Intelligent analysis method and system for hydroelectric generating set and medium

Abstract

The application relates to an intelligent analysis method and system for a hydroelectric generating set and a medium, the method comprising the following steps: a data source adopts a displacement reporting mode to upload hydroelectric generating set data to a server, the server collects data based on an IEC61850 protocol; specific trigger conditions are designed, an eventBus message bus is used to trigger test analysis message pushing monitoring of change events of key operation state signals of the hydroelectric generating set; when the trigger conditions are met, the eventBus pushes analysis trigger messages containing event types and time stamps, and the system immediately starts a corresponding special test analysis process to immediately analyze multi-dimensional operation data in an associated time window. The application not only realizes immediate analysis and processing of test data, but also establishes a complete traceable data chain through a persistent storage technology, effectively improving the intelligent level of power equipment state monitoring and the timeliness of data analysis.

Description

Technical Field

[0001] This application relates to the field of intelligent analysis of hydro-generator units, specifically to an intelligent analysis method, system, and medium for hydro-generator units based on multi-dimensional dynamic characteristics. Background Technology

[0002] Hydropower plays a vital role in the energy sector. Therefore, intelligent transformation of all aspects of hydropower operations, and improving the safety and stability of hydropower plant production processes through information technology, is of paramount importance. Current methods for information-based transformation of hydropower plants involve building monitoring platforms that integrate monitoring equipment, enabling the visualization of monitoring data. Based on this, data analysis is conducted to provide decision support for staff, significantly improving the efficiency of hydropower plant operations. Currently, most hydropower plant monitoring systems focus on major equipment such as turbines, generators, and main transformers, with very little research on monitoring auxiliary core equipment such as governors and excitation status.

[0003] The turbine governor is a crucial piece of basic automation equipment in a hydropower station, and its performance directly impacts the power generation quality and economic benefits of the station. The main functions of the turbine governor include starting up, stopping, increasing / decreasing load, and emergency shutdown of the turbine. The excitation system, as a vital component of the hydropower control system for voltage regulation, automatically adjusts the generator's excitation current according to the actual needs of the power grid. This ensures that the generator's output voltage and frequency remain stable, helping the power system maintain stability during faults or disturbances, and improving the reliability and stability of the power system. This is of great significance for preventing power system collapse and reducing power outages. Monitoring the governor and excitation system during operation is becoming increasingly difficult through routine inspections by maintenance personnel. Against this backdrop, developing an intelligent analysis platform for hydropower generator units based on multi-dimensional dynamic characteristics is particularly important.

[0004] Traditional monitoring systems focus on main equipment such as turbines and generators, with less than 40% coverage of the status perception of governors and excitation systems. Furthermore, they only collect basic parameters such as oil pressure, displacement, and voltage, making data analysis impossible. In addition, they cannot perform real-time data analysis of all parameters of the governor and excitation equipment to provide data-driven early warnings. Summary of the Invention

[0005] The purpose of this application is to provide an intelligent analysis method, system, and medium for hydro-generator units based on multi-dimensional dynamic characteristics, which provides a complete solution for intelligent operation and maintenance of power systems and has important engineering application value.

[0006] To achieve the above objectives, this application provides the following technical solution:

[0007] In a first aspect, embodiments of this application provide an intelligent analysis method for hydro-generator units based on multi-dimensional dynamic characteristics, comprising the following specific steps:

[0008] The data source uses a change-up reporting method to send the data of the hydro-generator unit to the server, and the server collects data based on the IEC61850 protocol.

[0009] Design specific triggering conditions and use the eventBus message bus to conduct triggering test analysis and message push monitoring of changes in key operating status signals of the hydro-generator unit. The key operating status signals include: start-up command, stop-up command, circuit breaker opening and closing status, guide vane opening command, frequency setpoint, unit active power, governor oil pressure and other status quantities or analog quantities collected and sent by intelligent terminal equipment.

[0010] When the triggering conditions are met, the eventBus pushes an analysis trigger message containing the event type and timestamp. The system then initiates the corresponding special test analysis process, performs real-time analysis on the multi-dimensional operational data within the associated time window, and calls the corresponding algorithm module to calculate a structured set of test feature parameters. Finally, the analysis results, along with the event metadata, are persistently stored in a relational or time-series database to form a traceable full lifecycle data chain. The multi-dimensional operational data includes guide vane opening, unit frequency, and active power.

[0011] The data source is the intelligent terminal equipment of the hydro-generator unit. The test analysis includes the test analysis of the hydro-generator unit's start-up process, shutdown process, low-frequency power oscillation and key operating conditions of primary frequency regulation performance, as well as the analysis of special test functions such as load shedding test, air disturbance test, idling test and static characteristic test.

[0012] The trigger condition for the start-up process test is that the start-up command changes from false to true. After the start-up process test is triggered, the guide vane opening, unit frequency, and opening setpoint data are recorded. During the start-up process test, the algorithm for calculating the maximum opening and maximum frequency of each of the three start-up processes is as follows:

[0013] Based on the guide vane opening, find the array index at the end of the first startup phase. Based on the startup command trigger time, if there are 4 consecutive moments where the difference between the opening and the 10th moment is greater than 2, it is determined that this moment is close to the end of the first startup phase. After that, if the difference between the opening at the current moment and the opening at the next moment is greater than 2, it is determined that the index of the current moment is the end of the first startup phase. Calculate the maximum value of the guide vane opening and the unit frequency during the time period from the startup command to the end of the first startup phase.

[0014] Based on the guide vane opening, find the array indices of the start and end times of the second stage. Starting from the end time of the first stage, if there are 30 consecutive opening times and the absolute value of the difference between the opening at the 30th time after that is greater than 2, then this time is determined to be the start time of the second stage. Based on the start time of the second stage, if the difference between the opening values ​​at 4 consecutive times and the opening value at the 4th time after that is greater than 4, then this is determined to be the end time of the second stage. Based on the start and end times of the second stage, calculate the maximum values ​​of the guide vane opening and the unit frequency within this time period.

[0015] Find the time corresponding to the minimum opening after the end of the second stage. From the end of the second stage to the end of the power-on, find the minimum opening value and the time corresponding to this minimum opening value.

[0016] Find the maximum opening degree and maximum frequency of the third stage. Within the time range of the minimum opening degree after the end of the second stage and the time range of the unit frequency climbing to 49.95HZ, find the maximum opening degree, which is the maximum opening degree of the third stage. From the time corresponding to the minimum opening degree after the end of the second stage to the end of the start-up time, find the maximum unit frequency, which is the maximum frequency of the third stage.

[0017] By introducing a step response analysis algorithm, the system automatically calculates key dynamic characteristic indicators such as overshoot, steady-state value, and maximum output. The step response analysis algorithm is based on the time response data of unit frequency or active power under step excitation, which includes start-up command triggering, frequency setpoint mutation, or load shedding events. The system extracts the corresponding output variable time series and its timestamp from the time series database as the original data source for calculating overshoot, steady-state value, and maximum output.

[0018] The implementation process of the step response analysis algorithm is as follows:

[0019] First, the time series data array and its corresponding time array are received as input parameters. The time series data array refers to the discrete sampled value sequence of the output variable of the controlled object extracted from the time series database after the test event that triggers the step response analysis. Specifically, it includes at least one of the following: unit frequency, active power, or guide vane opening.

[0020] Then, based on the data change trend, the maximum value and its corresponding time point are identified; next, the arithmetic mean of the last 100 sampling points of the data is taken as the steady-state value;

[0021] Then, a sliding window with a width of 100 data points is used to determine the steady-state time. When the cumulative deviation between the data in the window and the steady-state value is lower than a preset threshold, the time corresponding to the start point of the window is marked as the time when the system reaches steady state.

[0022] Finally, the overshoot index of the system is accurately calculated using the formula (maximum value - steady-state value) / (steady-state value - initial value).

[0023] In the shutdown process test, after receiving a remote shutdown command signal, the system automatically triggers the shutdown process test analysis procedure. The specific shutdown process test analysis procedure is as follows:

[0024] First, the shutdown start time value y1 and shutdown completion time value y2 are extracted from the historical database. Based on these two key parameters and the preset sequence length L, an ideal linear shutdown reference model is constructed: The slope coefficient ,intercept ;

[0025] Subsequently, the Euclidean distance between each discrete point on the actual shutdown curve and the linear shutdown reference model is calculated, and a smooth distance array is generated through mean filtering. Based on the curve characteristics analysis, the time point corresponding to the peak value D_max of the distance array is the inflection point of the first stage shutdown. Based on the opening value y3 at this inflection point, the system further establishes the second linear shutdown model with the opening value y2 at the shutdown completion time. The same distance calculation method and filtering process are used to determine the inflection point of the second stage shutdown by identifying the peak value of the distance array.

[0026] Finally, by integrating the timestamps corresponding to the inflection points of the first and second stages, and combining the shutdown start and end times, the duration of the first, second, and third stages, as well as the total time consumed by the overall shutdown process, were accurately calculated.

[0027] The specific process of the low-frequency power oscillation test is as follows:

[0028] The maximum, minimum, and power fluctuation values ​​are calculated based on the power data change trend. Detailed analysis is triggered when the fluctuation value exceeds a 10MW threshold. To ensure data integrity, the first and last intervals of the fluctuation interval array are removed, and the parity of the remaining interval count is checked. If the count is odd, the last interval is further removed to maintain an even length. Then, the direction of the first fluctuation interval is determined by judging the relationship between the inflection point value and the starting value, and key feature values ​​are extracted, including the maximum value p2 of the first rising interval, the minimum value p1 of the first falling interval, the maximum value p2n2 of the last rising interval, and the minimum value p2n1 of the last falling interval. The sampling time is calculated based on the total number of sampling points n. By combining the length of the fluctuation range, the number of fluctuations N = size / 2 and the frequency are derived. If the frequency is in the range of 0.02Hz to 2.5Hz, then the average damping ratio should be further calculated. Finally, when the damping ratio is below 3%, it is determined to be a low-frequency power oscillation event and recorded.

[0029] The air disturbance test is conducted manually. Before the test begins, the operator manually triggers the recording of the event start point and transmits waveform data to the front end for visualization in real time via WebSocket. After the test, the operator manually stops recording and marks the end time, and then the analysis process automatically starts: First, the frequency and frequency setpoint data for the test period are extracted from the time series database. Then, based on the step change moment of the frequency setpoint, the measured frequency data within a preset time window before and after it are extracted, and abnormal sampling points are removed to form a synchronized analysis dataset. Next, the step response analysis method is called to calculate dynamic characteristic parameters including overshoot, steady-state value, maximum output, rise time, peak time, settling time, number of fluctuations, and frequency decay. Finally, the complete analysis results are stored in a relational database.

[0030] The load shedding test specifically involves extracting historical data such as circuit breaker status, guide vane opening degree, and unit frequency from the time-series database according to the time interval selected by the user, and automatically completing the following analysis process: First, the tripping time is determined based on the circuit breaker signal change; then, the starting point of the action is determined through gradient analysis of the guide vane opening degree data, and the stationary time is calculated, along with the difference between the guide vane action time and the circuit breaker tripping time.

[0031] Extreme value analysis is performed on the unit frequency data to obtain the maximum frequency and its corresponding time, the minimum frequency and its corresponding time, and the speed rise time is calculated accordingly.

[0032] The system stabilization time is determined by a dual verification mechanism that first determines whether the frequency fluctuation has entered a set range, and then verifies the stability for 10 seconds.

[0033] After removing adjacent duplicate values ​​using data filtering, based on fluctuation threshold and inflection point detection algorithms, when the following conditions are met... Under certain conditions, the number of fluctuations is accurately counted, and a complete experimental analysis report is generated.

[0034] Secondly, embodiments of this application provide an intelligent analysis system for hydro-generator sets based on multi-dimensional dynamic characteristics. The system includes a memory and a processor. The memory includes a program for an intelligent analysis method for hydro-generator sets based on multi-dimensional dynamic characteristics. When the program for the intelligent analysis method for hydro-generator sets based on multi-dimensional dynamic characteristics is executed by the processor, it implements the steps of the intelligent analysis method for hydro-generator sets based on multi-dimensional dynamic characteristics as described above.

[0035] Thirdly, embodiments of this application provide a computer-readable storage medium storing program code, which, when executed by a processor, implements the steps of the intelligent analysis method for hydro-generator sets based on multi-dimensional dynamic characteristics as described above.

[0036] Compared with the prior art, the beneficial effects of the present invention are:

[0037] 1. Using the method of the present invention, the characteristic parameters of various test performances of the turbine unit under the action of the governor can be calculated quickly and accurately, thereby further evaluating the regulation performance of the governor.

[0038] 2. Using the method of this invention, the operating data of the turbine unit and the experimental analysis data related to the speed governor's regulation performance are serialized and stored in the power plant, providing support for the full life cycle data chain of the smart power plant.

[0039] 3. The method of the present invention can improve the level of intelligent automation of the speed control system. Attached Figure Description

[0040] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0041] Figure 1 This is a trend chart of the power-on process of the present invention.

[0042] Figure 2 This is a diagram showing the result of the three-stage power-on process of the present invention.

[0043] Figure 3 This is the trend curve of the three-stage shutdown process of the present invention.

[0044] Figure 4 This is the detection process for low-frequency power oscillations according to the present invention.

[0045] Figure 5 This is a waveform analysis diagram of the no-load frequency disturbance test of the present invention.

[0046] Figure 6 This is the complete waveform analysis process for the load shedding test of the present invention. Detailed Implementation

[0047] The technical solutions of the embodiments of this application will now be described with reference to the accompanying drawings. It should be noted that similar reference numerals and letters in the following drawings indicate similar items; therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings.

[0048] The terms “comprising,” “including,” or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase “comprising one…” does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0049] The terms “first,” “second,” etc., are used only to distinguish one entity or operation from another, and should not be construed as indicating or implying relative importance, nor as requiring or implying any such actual relationship or order between these entities or operations.

[0050] A smart analysis method for hydro-generator units based on multi-dimensional dynamic characteristics includes the following specific steps:

[0051] The data source uses a change-up reporting method to send the data of the hydro-generator unit to the server, and the server collects data based on the IEC61850 protocol.

[0052] Design specific triggering conditions and use the eventBus message bus to conduct triggering test analysis and message push monitoring of changes in key operating status signals of the hydro-generator unit. The key operating status signals include: start-up command, stop-up command, circuit breaker opening and closing status, guide vane opening command, frequency setpoint, unit active power, governor oil pressure and other status quantities or analog quantities collected and sent by intelligent terminal equipment.

[0053] When the triggering conditions are met, the eventBus pushes an analysis trigger message containing the event type and timestamp. The system then initiates the corresponding special test analysis process, performs real-time analysis on the multi-dimensional operational data within the associated time window, and calls the corresponding algorithm module to calculate a structured set of test feature parameters. Finally, the analysis results, along with the event metadata, are persistently stored in a relational or time-series database to form a traceable full lifecycle data chain. The multi-dimensional operational data includes guide vane opening, unit frequency, and active power.

[0054] Figure 1 This is a trend chart of the startup process. The startup process is triggered when the startup command changes from false to true, and data such as guide vane opening, unit frequency, and opening setpoint are recorded. The algorithm for calculating the maximum opening and maximum frequency for each of the three startup stages is as follows:

[0055] Figure 2This is the result diagram corresponding to the three-stage startup process. The array index at the end of the first startup stage is found based on the guide vane opening. Based on the startup command trigger time, if the difference between the opening at four consecutive moments and the opening at the tenth moment is greater than 2, If the opening value is close to the end of the first startup phase, then it is determined that this moment is nearing the end of the first startup phase. Afterwards, if the difference between the opening value at the current moment and the opening value at the next moment is greater than 2, If the index at the current moment is determined to be the end of the first stage of startup, then the maximum values ​​of the guide vane opening and unit frequency during the time period from the startup command to the end of the first stage of startup are calculated.

[0056] Based on the guide vane opening, find the array indices for the start and end times of the second segment. Starting from the end time of the first segment, if there are 30 consecutive opening times, the absolute value of the difference between the opening at the 30th consecutive opening time and that at the 30th consecutive opening time is greater than 2, If this time is greater than or equal to 4, then this moment is determined to be the start time of the second stage. From the start time of the second stage, if the difference between the opening value of four consecutive moments and the opening value of the fourth moment is greater than 4, then... If this is the start and end time of the second stage, then this is determined to be the end time of the second stage. Based on the start and end times of the second stage, the maximum values ​​of the guide vane opening and the unit frequency during this time period are calculated.

[0057] Find the time corresponding to the minimum opening after the end of the second stage. From the end of the second stage to the end of the power-on, find the minimum opening value and the time corresponding to this minimum opening value.

[0058] Find the maximum opening degree and maximum frequency of the third stage. Within the time range between the minimum opening degree after the end of the second stage and the time it takes for the unit frequency to climb to 49.95 Hz, find the maximum opening degree, which is the maximum opening degree of the third stage. From the time corresponding to the minimum opening degree after the end of the second stage to the end of the startup time, find the maximum unit frequency, which is the maximum frequency of the third stage.

[0059] Finally, a step response analysis algorithm is introduced, enabling the system to automatically calculate key dynamic characteristic indicators such as overshoot, steady-state value, and maximum output. The step response analysis algorithm is based on the time response data of the unit frequency or active power under step excitation, including start-up command triggering, frequency setpoint mutation, or load shedding events. The system extracts the corresponding output variable time series and its timestamp from the time series database as the original data source for calculating overshoot, steady-state value, and maximum output. The algorithm is implemented as follows: First, it receives a time series data array and its corresponding time array as input parameters. The time series data array refers to the discrete sampled value sequence of the output variable of the controlled object extracted from the time series database after the test event that triggers the step response analysis. Specifically, it includes at least one of the following: unit frequency, active power, or guide vane opening. Then, it identifies the maximum value and its corresponding time point based on the data change trend. Next, it takes the arithmetic mean of the last 100 sampling points as the steady-state value. Subsequently, it uses a sliding window (with a window width of 100 data points) to determine the steady-state time. When the cumulative deviation between the data in the window and the steady-state value is lower than a preset threshold, the time corresponding to the starting point of the window is marked as the time when the system reaches steady state. Finally, it accurately calculates the overshoot index of the system according to the formula (maximum value - steady-state value) / (steady-state value - initial value).

[0060] Figure 3 The trend curves of the three-stage shutdown process are shown. Upon receiving a remote shutdown command signal, the system automatically triggers the experimental analysis process, first retrieving the shutdown start value y1 and the shutdown completion value y2 from the historical database. Based on these two key parameters and the preset sequence length L, the system constructs an ideal linear shutdown reference model: The slope coefficient ,intercept .

[0061] Subsequently, the system calculates the Euclidean distance between each discrete point on the actual shutdown curve and the linear shutdown reference model, and generates a smoothed distance array through mean filtering. Based on curve characteristic analysis, the time point corresponding to the peak value D_max of this distance array is the inflection point (preliminary judgment value) of the first stage of shutdown. Based on this inflection point opening value y3, the system further establishes a second linear shutdown model with the opening value y2 at the shutdown completion time, using the same distance calculation method and filtering process, and determines the inflection point of the second stage of shutdown by identifying the peak value of the distance array.

[0062] Ultimately, by integrating the timestamps corresponding to the inflection points of the first and second stages, and combining them with the shutdown start and end times, the system accurately calculates the duration of the first, second, and third stages, as well as the total time consumed by the overall shutdown process.

[0063] Figure 4 The detection process for low-frequency power oscillations is demonstrated. This analysis is performed periodically every 60 seconds. First, the maximum, minimum, and their difference (power fluctuation value) are calculated based on the power data change trend. Detailed analysis is triggered when the fluctuation value exceeds a 10MW threshold. To ensure data integrity, the system removes the first and last intervals of the fluctuation interval array and checks the parity of the remaining interval count (if odd, the last interval is further removed to maintain an even length). Then, the direction (upward or downward) of the first fluctuation interval is determined by judging the relationship between the inflection point value (the second-to-last value in the interval sequence) and the starting value, and key feature values ​​are extracted (including the maximum value p2 of the first upward interval, the minimum value p1 of the first downward interval, and the maximum value p2n2 and minimum value p2n1 of the last upward interval). The sampling time is calculated based on the total number of sampling points n. Combining the fluctuation range length, we derive the fluctuation number N = size / 2 and the frequency f = N / T. If the frequency is within the range of 0.02Hz to 2.5Hz, then we further calculate the average damping ratio. Ultimately, when the damping ratio falls below 3%, it is identified as a low-frequency power oscillation event and recorded. This process, through multi-level condition judgments and parameter calculations, achieves accurate identification of low-frequency oscillations in power systems.

[0064] Figure 5 This is a waveform analysis diagram of an unloaded frequency disturbance test. The test supports three frequency disturbance modes: 4Hz, 2Hz, and 1Hz. It comprehensively analyzes key parameters such as steady-state value, maximum output, rise time, overshoot, peak time, settling time, frequency decay, and number of fluctuations. The test is manually controlled. Before the test begins, the operator manually triggers the recording of the event start point and transmits waveform data to the front end for visualization in real time via WebSocket. After the test, the operator manually stops recording and marks the end time. The system then automatically starts the analysis process: First, it extracts the frequency and frequency setpoint data for the test period from the time-series database. Then, based on the step change moment of the frequency setpoint, it extracts the measured frequency data within a preset time window before and after the step change, removing abnormal sampling points to form a synchronized analysis dataset. Next, it calls the step response analysis method to calculate dynamic characteristic parameters including overshoot, steady-state value, maximum output, rise time, peak time, settling time, number of fluctuations, and frequency decay. Finally, the complete analysis results are stored in a relational database.

[0065] Figure 6The complete waveform analysis process of a load shedding test is demonstrated. This test can accurately measure key parameters such as maximum frequency, minimum frequency, stationary time, speed rise time, settling time, time from minimum frequency to settling, and number of fluctuations. Based on the user-selected time interval, the system extracts historical data such as circuit breaker status, guide vane opening, and unit frequency from the time-series database. Through intelligent algorithms, it automatically completes the following analysis process: First, it determines the tripping time based on the circuit breaker signal change. Then, it determines the start point of the action and calculates the stationary time (the difference between the guide vane action time and the circuit breaker tripping time) through gradient analysis of the guide vane opening data. It performs extreme value analysis on the unit frequency data to obtain the maximum frequency and its corresponding time, the minimum frequency and its corresponding time, and calculates the speed rise time accordingly. It determines the system settling time through a dual verification mechanism (first determining if the frequency fluctuation enters the set interval, then verifying stability for 10 seconds). After removing adjacent duplicate values ​​using data filtering, it uses a fluctuation threshold and inflection point detection algorithm (when the condition is met...) (Under certain conditions) the number of fluctuations is accurately counted, ultimately generating a complete test analysis report. This analysis method achieves a multi-dimensional and accurate assessment of the dynamic characteristics of hydro-generator units under load shedding conditions.

[0066] In summary, the intelligent monitoring and analysis platform proposed in this application innovatively realizes full lifecycle management of power equipment data. Through a multi-source data acquisition architecture based on the IEC61850 protocol, combined with the real-time monitoring mechanism of the eventBus message bus, an intelligent analysis system with condition-triggered functionality is constructed. This solution not only enables real-time analysis and processing of test data but also establishes a complete and traceable data chain through persistent storage technology, effectively improving the intelligence level of power equipment condition monitoring and the timeliness of data analysis. The technology in this application provides a complete solution for intelligent operation and maintenance of power systems and has significant engineering application value.

[0067] This application provides an intelligent analysis system for hydro-generator sets based on multi-dimensional dynamic characteristics. The system includes a memory and a processor. The memory includes a program for an intelligent analysis method for hydro-generator sets based on multi-dimensional dynamic characteristics. When the program for the intelligent analysis method for hydro-generator sets based on multi-dimensional dynamic characteristics is executed by the processor, it implements the steps of the intelligent analysis method for hydro-generator sets based on multi-dimensional dynamic characteristics as described above.

[0068] This application provides a computer-readable storage medium storing program code. When the program code is executed by a processor, it implements the steps of the intelligent analysis method for hydro-generator sets based on multi-dimensional dynamic characteristics as described above.

[0069] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0070] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0071] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0072] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0073] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0074] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0075] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0076] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A smart analysis method for hydro-generator sets, characterized in that, The specific steps include the following: The data source uses a change-up reporting method to send the data of the hydro-generator unit to the server, and the server collects data based on the IEC61850 protocol. Design specific triggering conditions and use the eventBus message bus to conduct triggering test analysis and message push monitoring of changes in key operating status signals of the hydro-generator unit. The key operating status signals include: start-up command, stop-up command, circuit breaker opening and closing status, guide vane opening command, frequency setpoint, unit active power, governor oil pressure, etc. These status quantities or analog quantities are collected and sent by intelligent terminal equipment. When the triggering conditions are met, the eventBus pushes an analysis trigger message containing the event type and timestamp. The system then initiates the corresponding special test analysis process, performs real-time analysis on the multi-dimensional operational data within the associated time window, and calls the corresponding algorithm module to calculate a structured set of test feature parameters. Finally, the analysis results, along with the event metadata, are persistently stored in a relational or time-series database to form a traceable full lifecycle data chain. The multi-dimensional operational data includes guide vane opening, unit frequency, and active power.

2. The intelligent analysis method for a hydro-generator unit according to claim 1, characterized in that, The data source is the intelligent terminal equipment of the hydro-generator unit. The test analysis includes the test analysis of the hydro-generator unit's start-up process, shutdown process, low-frequency power oscillation and key operating conditions of primary frequency regulation performance, as well as the analysis of special test functions such as load shedding test, air disturbance test, idling test and static characteristic test.

3. The intelligent analysis method for a hydro-generator unit according to claim 2, characterized in that, The trigger condition for the start-up process test is that the start-up command changes from false to true. After the start-up process test is triggered, the guide vane opening, unit frequency, and opening setpoint data are recorded. During the start-up process test, the algorithm for calculating the maximum opening and maximum frequency of each of the three start-up processes is as follows: Based on the guide vane opening, find the array index at the end of the first startup phase. Based on the startup command trigger time, if there are 4 consecutive moments where the difference between the opening and the 10th moment is greater than 2, it is determined that this moment is close to the end of the first startup phase. After that, if the difference between the opening at the current moment and the opening at the next moment is greater than 2, it is determined that the index of the current moment is the end of the first startup phase. Calculate the maximum value of the guide vane opening and the unit frequency during the time period from the startup command to the end of the first startup phase. Based on the guide vane opening, find the array indices of the start and end times of the second stage. Starting from the end time of the first stage, if there are 30 consecutive opening times and the absolute value of the difference between the opening at the 30th time after that time is greater than 2, then this time is determined to be the start time of the second stage. Based on the start time of the second stage, if the difference between the opening values ​​at 4 consecutive times and the opening value at the 4th time after that time is greater than 4, then this is determined to be the end time of the second stage. Based on the start and end times of the second stage, calculate the maximum values ​​of the guide vane opening and the unit frequency within this time period. Find the time corresponding to the minimum opening after the end of the second stage. From the end of the second stage to the end of the power-on, find the minimum opening value and the time corresponding to this minimum opening value. Find the maximum opening degree and maximum frequency of the third stage. Within the time range of the minimum opening degree after the end of the second stage and the time range of the unit frequency climbing to 49.95HZ, find the maximum opening degree, which is the maximum opening degree of the third stage. From the time corresponding to the minimum opening degree after the end of the second stage to the end of the start-up time, find the maximum unit frequency, which is the maximum frequency of the third stage. By introducing a step response analysis algorithm, the system automatically calculates key dynamic characteristic indicators such as overshoot, steady-state value, and maximum output. The step response analysis algorithm is based on the time response data of unit frequency or active power under step excitation, which includes start-up command triggering, frequency setpoint mutation, or load shedding events. The system extracts the corresponding output variable time series and its timestamp from the time series database as the original data source for calculating overshoot, steady-state value, and maximum output.

4. The intelligent analysis method for a hydro-generator unit according to claim 3, characterized in that, The implementation process of the step response analysis algorithm is as follows: First, the time series data array and its corresponding time array are received as input parameters; the time series data array refers to the discrete sampled value sequence of the output variable of the controlled object extracted from the time series database after the test event that triggers the step response analysis, specifically including at least one of the unit frequency, active power or guide vane opening; Then, based on the data change trend, the maximum value and its corresponding time point are identified; next, the arithmetic mean of the last 100 sampling points of the data is taken as the steady-state value; Then, a sliding window with a width of 100 data points is used to determine the steady-state time. When the cumulative deviation between the data in the window and the steady-state value is lower than a preset threshold, the time corresponding to the start point of the window is marked as the time when the system reaches steady state. Finally, the overshoot index of the system is accurately calculated using the formula (maximum value - steady-state value) / (steady-state value - initial value).

5. The intelligent analysis method for a hydro-generator unit according to claim 2, characterized in that, In the test analysis of the shutdown process, the system automatically triggers the shutdown process test analysis procedure after receiving a remote shutdown command signal. The specific shutdown process test analysis procedure is as follows: First, the shutdown start time value y1 and shutdown completion time value y2 are extracted from the historical database. Based on these two key parameters and the preset sequence length L, an ideal linear shutdown reference model is constructed: The slope coefficient ,intercept ; Subsequently, the Euclidean distance between each discrete point on the actual shutdown curve and the linear shutdown reference model is calculated, and a smooth distance array is generated through mean filtering. Based on the curve characteristics analysis, the time point corresponding to the peak value D_max of the distance array is the inflection point of the first stage shutdown. Based on the opening value y3 at this inflection point, the system further establishes the second linear shutdown model with the opening value y2 at the shutdown completion time. The same distance calculation method and filtering process are used to determine the inflection point of the second stage shutdown by identifying the peak value of the distance array. Finally, by integrating the timestamps corresponding to the inflection points of the first and second stages, and combining the shutdown start and end times, the duration of the first, second, and third stages, as well as the total time consumed by the overall shutdown process, were accurately calculated.

6. The intelligent analysis method for a hydro-generator unit according to claim 2, characterized in that, The specific process of the low-frequency power oscillation test is as follows: The maximum, minimum, and power fluctuation values ​​are calculated based on the power data change trend. Detailed analysis is triggered when the fluctuation value exceeds a 10MW threshold. To ensure data integrity, the first and last intervals of the fluctuation interval array are removed, and the parity of the remaining interval count is checked. If the count is odd, the last interval is further removed to maintain an even length. Then, the direction of the first fluctuation interval is determined by judging the relationship between the inflection point value and the starting value, and key feature values ​​are extracted, including the maximum value p2 of the first rising interval, the minimum value p1 of the first falling interval, the maximum value p2n2 of the last rising interval, and the minimum value p2n1 of the last falling interval. The sampling time is calculated based on the total number of sampling points n. The number of fluctuations is derived by combining the length of the fluctuation range. and frequency If the frequency is in the range of 0.02Hz to 2.5Hz, then the average damping ratio should be further calculated. Finally, when the damping ratio is below 3%, it is determined to be a low-frequency power oscillation event and recorded.

7. The intelligent analysis method for a hydro-generator unit according to claim 2, characterized in that, The air disturbance test is conducted manually. Before the test begins, the operator manually triggers the recording of the event start point and transmits waveform data to the front end for visualization in real time via WebSocket. After the test, the operator manually stops recording and marks the end time, and then the analysis process automatically starts: First, the frequency and frequency setpoint data for the test period are extracted from the time series database. Then, based on the step change moment of the frequency setpoint, the measured frequency data within a preset time window before and after it are extracted, and abnormal sampling points are removed to form a synchronized analysis dataset. Next, the step response analysis method is called to calculate dynamic characteristic parameters including overshoot, steady-state value, maximum output, rise time, peak time, settling time, number of fluctuations, and frequency decay. Finally, the complete analysis results are stored in a relational database.

8. The intelligent analysis method for a hydro-generator unit according to claim 2, characterized in that, The load shedding test specifically involves extracting historical data such as circuit breaker status, guide vane opening degree, and unit frequency from the time-series database according to the time interval selected by the user, and automatically completing the following analysis process: First, the tripping time is determined based on the circuit breaker signal change; then, the starting point of the action is determined through gradient analysis of the guide vane opening degree data, and the stationary time is calculated, along with the difference between the guide vane action time and the circuit breaker tripping time. Extreme value analysis is performed on the unit frequency data to obtain the maximum frequency and its corresponding time, the minimum frequency and its corresponding time, and the speed rise time is calculated accordingly. The system stabilization time is determined by a dual verification mechanism that first determines whether the frequency fluctuation has entered a set range, and then verifies the stability for 10 seconds. After removing adjacent duplicate values ​​using data filtering, based on fluctuation threshold and inflection point detection algorithms, when the following conditions are met... Under certain conditions, the number of fluctuations is accurately counted, and a complete experimental analysis report is generated.

9. An intelligent analysis system for a hydro-generator set, characterized in that, The system includes a memory and a processor. The memory contains a program for an intelligent analysis method for hydro-generator sets. When the program for the intelligent analysis method for hydro-generator sets is executed by the processor, it implements the steps of the intelligent analysis method for hydro-generator sets as described above.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores program code, which, when executed by a processor, implements the steps of the intelligent analysis method for hydro-generator sets as described in any one of claims 1 to 8.

Images