Hydroelectric generating set stopping necessary detection online diagnosis method and system based on working condition identification

Abstract

The invention relates to a hydroelectric generating set stopping necessary inspection online diagnosis method based on working condition identification, and the method comprises the steps: carrying out the deep fusion of a plurality of technical means, such as intelligent working condition identification, real-time monitoring and early warning, efficient diagnosis and analysis, and comprehensive health assessment; the hydroelectric generating set online stopping necessary detection diagnosis and analysis method based on working condition recognition is established, the problems that a traditional mode is low in efficiency, high in subjectivity, limited in information amount and the like are solved, an efficient, accurate and safe solution is provided for operation and maintenance management of a hydroelectric generating set, and the overall operation and maintenance level of the hydroelectric industry can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of hydroelectric generating set monitoring, in particular to a hydroelectric generating set stop-check online diagnosis method and system based on working condition identification, a computer device and a computer readable storage medium. BACKGROUND

[0002] Hydroelectric generating sets have the characteristics of rapid start and flexible regulation. With the construction of new power systems, large-scale new energy construction and access, it is particularly important to use the flexibility of hydroelectric generating sets to regulate the supply and demand balance of the power system.

[0003] In the running process, hydroelectric generating sets are faced with frequent start-stop and working condition adjustment, and the requirements for unit performance state operation and maintenance management are also increasing. Most power plants need to carry out routine inspection and analysis on the unit after shutdown to ensure the normal state of the unit, prevent potential failures, and ensure the continuity and safety of power supply.

[0004] In related technologies, unit stop-check mainly adopts manual inspection method. Although this method has a long history, it has limitations in actual operation: first, the efficiency is low, manual inspection requires a lot of time and manpower, especially in large hydroelectric generating sets, the inspection process may be very time-consuming; second, it is subjective, manual inspection depends on the experience and skills of the inspector, different personnel may have different judgment standards, resulting in poor consistency of the inspection results; third, the amount of information is limited, manual inspection can only observe the external state of the unit, and it is difficult to obtain detailed running data inside the unit, which leads to inaccurate monitoring results. SUMMARY

[0005] The embodiments of the present application provide a hydroelectric generating set stop-check online diagnosis method, device, system, computer equipment and computer readable storage medium based on working condition identification, to at least solve the problem of low efficiency in related technologies.

[0006] In a first aspect, the embodiments of the present application provide a hydroelectric generating set stop-check online diagnosis method based on working condition identification, characterized in that the method comprises:

[0007] Collecting running working condition data of the hydroelectric generating set, performing unit working condition identification based on the running working condition data, and obtaining a working condition identification result;

[0008] Obtaining online monitoring data of the hydroelectric generating set through a state monitoring system, and obtaining deep diagnosis information based on the online monitoring data through a state feature extraction model, wherein the online monitoring data includes key running information, alarm information and state information;

[0009] The water turbine generator is subjected to multi-dimensional diagnostic analysis according to the working condition recognition result and the deep-level diagnostic information through a preset diagnostic analysis model, and maintenance suggestions and diagnostic results are generated.

[0010] In some embodiments, operation data of the water turbine generator is collected, and working condition recognition of the unit is performed based on the operation data to obtain a working condition recognition result.

[0011] Operation data of the water turbine generator is collected, wherein the operation data includes rotation speed, power and flow rate.

[0012] Working condition recognition is performed based on the operation data through a working condition recognition model to obtain the working condition recognition result, wherein the working condition recognition result includes a shutdown working condition and a running working condition, and the running working condition includes a starting process state, a shutdown process state, a stable power generation state, a load change state, an idle state and an idling state.

[0013] In some embodiments, the state feature extraction model includes a time sequence feature model, a statistical feature model and a mutation feature model.

[0014] Features are extracted from the online monitoring data through the time sequence feature model, the statistical feature model and the mutation feature model respectively, and the extracted features are classified according to time sequence and working condition type respectively to obtain time sequence features, statistical features and mutation features respectively.

[0015] The deep-level diagnostic information is obtained through comprehensive analysis of the time sequence features, the statistical features and the mutation features.

[0016] In some embodiments, the water turbine generator is subjected to multi-dimensional diagnostic analysis according to the working condition recognition result and the deep-level diagnostic information through a preset diagnostic analysis model, and maintenance suggestions and diagnostic results are generated.

[0017] Complete operation information in any complete unit operation process is obtained, and significant features in the unit operation process are calculated based on the complete operation information and the working condition recognition result.

[0018] A neural network model is used to train historical alarm information, disposal records and equipment working conditions and operation data in the unit operation process to obtain an abnormality matching model.

[0019] The specific alarm information in any operation is associated with alarm information in a historical database and corresponding disposal records and equipment working conditions through the abnormality matching model, a preset disposal process corresponding to the specific alarm information is determined according to the association result, and a disposal suggestion is generated according to the preset disposal process.

[0020] based on the historical data corresponding to the state information, training and prediction are performed through a machine learning model to obtain an abnormal diagnosis result; based on the state information, an expert analysis result is determined through an expert experience rule base;

[0021] Through a comprehensive evaluation model, the abnormal diagnosis result and the expert analysis result are combined to obtain a comprehensive diagnosis result.

[0022] In some embodiments, the significant features include: operation time of the unit, stable operation time, operation time in each load interval, vibration zone operation time, number of working condition changes, average load, average water head, and average flow rate;

[0023] The preset treatment process corresponds to any alarm information, and is a standardized process written based on historical treatment records and expert experience.

[0024] In some embodiments, the method further includes:

[0025] Through the machine learning model, running threshold intervals of the hydroelectric generating unit under different working conditions are extracted, wherein the running threshold intervals include: a normal interval, a pre-warning interval, and a fault interval;

[0026] Based on the working condition recognition result and the online monitoring data in any unit operation process, data contained in the running threshold intervals are matched, and in the case that the hydroelectric generating unit is in the pre-warning interval and the fault interval, an abnormal reminder information is outputted;

[0027] In the case that the hydroelectric generating unit is in the normal interval, a health quantitative evaluation result is obtained based on the running working condition data and the deep diagnosis information.

[0028] In some embodiments, the comprehensive evaluation model dynamically updates the health quantitative evaluation result of the hydroelectric generating unit according to changes in the running state and working condition data of the generator unit.

[0029] In a second aspect, the embodiments of the present application provide a hydroelectric generating unit stop-for-inspection online diagnosis system based on working condition recognition, which includes: a collection module, a monitoring module, and a diagnosis module, wherein:

[0030] The collection module is configured to collect running working condition data of the hydroelectric generating unit, perform unit working condition recognition based on the running working condition data, and obtain a working condition recognition result.

[0031] The monitoring module is configured to obtain online monitoring data of the hydroelectric generating unit through a state monitoring system, and obtain deep diagnosis information through a state feature extraction model based on the online monitoring data, wherein the online monitoring data includes: key operation information, alarm information, and state information.

[0032] The diagnostic module is configured to perform multi-dimensional diagnostic analysis on the hydroelectric generating set according to the working condition recognition result and the deep diagnosis information by a preset diagnostic analysis model, and generate a maintenance suggestion and a diagnosis result.

[0033] In a third aspect, an embodiment of the present application provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the method in the first aspect when executing the computer program.

[0034] In a fourth aspect, an embodiment of the present application provides a computer readable storage medium, which stores a computer program executable by a processor to implement the method in the first aspect.

[0035] Compared with the related art, the online diagnosis method for the hydroelectric generating set based on working condition recognition provided by the embodiment of the present application deeply integrates various technical means such as intelligent working condition recognition, real-time monitoring and early warning, efficient diagnostic analysis, and comprehensive health evaluation, establishes an online diagnosis analysis method for the hydroelectric generating set based on working condition recognition, solves the problems of low efficiency, strong subjectivity, and limited information quantity of the traditional method, and provides an efficient, accurate, and safe solution for the operation and maintenance management of the hydroelectric generating set, which helps to improve the overall operation and maintenance level of the hydroelectric industry. BRIEF DESCRIPTION OF DRAWINGS

[0036] The accompanying drawings, which are included to provide a further understanding of the present application, constitute a part of the present application, and the illustrative embodiments of the present application and their description serve to explain the present application, and do not constitute improper limitations on the present application. In the drawings:

[0037] Figure 1 FIG. 1 is a flowchart of a diagnosis method for a hydroelectric generating set based on working condition recognition according to an embodiment of the present application;

[0038] Figure 2 FIG. 2 is a structural block diagram of an online diagnosis system for a hydroelectric generating set based on working condition recognition according to an embodiment of the present application;

[0039] Figure 3 FIG. 3 is a running flowchart of an online diagnosis system for a hydroelectric generating set based on working condition recognition according to an embodiment of the present application;

[0040] Figure 4 FIG. 4 is a schematic diagram of the internal structure of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0041] In order to make the objects, technical solutions and advantages of the present application clearer, the present application will be described and illustrated below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and should not be used to limit the present application. Based on the embodiments provided by the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort belong to the scope of the present application.

[0042] Obviously, the accompanying drawings in the following description are only some examples or embodiments of the present application, and for those of ordinary skill in the art, the present application can also be applied to other similar scenarios without creative effort based on the accompanying drawings. In addition, it can be understood that although the efforts made in the development process can be complex and lengthy, for those of ordinary skill in the art related to the content disclosed in the present application, some design, manufacture or production changes based on the technical content disclosed in the present application are only routine technical means, and should not be understood as insufficient disclosure of the content disclosed in the present application.

[0043] In the present application, "embodiments" means that the specific features, structures or characteristics described in conjunction with the embodiments can be included in at least one embodiment of the present application. The phrase appears at various places in the specification does not necessarily refer to the same embodiment, nor is it mutually exclusive or alternative embodiments. It is explicitly and implicitly understood by those of ordinary skill in the art that the embodiments described in the present application can be combined with other embodiments without conflict.

[0044] Unless otherwise defined, technical terms and scientific terms used in the present application shall have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. The terms "a", "an", "one", "this", and similar referents in the context of describing the application are to be construed to be open-ended, referring to one or more than one, unless otherwise noted. The terms "comprising", "comprises", "including", "includes" and "containing", "contains" shall be construed as containing the stated steps, elements or features but do not preclude the presence or addition of one or more other steps, elements or features. The term "connected" and / or "coupled" to be construed as not necessarily being direct connections or couplings, but can be indirect connections or couplings through one or more other elements. The term "plurality" refers to two or more. The term "and / or" describes associated objects, and can mean three conditions: both, one, or none. The term " / " generally means "or". The terms "first", "second", "third", etc. are used to distinguish similar objects, and do not represent a specific order.

[0045] Hydroelectric generating units have the characteristics of fast start and flexible regulation. With the construction of new power systems, large-scale new energy construction and access, the use of the flexibility of hydroelectric generating units to regulate the supply and demand balance of the power system becomes particularly important. Frequent start-stop and working condition adjustment of hydroelectric generating units also require higher performance state operation and maintenance management. Most power plants do need to carry out routine inspection and analysis on the units after shutdown, in order to ensure the normal state of the units, prevent potential failures, and ensure the continuity and safety of power supply.

[0046] Traditional unit inspection after shutdown mainly adopts manual inspection method. Although this method has a long history, it has limitations in actual operation: first, the efficiency is low, manual inspection requires a lot of time and manpower, especially in large hydroelectric generating units, the inspection process can be very time-consuming; second, it is subjective, manual inspection depends on the experience and skills of the inspector, different personnel may have different judgment standards, resulting in poor consistency of inspection results; third, the amount of information is limited, manual inspection can only observe the external state of the unit, it is difficult to obtain detailed internal operation data of the unit.

[0047] The application scheme establishes an online stop-by-stop inspection diagnosis analysis method and system for hydroelectric generating units based on working condition recognition, realizes real-time online diagnosis analysis of stop-by-stop inspection of the unit by using online monitoring, machine learning, diagnosis model and other technical methods, solves the problems of low efficiency, strong subjectivity and limited information quantity of traditional methods, provides an efficient, accurate and safe solution for operation and maintenance management of hydroelectric generating units, and helps to improve the overall operation and maintenance level of the hydroelectric industry.

[0048] Figure 1 is a flowchart of a hydroelectric generating unit stop-by-stop inspection diagnosis method based on working condition recognition according to an embodiment of the application, as shown in Figure 1 The flowchart includes the following steps:

[0049] S101, collect the running working condition data of the hydroelectric generating unit, perform working condition recognition of the unit based on the running working condition data, and obtain a working condition recognition result;

[0050] The working condition recognition of the hydroelectric generating unit is an indispensable prerequisite for carrying out online stop-by-stop inspection diagnosis analysis. The hydroelectric generating unit has complex and diverse working condition classification, which is mainly divided into two categories of shutdown and running. In the running state, it is further subdivided into multiple different stages, each stage having unique characteristics.

[0051] In the starting process, the unit starts from a stationary state and gradually accelerates operation, and each component starts to work cooperatively. At this time, parameters such as speed and torque gradually change, and related auxiliary systems such as lubrication and cooling are started and gradually enter the working state. As the unit enters the shutdown process, each operating parameter gradually decreases according to a certain rule until the unit completely stops running.

[0052] In the stable power generation stage, the unit outputs stable power while maintaining a relatively constant speed, and each operating and monitoring parameter is maintained in a relatively stable interval, but there are still slight fluctuations.

[0053] When the unit is in the load change stage, due to the influence of external power demand and other factors, the power output of the unit will change, and in turn the speed, flow and other parameters will also change accordingly.

[0054] In the no-load state, the unit is running but does not output power to the outside, and the internal energy consumption is mainly used to maintain its basic operation such as idling; when idling, the unit maintains a basic mechanical rotation state with low energy consumption.

[0055] In different operating states, various operating and monitoring parameters of the unit will have significant differences. In this embodiment, the goal of working condition recognition is to recognize the current working condition of the unit by collecting and analyzing the operating data (such as rotating speed, power, flow, etc.) of the unit, establishing a model algorithm, and recording the current working condition to provide a basis for subsequent diagnosis. The working condition record is divided into complete unit operation record (from starting to stopping) and subdivided working condition record in the operation process (start-up process, stopping process, stable power generation process, load change process, no-load process, idling process, etc.).

[0056] In S102, online monitoring data of the hydroelectric generating unit is acquired through the state monitoring system, and deep diagnosis information is obtained through prediction based on the online monitoring data by a state feature extraction model, wherein the online monitoring data includes key operating information, alarm information and state information.

[0057] It should be noted that in the field of modern hydroelectric engineering, by accessing computer monitoring and unit state monitoring systems of a hydropower station, all-round real-time monitoring of key parameters of the hydroelectric generating unit can be achieved.

[0058] Among them, the monitoring data including load, rotating speed, flow, current, voltage and other key operating information can reflect the behavior indicators of the unit. The change of load reflects the dynamic adjustment of the output power of the unit to the outside world. When the electricity demand fluctuates, the load will change accordingly, and the coordinated change of the rotating speed, flow and other parameters is closely related. The stability of the rotating speed is directly related to the smoothness of the operation of the unit. The rotating speed has a specific range and change trend under different working conditions. The flow is closely related to the energy conversion efficiency, and reasonable flow control can ensure the efficient operation of the unit. The current and voltage data not only reflect the transmission of electric energy, but also reflect the health status of the electrical system.

[0059] At the same time, the state data such as vibration, swing, temperature and pressure can reflect the “health signs” of the unit. During the operation of the unit, the vibration and swing data can reflect the running stability of the mechanical parts. Abnormal vibration or swing may indicate problems such as part wear, looseness or imbalance. The temperature data monitor the thermal state of the equipment. Excessive temperature may cause equipment damage, insulation aging and other faults. Further, the pressure data are crucial for systems involving fluid transmission. Abnormal pressure changes may indicate risks such as system leakage, blockage or equipment failure.

[0060] The above various information is transmitted to the monitoring system through a reliable communication network for real-time analysis. The operating data plays an important role in the entire monitoring system. On the one hand, it is used for working condition analysis to help identify the current operating stage of the unit and provide a basis for subsequent operation and maintenance. On the other hand, it is used for correlation analysis with the state data to mine potential problems through the internal relationship between the data.

[0061] In addition, the alarm information and the state data can be combined for abnormal treatment and diagnostic analysis. Once the alarm information is received, the operation and maintenance personnel can quickly combine the state data to make a preliminary judgment to determine the approximate direction of the fault, so as to take emergency measures in time.

[0062] In the online monitoring process, the establishment of various state feature extraction models is the key to improving the diagnostic efficiency. These models can extract the feature information of important state quantities of the unit from massive data in real time. The specific model training implementation method is not described again in this embodiment.

[0063] Specifically, the time sequence feature can reveal the change rule of the data over time, the potential feature can reflect the distribution trend of the data in the statistical sense, and the mutation feature can help quickly locate the abnormal fluctuation point of the data. Detailed recording of these feature information can obtain deep diagnostic information for the subsequent diagnostic system, thereby significantly prompting the efficiency and accuracy of the diagnostic analysis, and providing strong support for ensuring the safe and stable operation of the hydroelectric generating unit.

[0064] In addition, it should be noted that the working condition recognition is usually performed at a specific time point or stage, for example, when the operating state of the unit changes, working condition recognition needs to be performed to determine the current working condition type, to provide accurate basic information for subsequent diagnosis and analysis. Online detection runs throughout the entire process of the hydroelectric generating unit operation, continuously monitoring the unit, regardless of the working condition of the unit, data is collected and analyzed in real time, so as to respond immediately when an abnormal situation occurs, thereby obtaining deeper feature information.

[0065] The ultimate goal of online monitoring and working condition recognition is to ensure the safe and stable operation of the hydroelectric generating unit. Online monitoring discovers and solves problems in the operation of the unit in time through real-time monitoring and abnormal treatment; working condition recognition provides a basis for subsequent diagnostic analysis by accurately judging the working condition of the unit, thereby helping to prevent faults from occurring in advance and improving the reliability and operation efficiency of the unit.

[0066] S103, according to the working condition recognition result and the deep diagnostic information, a multi-dimensional diagnostic analysis of the hydroelectric generating unit is performed through a preset diagnostic analysis model, to generate a maintenance suggestion and a diagnostic result.

[0067] Among them, when the hydroelectric generating unit is shut down, targeted inspection will be automatically performed according to the previous working condition recognition result. By analyzing the operating condition and the data, the unit is diagnosed and analyzed in each aspect according to the set diagnostic analysis model, to detect potential hidden faults; if an abnormality or a fault trend is detected, the system will automatically generate a maintenance suggestion or an alarm information.

[0068] Specifically:

[0069] 1) For a complete unit operation process (start-up - shutdown), combined with the results of the working condition recognition model, the significant features such as the operation time, stable operation time, operation time in each load interval, operation time in vibration zone, number of working condition changes, average load, average water head, average flow, etc. are calculated and stored, providing basic information for subsequent analysis of larger time dimension;

[0070] Among them, the associated working condition information in the analysis of operation information is one of the important inputs of the machine learning model. Under different operating conditions, the state quantities of the unit behave differently. By combining the working condition data in the operation information, the machine learning model can more accurately extract the operating threshold interval under different working conditions, thereby better judging the operation health status of the equipment.

[0071] In addition, each parameter in the significant features can also provide a reference for subsequent rule-based expert experience models. When establishing the experience rule library, experts can combine the operation time, load interval, water head flow, etc. under different operating conditions to develop rules that are more in line with actual conditions. For example, according to the historical operation data of the unit in a specific load interval and working condition, the expert can determine which state quantities are the key monitoring objects and the corresponding abnormal judgment criteria under this condition.

[0072] 2) Analysis of alarm information; after summarizing, classifying, and correlating the process alarm information that occurs during the operation of the unit, the neural network model is trained using historical alarm information, disposal records, equipment working conditions, and operation data to automatically correlate the current alarm information with past historical data and disposal procedures. After analyzing the alarm information, processing suggestions will be automatically generated based on the alarm level and type, combined with the pre-set disposal procedures.

[0073] Among them, the pre-set disposal procedure is a standardized procedure based on past historical disposal records and expert experience; the system pushes the corresponding disposal suggestions to the operator according to the content of the alarm information, helping them to respond quickly.

[0074] In an exemplary embodiment, when new alarm information is generated, the key features of the alarm information are extracted and analyzed, and then the system matches and correlates these features with the stored historical data to find similar historical alarm cases and their corresponding disposal procedures. At the same time, combined with the pre-set disposal procedures based on historical disposal records and expert experience, processing suggestions suitable for the current alarm situation are automatically generated. These processing suggestions are pushed to the operator in a clear and understandable way, and the operator can quickly understand the preliminary actions to be taken according to the pushed content, so as to respond to the alarm situation in the shortest possible time, reduce the loss caused by the fault, and improve the safety and stability of the unit operation.

[0075] 3) Analysis of state information; mainly divided into three aspects, including machine learning model based on historical data, rule-based expert experience model and comprehensive evaluation model, among which:

[0076] Machine learning model based on historical data: Different machine learning algorithms are selected to train the model based on historical data of different state quantities (monitoring values, associated working conditions) to extract their running threshold intervals (divided into normal, early warning and failure) under different running conditions. Through the monitoring data and working conditions of this unit running project and the threshold intervals trained historically, the running health status of the equipment is judged. If it is in the early warning or failure interval, an abnormal reminder information is issued, and if it is in the normal interval, a further quantitative health evaluation is carried out.

[0077] Rule-based expert experience model: By combining the experience of power plant professionals for each specific object (such as key components or state quantities of the unit), an expert experience rule base is established to quickly identify abnormal conditions during equipment operation. The rule-based expert experience model is configured and run based on the actual situation of each power plant and unit.

[0078] Comprehensive evaluation model: The quantitative health score of the machine learning model and the analysis results of the expert experience model are combined to establish a hierarchical health state evaluation system to evaluate the health status of the hydroelectric generating unit according to different levels (state quantities, components and whole machine). In the comprehensive evaluation process, different weight coefficients are set for different state quantities, components and working conditions. The comprehensive evaluation model is a dynamic system that updates the health evaluation of the equipment in real time according to the changes in the running state and working condition of the equipment.

[0079] Through the above steps S101 to S103, intelligent working condition recognition, real-time monitoring and early warning, efficient diagnosis and analysis and comprehensive health evaluation are deeply integrated to establish an online stop-by-stop diagnosis and analysis method for hydroelectric generating units based on working condition recognition, which solves the problems of low efficiency, strong subjectivity and limited information quantity of traditional methods, and provides an efficient, accurate and safe solution for the operation and maintenance management of hydroelectric generating units, which helps to improve the overall operation and maintenance level of the hydroelectric industry.

[0080] It should be noted that the steps shown in the above process or the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0081] The present application also provides a hydroelectric generating unit stop-by-stop online diagnosis system based on working condition recognition, Figure 2is a structural block diagram of a water-turbine-generator-set online diagnosis system based on working condition recognition according to an embodiment of the present application, as shown in Figure 2 The system comprises a collection module 20, a monitoring module 21 and a diagnosis module 22, wherein

[0082] The collection module 20 is configured to collect the running working condition data of the water-turbine-generator set, perform working condition recognition on the water-turbine-generator set based on the running working condition data, and obtain a working condition recognition result.

[0083] The monitoring module 21 is configured to acquire online monitoring data of the water-turbine-generator set through a state monitoring system, and perform prediction based on the online monitoring data through a state feature extraction model to obtain deep diagnosis information, wherein the online monitoring data comprises key running information, alarm information and state information.

[0084] The diagnosis module 22 is configured to perform multi-dimensional diagnosis analysis on the water-turbine-generator set according to the working condition recognition result and the deep diagnosis information through a preset diagnosis analysis model, and generate a maintenance suggestion and a diagnosis result.

[0085] Through the above system, intelligent working condition recognition, real-time monitoring and early warning, efficient diagnosis analysis and comprehensive health assessment and other technical modules are deeply integrated to establish an online water-turbine-generator-set diagnosis analysis method based on working condition recognition, which solves the problems of low efficiency, strong subjectivity and limited information quantity of traditional methods, and provides an efficient, accurate and safe solution for the operation and maintenance management of water-turbine-generator sets, which helps to improve the overall operation and maintenance level of the water and electricity industry.

[0086] In addition, Figure 3 is a running flowchart of a water-turbine-generator-set online diagnosis system based on working condition recognition according to an embodiment of the present application.

[0087] In one embodiment, Figure 4 is an internal structure diagram of an electronic device according to an embodiment of the present application, as shown in Figure 4 An electronic device is provided, which can be a server, and the internal structure diagram thereof can be as shown in Figure 4 The electronic device comprises a processor, a memory, a network interface and a database connected through a system bus. The processor of the electronic device is configured to provide computing and control capabilities. The memory of the electronic device comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The database of the electronic device is configured to store data. The network interface of the electronic device is configured to communicate with an external terminal through a network connection. The computer program is executed by the processor to implement a water-turbine-generator-set online diagnosis method based on working condition recognition.

[0088] Those skilled in the art can understand that, Figure 4 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the electronic device to which the scheme of the present application is applied. The specific electronic device can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0089] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by instructing the relevant hardware through a computer program, which can be stored in a non-volatile computer readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiments of each method. Any reference to memory, storage, database or other medium used in each embodiment provided by the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0090] Those skilled in the art should understand that each technical feature of the above-mentioned embodiments can be combined arbitrarily, and in order to make the description simple, each technical feature in the above-mentioned embodiments is not described all possible combinations, however, as long as the combination of these technical features does not exist contradictory, it should be considered as the scope of the present application.

[0091] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the patent. It should be noted that for those skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are within the scope of the present application. Therefore, the scope of protection of the patent of the present application should be subject to the appended claims.

Claims

1. An online diagnostic method for hydropower units based on operating condition identification, characterized in that: The method comprises: Collecting operating condition data of the hydropower unit, performing unit operating condition identification based on the operating condition data, and obtaining an operating condition identification result; Obtain online monitoring data of the hydropower unit through a condition monitoring system, and perform prediction based on the online monitoring data using a condition feature extraction model to obtain in-depth diagnostic information, wherein the online monitoring data includes: key operating information, alarm information, and condition information; By using a preset diagnostic analysis model, a multi-dimensional diagnostic analysis is performed on the hydropower unit according to the operating condition identification result and the deep diagnostic information, and maintenance recommendations and diagnostic results are generated.

2. The method according to claim 1, characterized in that Collecting the operating data of the hydropower unit, and identifying the unit operating condition based on the operating data, and obtaining the operating condition identification results include: Collecting operating data of the hydropower unit, wherein the operating data includes speed, power and flow; The operating condition identification model is used to perform operating condition identification based on the operating data to obtain the operating condition identification result, wherein the operating condition identification result includes: shutdown condition and operating condition, and the operating condition includes: startup process state, shutdown process state, stable power generation state, load change state, no-load state, and idling state.

3. The method according to claim 1, characterized in that The state feature extraction model includes: a time series feature model, a general trend feature model and a mutation feature model: Extracting features from the online monitoring data using the time series feature model, the general trend feature model, and the mutation feature model, and classifying the extracted features according to time sequence and operating condition type to obtain time series features, statistical features, and mutation features, respectively; The deep diagnostic information is obtained by comprehensively analyzing the time series features, the statistical features and the mutation features.

4. The method according to claim 1, wherein By using a preset diagnostic analysis model, and based on the operating condition identification results and the deep diagnostic information, a multi-dimensional diagnostic analysis of the hydropower unit is performed, including: Obtaining complete operating information during any complete unit operation process, and combining the complete operating information with the operating condition identification result to calculate significant features during the unit operation process; Using a neural network model, historical alarm information, disposal records, equipment operating conditions, and operating data during the operation of the unit are trained to obtain an anomaly matching model; By using the anomaly matching model, any specific alarm information is associated with the alarm information in the historical database, its corresponding handling records and equipment operating conditions, and based on the association results, a preset handling process corresponding to the specific alarm information is determined, and a handling suggestion is generated based on the preset handling process; Based on the historical data corresponding to the state information, a machine learning model is used to perform training and prediction to obtain abnormal diagnosis results; based on the state information, an expert analysis result is determined through an expert experience rule base; The comprehensive evaluation model is used to combine the abnormal diagnosis results and the expert analysis results to obtain the comprehensive diagnosis results.

5. The method according to claim 4, characterized in that The significant characteristics include: the unit's operating time, stable operating time, operating time in each load range, operating time in the vibration range, number of operating condition changes, average load, average water head and average flow; The preset handling process is a standardized process corresponding to any alarm information and is compiled based on historical handling records and expert experience.

6. The method according to claim 4, characterized in that The method further comprises: Extracting the operating threshold intervals of the hydropower unit under different operating conditions through the machine learning model, wherein the operating threshold intervals include: a normal interval, a warning interval, and a fault interval; Based on the operating condition identification result and the online monitoring data during any operation of the unit, matching is performed with the data contained in the operating threshold interval, and when the hydropower unit is in the warning interval and the fault interval, outputting abnormal reminder information; When the hydropower unit is in the normal range, a health quantitative evaluation result is obtained based on the operating condition data and the deep diagnostic information.

7. The method according to claim 6, characterized in that The comprehensive evaluation model dynamically updates the health quantitative evaluation results of the hydropower unit according to the operating status and operating condition data of the generator unit.

8. An online diagnostic system for hydropower units that requires inspection every time they are shut down based on working condition identification, characterized in that: include: Acquisition module, monitoring module and diagnosis module, among which; The acquisition module is used to collect the operating condition data of the hydropower unit, perform unit operating condition identification based on the operating condition data, and obtain an operating condition identification result; The monitoring module is used to obtain online monitoring data of the hydropower unit through a state monitoring system, and to perform prediction based on the online monitoring data using a state feature extraction model to obtain in-depth diagnostic information, wherein the online monitoring data includes: key operation information, alarm information and state information; The diagnostic module is used to perform multi-dimensional diagnostic analysis on the hydropower unit based on the working condition identification result and the deep diagnostic information through a preset diagnostic analysis model, and generate maintenance suggestions and diagnostic results.

9. A computer 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 method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images