% of the total number of data groups, the fault level of the device is a minor fault;

[0024] If the number is less than 1% to 5% of the total number of data groups, the fault level of the device is a general fault;

[0025] If the number is greater than 5% of the total number of data groups, the fault level of the device is a serious fault.

[0026] Preferably, it further includes:

[0027] Classify the current devices into power generation core devices or non-core devices. The power generation core devices are the related devices directly participating in power generation, and the non-core devices are the remaining devices other than the related devices directly participating in power generation;

[0028] If it is determined that the device corresponding to the data group with a current fault is a power generation core device, the corresponding fault level is corrected to a more serious level.

[0029] Preferably, the scoring based on the fault level and distance information includes:

[0030]

[0031] In the formula, F g is the scoring value, S i is the average distance between the robot and all devices, S o is the distance between the robot and a certain device, λ is a calculation coefficient, ξ a is the total number of data in the data groups with faults, ξ b is the number of faulty data in the data groups with faults.

[0032] In the second aspect, the present invention also provides a hydropower station management system based on data analysis, including:

[0033] A fault judgment module, configured to obtain the basic data of the current hydropower station, preprocess the basic data, establish a hydropower station fault diagnosis model, diagnose according to the basic data through the hydropower station fault diagnosis model, and output whether there is a fault in the current hydropower station; if there is no fault, no processing is performed, if there is a fault, the data group with the fault is encrypted and sent to the remote transceiver, and after receiving the encrypted fault data group, the remote transceiver decrypts and sends it to the inspection end, and the inspection end includes a robot control end and an MR control end;

[0034] A database processing module, configured to receive and display the data group with the fault by the MR control end, establish a hydropower station database, where the hydropower station database includes a three-dimensional data model of the hydropower station, equipment appearance diagrams and standard data of corresponding equipment, and map all equipment to the three-dimensional data model of the hydropower station; the robot control end obtains all the corresponding faulty equipment of the data group based on the received data group with the fault, obtains the current positioning information of the robot and maps it to the three-dimensional data model of the hydropower station, and obtains the distance information between the current robot and the faulty equipment through the three-dimensional data model of the hydropower station;

[0035] A robot control module, configured to set the fault level of the faulty equipment, respectively judge the fault level of the current faulty equipment, score based on the fault level and distance information, sort the score values, output the viewing order of the robot for the faulty equipment according to the sorting, and generate a robot walking path through the viewing order; when the robot reaches the faulty equipment, check through the neural network image recognition model, if the check is correct, continue to obtain the status data of the equipment and send the status data of the current equipment back to the MR control end, if the check is incorrect, an error is reported.

[0036] In a third aspect, the present invention provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the computer program, the above-mentioned hydropower station management method based on data analysis is implemented.

[0037] In a fourth aspect, the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the above-mentioned hydropower station management method based on data analysis is implemented.

[0038] The technical solution of the present invention has at least the following advantages and beneficial effects:

[0039] 1. The above method provided by the present invention mainly includes establishing a fault diagnosis model for a hydropower station, diagnosing according to basic data through the fault diagnosis model of the hydropower station, the robot control terminal obtaining all corresponding faulty devices of the data group based on the received faulty data group, scoring based on the fault level and distance information, sorting the scoring values, and outputting the viewing order of the robot for the faulty devices according to the sorting, and generating a robot walking path through the viewing order; when the robot reaches the faulty device, it is verified through a neural network image recognition model. If the verification is correct, the status data of the device is continuously obtained and the status data of the current device is transmitted back to the MR control terminal. If the verification is incorrect, an error is reported. Through the above method, research and innovation in aspects such as big data, cloud computing, Internet of Things, and MR technology provide reference for the intelligent management of the hydropower station industry.

[0040] 2. The establishment of a digital asset management system and the practical application of MR technology will revolutionize the operation and maintenance mode of hydropower stations at multiple levels, greatly improving the efficiency of equipment operation and maintenance, and significantly enhancing operation safety at the same time.

[0041] 3. The method of using the control of the robot to replace manual inspection improves the operation and maintenance efficiency. A database is established for various data of the robot's historical inspections, and big data technology is used to analyze a large amount of inspection data. Combining real-time monitoring data to analyze, diagnose, and warn the equipment status, and at the same time provide information statistical analysis and preliminary auxiliary diagnosis results. After obtaining an alarm, the operation and maintenance personnel can promptly call the robot to quickly reach the specified device, or the robot can also reach the specified location for inspection by itself through this method, and promptly view and verify the alarm information in order to quickly formulate countermeasures. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required to be used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0043] Figure 1 is a flow schematic diagram of the present invention;

[0044] Figure 2 is a system structure schematic diagram of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0045] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and illustrated herein can be arranged and designed in various different configurations.

[0046] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. The naming or numbering of steps in this application does not mean that the steps in the method flow must be executed in the time / logical sequence indicated by the naming or numbering. The named or numbered process steps can be changed in the execution order according to the technical purpose to be achieved, as long as the same or similar technical effects can be achieved.

[0047] The independently described modules or sub-modules can be physically separated or not: they can be implemented in software or in hardware, and some of the modules or sub-modules can be implemented in software and the functions of these modules or sub-modules can be called by a processor, and the other parts of the modules or sub-modules can be implemented in hardware, for example, through a hardware circuit. In addition, some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this application.

[0048] Please refer to Figure 1 - Figure 2 , the hydropower station management method based on data analysis provided by the present invention includes:

[0049] S101: Obtain the basic data of the current hydropower station, preprocess the basic data, establish a hydropower station fault diagnosis model, and perform diagnosis according to the basic data through the hydropower station fault diagnosis model, and output whether there is a fault in the current hydropower station;

[0050] Among them, the basic data of the current hydropower station includes, for example, the capacity, speed, net head, efficiency, etc. of the water turbine, the speed, voltage, power factor, efficiency, etc. of the generator, the capacity, speed, efficiency, water pumping volume, pressure of the water pump, and the shaft power, power generation speed, vibration value, oil pressure, temperature and other state index parameters of the remaining equipment. These parameters are of various types and large in number, and are constantly updated and generated. Therefore, it is necessary to perform a preprocessing on these parameters, such as data cleaning and label marking, to provide a basis for subsequent data use.

[0051] S102: If there is no fault, no processing is performed. If there is a fault, the data group with the fault is encrypted and sent to the remote transceiver. After receiving the encrypted data group with the fault, the remote transceiver decrypts it and sends it to the inspection terminal, where the inspection terminal includes a robot control terminal and an MR control terminal;

[0052] Among them, end-to-end communication transmission can all use wireless transmission. In this embodiment, the data confidentiality level of the hydropower station is relatively high, and the data to be sent needs to be encrypted. Among them, the MR control terminal can be an MR wearable device worn by the staff to perform an MR display, seamlessly integrating virtual information with the real environment, providing intuitive and real-time operation guidance and fault diagnosis solutions for the operation and maintenance personnel. Through this innovative interaction method, multiple users are supported to interact and discuss in the same virtual space. Users can operate through gestures, voices, or touches to achieve a natural interaction experience. On-site experts can provide real-time guidance to remote employees through MR technology, providing visual operation guidelines, reducing training costs, and improving work efficiency. MR can superimpose virtual objects onto the real environment, enabling users to feel a more real scene and operation experience during the collaboration process. The remote collaboration technology based on MR not only provides a new collaboration method for the team but also demonstrates great potential in improving efficiency, reducing costs, and enhancing the user experience, providing strong guarantee for safe production.

[0053] S103: The MR control terminal receives the data group with the fault and displays it, and establishes a hydropower station database. The hydropower station database includes a three-dimensional data model of the hydropower station, equipment appearance diagrams, and standard data of the corresponding equipment, and maps all equipment to the three-dimensional data model of the hydropower station;

[0054] S104: The robot control terminal obtains all the corresponding faulty equipment of the data group based on the received data group with the fault, obtains the current positioning information of the robot and maps it to the three-dimensional data model of the hydropower station, and obtains the distance information between the current robot and the faulty equipment through the three-dimensional data model of the hydropower station;

[0055] S105: Set the fault level of the faulty equipment, and respectively judge the fault level of the current faulty equipment. Score based on the fault level and distance information, sort the score values, output the viewing order of the robot for the faulty equipment according to the sorting, and generate a robot walking path through the viewing order;

[0056] Among them, the three-dimensional data model needs to write all the routes from the channel to the device. After the viewing order is confirmed by the robot's control system, retrieve all the routes from the robot's positioning point to the target device in the three-dimensional data model, and select the nearest route to move forward. In addition, for positioning, high-precision Beidou positioning or GPS positioning can be used, or the method of sending communication signals can be used for positioning. This embodiment does not make restrictions, as long as the positioning function can be achieved.

[0057] S106: When the robot arrives at the faulty device, check it through the neural network image recognition model. If the check is correct, continue to obtain the status data of the device and transmit the status data of the current device back to the MR control terminal. If the check is incorrect, an error will be reported.

[0058] Among them, the data of the device can be set to be wirelessly transmitted to other ports. Then, after the robot is within the receiving range, it can directly receive the status data of the device. By using this method, the inconvenience of the robot still needing to connect to the device interface can be avoided.

[0059] In this embodiment, the image recognition technology used by the robot is based on the traditional deep learning mode and combines six algorithms such as the convolutional neural network algorithm and the spatial pyramid pooling algorithm. Compared with the traditional image recognition algorithm, through processes such as continuous image collection, in-depth processing, deep learning, and high-order control, as the time of using the robot system goes by, the image recognition rate and the adaptability to special inspection points in the scene are continuously enhanced and improved. Combining various perception data such as device status recognition data, infrared thermal imaging temperature field data, and on-site sounds, through complex judgment logic and calculations, accurately analyze and judge the on-site setting status to avoid false alarms. Establish a database for various data of the robot's historical inspections, use big data technology to analyze a large amount of inspection data, and jointly analyze and diagnose and predict the device status with real-time monitoring data. At the same time, provide information statistical analysis and preliminary auxiliary diagnosis results. After obtaining the alarm, the operation and maintenance personnel can promptly call the robot to quickly reach the specified device, check and verify the alarm information in time, so as to quickly formulate countermeasures.

[0060] The intelligent inspection robot system proposed in this project has the advantages of being able to operate all-weather and having stronger adaptability to harsh environments. According to the pre-set inspection content, time, cycle, route and other parameter information, it autonomously starts to complete routine inspection tasks, classifies and stores according to the alarm level, matter source, etc. and realizes intelligent alarm. By collecting various data such as images, sounds, temperatures, gases, etc., it realizes the comprehensive monitoring of the equipment operation status, instrument data, control cabinet indicator light status, etc. It is beneficial for operation and maintenance personnel to comprehensively and real-time understand the health level of electrical equipment, realize unattended and automatic inspection. Its inspection data is timely entered into the information management system, with operation flexibility and intelligence, assisting operation and maintenance personnel to carry out routine inspection work, greatly improving the inspection efficiency. This all-round data collection and processing ability ensures the accuracy and reliability of the inspection.

[0061] In the present invention, a data architecture is established. The data architecture is crucial for data management. It not only affects the performance and maintainability of the system, but also relates to the consistency, security and scalability of data. The architecture includes equipment monitoring data, core process parameters, operation and maintenance information and business data, forming an interconnected time series data set. It will process high-density and large-concurrency data from sources such as pre-installed sensors, post-installed professional monitoring systems, PLC master computers, SCADA and SAP.

[0062] An exemplary embodiment of the present invention, the preprocessing of basic data includes:

[0063] Set a first classification process and a second classification process. The first classification process is used to send a control signal for grouping the basic data sent by the same device as a group, and the second classification process is used to send a control signal for grouping the same type of basic data as a group; for the received external control instructions, select different classification processes to classify the data.

[0064] In this embodiment, 2 classification methods are adopted to facilitate the direct call of the subsequent system. One is based on the device as a unit to form an independent data group, and the other is to classify with the type as an independent data group. For example, all temperature data is a group, and all pressure data is a group.

[0065] Specifically, it further includes:

[0066] Obtain the data groups after classifying the basic data, and perform label marking on each data group with different classification processes;

[0067] The classified data groups are marked with labels, so that when the subsequent control terminal performs retrieval, it can more quickly find the required data groups.

[0068] Obtain the basic data within the data group with the marker tags completed, determine whether there are missing values in the data group. If there are, save the positions of the missing values, obtain the specific values of this data at different times from the basic data, and select the average value of the said values to fill the missing positions.

[0069] An exemplary embodiment of the present invention, establishing a hydropower station fault diagnosis model includes:

[0070] Obtain any data group classified by the first classification process as the target data group, obtain the historical operation databases of all devices, and obtain the device to which the current data group belongs through label recognition of the target data group;

[0071] Obtain the historical data of the device corresponding to the target data group from the historical operation database, compare the historical data with the data within the target data group, set an error threshold. If there is data in the target data group whose difference from the historical data is greater than the error threshold, mark this data as faulty, and record the data value of the data group at this moment, and output it as the data group with a fault. If there is no data in the target data group whose difference from the historical data is greater than the error threshold, then judge the next data group.

[0072] An exemplary embodiment of the present invention, the setting of the fault level of the faulty device and the separate judgment of the fault level of the current faulty device include:

[0073] The fault levels include minor faults, general faults, and severe faults, and the severity increases in sequence;

[0074] In this embodiment, a minor fault is generally a fault that can be handled within a certain period of time, a general fault is a fault that needs to be handled within a shorter time than a minor fault, and a severe fault is a fault that needs to be handled immediately.

[0075] Obtain the number of data marked as faulty in the data group with a fault. If the number is less than one percent of the total number of data groups, output the fault level of the device as a minor fault;

[0076] If the number is less than one percent to five percent of the total number of data groups, output the fault level of the device as a general fault;

[0077] If the number is greater than five percent of the total number of data groups, output the fault level of the device as a severe fault.

[0078] An exemplary embodiment of the present invention further includes:

[0079] Classify the current device as a power generation core device or a non-core device. The power generation core device is a device directly involved in power generation, and the non-core device is the remaining devices other than the devices directly involved in power generation;

[0080] For example, equipment directly involved in power generation such as water turbines, generators, and transformers are core equipment, while equipment such as transmission cables and water pumps are non-core equipment.

[0081] If it is determined that the equipment corresponding to the currently faulty data group is a core power generation equipment, then the corresponding fault level is corrected to a more serious level.

[0082] In an exemplary embodiment of the present invention, the scoring based on the fault level and distance information includes:

[0083]

[0084] In the formula, F g is the scoring value, S i is the average distance between the robot and all equipment, S o is the distance between the robot and a certain equipment, λ is a calculation coefficient, ξ a is the total number of data in the faulty data group, ξ b is the number of faulty data in the faulty data group.

[0085] In the above content, the higher the scoring value, the more urgent the situation of the current equipment. Therefore, the control system of the robot can view the corresponding equipment separately according to the scoring value from high to low. Among them, when the fault level is a minor fault, the calculation coefficient λ = 1; when the fault level is a general fault, the calculation coefficient λ = 2; when the fault level is a serious fault, the calculation coefficient λ = 3.

[0086] In a second aspect, the present invention also provides a water station management system based on data analysis, including:

[0087] A fault judgment module, configured to obtain the basic data of the current hydropower station, preprocess the basic data, establish a hydropower station fault diagnosis model, diagnose according to the basic data and through the hydropower station fault diagnosis model, and output whether there is a fault in the current hydropower station; if there is no fault, no processing is performed; if there is a fault, the faulty data group is encrypted and sent to the remote transceiver. After receiving the encrypted faulty data group, the remote transceiver decrypts and sends it to the inspection end, and the inspection end includes a robot control end and an MR control end;

[0088] The database processing module is configured to receive and display the faulty data group at the MR control end, establish a hydropower station database, where the hydropower station database includes a three-dimensional data model of the hydropower station, equipment appearance diagrams, and standard data of corresponding equipment, and map all equipment to the three-dimensional data model of the hydropower station; the robot control end obtains all the faulty equipment corresponding to the data group based on the received faulty data group, obtains the current positioning information of the robot and maps it to the three-dimensional data model of the hydropower station, and obtains the distance information between the current robot and the faulty equipment through the three-dimensional data model of the hydropower station.

[0089] The robot control module is configured to set the fault level of the faulty equipment, respectively judge the fault level of the current faulty equipment, score based on the fault level and distance information, sort the score values, output the viewing order of the robot for the faulty equipment according to the sorting, and generate a robot walking path through the viewing order; when the robot reaches the faulty equipment, it is verified through the neural network image recognition model. If the verification is correct, the status data of the equipment is continuously obtained and the status data of the current equipment is transmitted back to the MR control end. If the verification is incorrect, an error is reported.

[0090] The main control device, the main control device is used with the fault judgment module, the database processing module, and the robot control module to execute the above-mentioned hydropower station management method based on data analysis.

[0091] In addition, in each embodiment of the present invention, each functional unit can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0092] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. This computer software product is stored in a storage medium and includes several instructions to enable 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 in each embodiment of the present invention. And the foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read—On l yMemory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0093] The above are only the preferred embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A hydropower station management method based on data analysis, characterized in that: include: Obtain the basic data of the current hydropower station, pre-process the basic data, establish a hydropower station fault diagnosis model, perform diagnosis based on the basic data and through the hydropower station fault diagnosis model, and output whether there is a fault in the current hydropower station; If there is no fault, no processing is performed. If there is a fault, the data group with the fault is encrypted and sent to the remote transceiver. After receiving the encrypted data group of the fault, the remote transceiver decrypts it and sends it to the inspection end, which includes the robot control end and the MR control end. The MR control terminal receives and displays the data group of the fault, establishes a hydropower station database, and the hydropower station database includes a three-dimensional data model of the hydropower station, an equipment appearance diagram and standard data of corresponding equipment, and maps all equipment to the three-dimensional data model of the hydropower station; The robot control end obtains all corresponding faulty devices in the data group based on the received faulty data group, obtains the current positioning information of the robot and maps it to the three-dimensional data model of the hydropower station, and obtains the distance information between the current robot and the faulty device through the three-dimensional data model of the hydropower station; Set the fault level of the faulty device, and determine the fault level of the current faulty device respectively, score based on the fault level and distance information, and sort the score values. Output the robot's viewing order for the faulty device according to the sorting, and generate the robot's walking path based on the viewing order; When the robot reaches the faulty device, it verifies it through the neural network image recognition model. If the verification is correct, it continues to obtain the device status data and transmits the current device status data back to the MR control end. If the verification is incorrect, an error is reported.

2. The hydropower station management method based on data analysis according to claim 1 is characterized in that: The preprocessing of basic data includes: Setting a first classification process and a second classification process, wherein the first classification process is used to send a control signal for grouping by taking basic data sent by the same device as one group, and the second classification process is used to send a control signal for grouping by taking the same type of basic data as one group; According to the received external control instructions, different classification processes are selected to classify the data.

3. The hydropower station management method based on data analysis according to claim 2 is characterized in that: Also includes: Obtaining data groups that have completed classification of basic data, and labeling each data group with different classification processes; Obtain the basic data in the data group with the marked labels, determine whether there are missing values ​​in the data group, and if so, save the missing positions. Obtain the specific values ​​of the data at different times in the basic data, and select the average value of the values ​​to fill the missing positions.

4. The hydropower station management method based on data analysis according to claim 3 is characterized in that: The establishment of a hydropower station fault diagnosis model comprises: Obtain any data group classified by the first classification process as a target data group, obtain a historical operation database of all devices, and obtain a label of the target data group to identify the device to which the current data group belongs; Obtain the historical data of the equipment corresponding to the target data group from the historical operation database, compare the historical data with the data in the target data group, set an error threshold, and if there is data in the target data group that differs from the historical data by a value greater than the error threshold, mark the data as faulty, and record the data value of the data group at that moment, and output it as the data group with a fault. If there is no data in the target data group that differs from the historical data by a value greater than the error threshold, make a judgment on the next data group.

5. The hydropower station management method based on data analysis according to claim 4 is characterized in that: The step of setting the fault level of the faulty device and respectively determining the fault level of the current faulty device comprises: The fault levels include minor faults, general faults and serious faults, with the severity increasing in sequence; Obtaining the number of data marked as faults in the data group with faults, if the number is less than 1% of the total number of the data group, the fault level of the output device is a minor fault; If the number is less than 1% to 5% of the total number of data sets, the fault level of the output device is a general fault; If the number is less than 5% of the total number of data sets, the fault level of the output device is a serious fault.

6. The hydropower station management method based on data analysis according to claim 5 is characterized in that: Also includes: The current equipment is divided into power generation core equipment or non-core equipment, wherein the power generation core equipment is related equipment directly involved in power generation, and the non-core equipment is other equipment except the related equipment directly involved in power generation; If it is determined that the device corresponding to the data group currently having a fault is a core power generation device, the corresponding fault level will be corrected to a more serious level.

7. The hydropower station management method based on data analysis according to claim 5 is characterized in that: The scoring based on fault level and distance information includes: In the formula, F g is the score, S i is the average distance between the robot and all devices, S o is the distance between the robot and a certain device, λ is the calculation coefficient, ξ a is the total number of data in the data group with faults, ξ b is the number of faulty data in the data group with faults.

8. A hydropower station management system based on data analysis, characterized in that: include: The fault judgment module is configured to obtain basic data of the current hydropower station, pre-process the basic data, establish a fault diagnosis model for the hydropower station, perform diagnosis based on the basic data and through the fault diagnosis model for the hydropower station, and output whether there is a fault in the current hydropower station; If there is no fault, no processing is performed. If there is a fault, the data group with the fault is encrypted and sent to the remote transceiver. After receiving the encrypted data group of the fault, the remote transceiver decrypts it and sends it to the inspection end, which includes the robot control end and the MR control end. The database processing module is configured to receive and display the data group of the fault at the MR control end, establish a hydropower station database, and the hydropower station database includes a three-dimensional data model of the hydropower station, an equipment appearance diagram and standard data of corresponding equipment, and map all equipment to the three-dimensional data model of the hydropower station; the robot control end obtains all corresponding faulty equipment in the data group based on the received fault data group, obtains the current positioning information of the robot and maps it to the three-dimensional data model of the hydropower station, and obtains the distance information between the current robot and the faulty equipment through the three-dimensional data model of the hydropower station; The robot control module is configured to set the fault level of the faulty equipment, and judge the fault level of the current faulty equipment respectively, score based on the fault level and distance information, sort the score values, output the robot's viewing order of the faulty equipment according to the sorting, and generate the robot's walking path through the viewing order; when the robot arrives at the faulty equipment, it verifies through the neural network image recognition model. If the verification is correct, it continues to obtain the status data of the equipment and transmits the current status data of the equipment back to the MR control end. If the verification is incorrect, an error is reported.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the hydropower station management method based on data analysis described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the hydropower station management method based on data analysis as described in any one of claims 1 to 7 is implemented.

Images