A digital small hydropower station fault identification device and method

Abstract

The application relates to the technical field of hydropower station operation fault monitoring, in particular to a digital small hydropower station fault identification device and method, which comprises the following steps: collecting and preprocessing multi-source electrical time sequence data during operation of a digital small hydropower station; constructing an environmental noise interference condition of a monitoring interval; constructing an energy magnetic field fault condition of the monitoring interval; obtaining a hydropower station fault factor, and judging whether the digital small generator is faulty according to whether the hydropower station fault factor is greater than a preset threshold value. The application aims to improve fault identification precision and the flexible adjustment capability of a small hydropower station in a new power system.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of hydropower station operation fault monitoring, in particular to a digital small hydropower station fault identification device and method. BACKGROUND

[0002] With the increasing use of hydropower resources, small hydropower stations, as an important renewable energy power source in the process of energy supply in mountainous and remote areas, are gradually developing towards intelligence and digitization. However, most small hydropower stations were built early, are located in remote areas, and have low informatization level, less online monitoring data, and difficult fault handling. These inherent defects not only lead to low operation efficiency of small hydropower stations and frequent equipment failures, which may cause serious safety accidents, but also increase the difficulty of operation and maintenance, and are difficult to meet the green, intelligent and efficient development demand under the new energy system.

[0003] In the process of digital small hydropower station operation, according to the disclosure in "Application of Digital Twin Technology in Hydropower Fault Prediction and Diagnosis", the traditional fault diagnosis and identification method of digital small hydropower station mainly relies on analysis of historical data. However, the operation data of small hydropower stations are seriously affected by the humid environment and electromagnetic interference of hydropower stations, causing signal distortion of key parameters, which seriously damages the accuracy and reliability of fault feature extraction. At the same time, analysis of historical data seriously affected by environmental interference will further aggravate the risk of lagging diagnosis of small hydropower station fault identification, resulting in poor early fault capture sensitivity and high fault misjudgment rate of digital small hydropower station, which seriously restricts the flexible adjustment ability of small hydropower station in the new power system. SUMMARY

[0004] In order to solve the above technical problems, the present application provides a digital small hydropower station fault identification device and method, and the technical solution adopted is as follows:

[0005] In the first aspect, one embodiment of the present application provides a digital small hydropower station fault identification method, which comprises the following steps:

[0006] a) Collecting and preprocessing multi-source electrical time series data of digital small hydropower station during operation, including generator rotor current, generator active power, reactive power and excitation voltage data;

[0007] b) Based on the high-frequency frequency component of the generator rotor current data and the low-frequency frequency component of the active power data in the monitoring interval, and combining the average value difference of the reactive power data between the fluctuation intervals and the speed of the range of the excitation voltage data under the change of time in the monitoring interval, the environmental noise interference condition of the monitoring interval is constructed;

[0008] c) constructing an energy magnetic field fault condition of the monitoring interval based on the trend item difference of all adjacent time points of the generator active power data in the monitoring interval and the similarity between the active power and the reactive power, and combining the periodic item abnormality degree of the generator rotor current data and the fitted straight line slope of the excitation voltage data in the monitoring interval;

[0009] d) obtaining a hydropower station fault factor by using the ratio of the environmental noise interference condition and the energy magnetic field fault condition, and judging whether the digital small generator is faulty according to whether the hydropower station fault factor is greater than a preset threshold.

[0010] Preferably, the environmental noise interference condition is constructed in the following manner:

[0011] Based on the high-frequency frequency component of the generator rotor current data and the low-frequency frequency component of the active power data in the monitoring interval, the damp interference of the monitoring interval is determined.

[0012] Based on the average value difference of the reactive power data between fluctuation intervals and the speed of the excitation voltage data under the time change of the range of fluctuation, the electromagnetic interference of the monitoring interval is determined.

[0013] The normalized value of the product of the damp interference and the electromagnetic interference is taken as the environmental noise interference condition of the monitoring interval.

[0014] Preferably, the determination method of the damp interference of the monitoring interval is that the product of the variance of the maximum energy value of all high-frequency frequency components of the generator rotor current data in the monitoring interval and the difference between the maximum energy value and the minimum energy value of all low-frequency frequency components of the active power data is taken as the damp interference of the monitoring interval.

[0015] Preferably, the determination method of the electromagnetic interference of the monitoring interval is that the product of the absolute value accumulation result of the difference between the data average values of the reactive power data in all fluctuation intervals and the ratio accumulation result of the difference between the maximum excitation voltage and the minimum excitation voltage of the excitation voltage data in all fluctuation intervals and the time interval corresponding to the fluctuation interval is taken as the electromagnetic interference of the monitoring interval.

[0016] Preferably, the energy magnetic field fault condition is constructed in the following manner:

[0017] Based on the trend item difference of all adjacent time points of the generator active power data in the monitoring interval and the similarity between the active power and the reactive power, the energy conversion abnormality of the monitoring interval is determined.

[0018] Based on the periodic item abnormality degree of the generator rotor current data and the fitted straight line slope of the excitation voltage data in the monitoring interval, the electromagnetic conversion abnormality of the monitoring interval is determined.

[0019] The normalized value of the product of the energy conversion abnormality and the electromagnetic conversion abnormality is taken as the energy magnetic field fault condition of the monitoring interval.

[0020] Preferably, the determination method of the energy conversion abnormality of the monitoring interval is that the cumulative result of the absolute value of the trend item difference of all adjacent time points in the generator active power data in the monitoring interval is taken as the index of the exponential function with the natural constant as the base number, and the ratio of the calculation result of the exponential function and the cosine similarity between the active power and the reactive power is taken as the energy conversion abnormality of the monitoring interval.

[0021] Preferably, the determination method of the electromagnetic conversion abnormality of the monitoring interval is that the product of the coefficient of variation of the periodic term sequence of the generator rotor current data and the slope of the fitted straight line of the excitation voltage data in the monitoring interval is taken as the electromagnetic conversion abnormality of the monitoring interval.

[0022] Preferably, the high-frequency frequency component is a frequency component greater than or equal to 50 Hz after the data is converted to the frequency domain; and the low-frequency frequency component is a frequency component greater than 0 Hz and less than 5 Hz after the data is converted to the frequency domain.

[0023] Preferably, the fluctuation interval is the time range between adjacent peak data in the data in the monitoring interval.

[0024] In a second aspect, another embodiment of the present application further provides a digital small hydropower station fault identification device, which comprises a data acquisition module, a data analysis module, a data transmission module, and a man-machine interaction module.

[0025] The data acquisition module is configured to acquire and pre-process multi-source electrical time series data of the digital small hydropower station during operation.

[0026] The data analysis module is connected with the data acquisition module and is configured to execute the digital small hydropower station fault identification method to output a fault analysis result.

[0027] The data transmission module is configured to transmit data between modules.

[0028] The man-machine interaction module is configured to display the fault analysis result.

[0029] The present application has at least the following beneficial effects:

[0030] 1、The application is directed to the technical problem of small hydropower station harsh environment interference causing key parameter signal distortion, making the water power operation fault feature extraction precision damaged, and may further aggravate the small hydropower station fault recognition lag diagnosis risk, through the data acquisition module to collect the small hydropower station multi-source data, and according to the data analysis module provides a specific calculation quantization method of environmental noise interference condition and energy fault condition, which can accurately reflect the interference condition of small hydropower station running process of damp environment and electromagnetic interference on key parameters, and based on the multi-source data characteristics of hydropower station, the energy efficiency conversion and electromagnetic energy imbalance abnormality in the running process of small hydropower station are accurately captured, which lays a digital foundation for small hydropower station fault recognition.

[0031] 2、The application evaluates the fault condition of the fault factors of each generator of the small hydropower station through the man-machine interaction module, which can ensure that the fault feature of the small hydropower station with slight environmental interference is effectively captured, reduce the fault recognition lag diagnosis risk of digital small hydropower station, improve the fault recognition precision and the flexible adjustment ability of small hydropower station in new power system, and realize a kind of digital small hydropower station fault recognition device and method. BRIEF DESCRIPTION OF DRAWINGS

[0032] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiment or prior art description. Obviously, the drawings in the following description are only some embodiments of the application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0033] Figure 1 A flow chart of a digital small hydropower station fault recognition method provided by an embodiment of the application. DETAILED DESCRIPTION

[0034] The digital small hydropower station fault recognition device provided by an embodiment of the application comprises a data acquisition module, a data analysis module, a data transmission module, and a man-machine interaction module.

[0035] The data acquisition module is used to acquire and preprocess the multi-source electrical time series data of the digital small hydropower station during operation.

[0036] The data analysis module is connected with the data acquisition module and is used to execute a digital small hydropower station fault recognition method to output fault analysis results.

[0037] The data transmission module is used to transmit data between modules.

[0038] Specifically, the application adopts an independent LonWorks high-speed data communication network interface to realize data transmission between various functional modules of the digital small hydropower station, and the maximum communication rate can reach 1.25 Mbits / s.

[0039] In order to prevent the relevant data obtained by the data acquisition module from being lost due to network fluctuations or environmental interference during data transmission, the relevant data of the digital small hydropower station transmitted to the data analysis module are respectively taken as inputs, and a linear interpolation method is used for missing value filling processing.

[0040] The man-machine interaction module is used for displaying the fault analysis result.

[0041] Specifically, the visualization interface is accessed, which is responsible for displaying and outputting the fault analysis result of the small hydropower station output by the data analysis module, assisting relevant personnel in assessing the operation condition of the small hydropower station and formulating a response decision.

[0042] When the fault analysis result of the small hydropower station is displayed and output through the visualization interface, the model, position and fault factor of each fault generator are labeled in real time, which assists relevant personnel in timely mastering the operation condition of each generator in the digital small hydropower station and avoiding the risk of delayed diagnosis of fault identification. Since the access of the visualization interface is a known technology, the specific acquisition process will not be described in detail.

[0043] One embodiment of the application provides a digital small hydropower station fault identification method, which is configured in the data analysis module of the device and cooperates with the data acquisition module to execute the method. For details, please refer to Figure 1 The method comprises the following steps:

[0044] a) Collecting and preprocessing multi-source electrical time series data of the digital small hydropower station during operation, including generator rotor current, generator active power, reactive power and excitation voltage data.

[0045] By deploying relevant sensors in the digital small hydropower station, relevant data in the operation process of the digital small hydropower station are obtained, wherein the relevant sensors include current transformers, power meters, voltage sensors and current mutual inductors. The current transformers and power meters are deployed on the output circuits of each generator set of the digital small hydropower station, to obtain rotor current and active power and reactive power time series data of the generator of the hydropower station; the voltage sensor is connected in parallel to the excitation circuit in the excitation system of the digital small hydropower station, to obtain excitation voltage time series data in the operation process of the hydropower station. The multi-source electrical time series data collection frequency in the operation process of the above-mentioned digital small hydropower station is set to 1KHZ, and the IEEE 1588 protocol is used for clock unification.

[0046] To prevent the influence of different dimensions between different data on subsequent analysis, the application uses Z-Score algorithm to standardize the above-mentioned acquired multi-source electrical time series data of the digital small hydropower station during operation. Thus, the generator rotor current data sequence, the generator active power data sequence, the reactive power data sequence, and the excitation voltage data sequence during the operation of the digital small hydropower station can be acquired through the above-mentioned manner.

[0047] b) Based on the high-frequency frequency component of the generator rotor current data and the low-frequency frequency component of the active power data in the monitoring interval, and combined with the average value difference of the reactive power data between the fluctuation intervals and the speed of the range of the excitation voltage data in the fluctuation interval under the change of time, the environmental noise interference condition of the monitoring interval is constructed.

[0048] During the operation of the digital small hydropower station, the external environmental interference mainly comes from the hydropower station humid environment and external strong electromagnetic interference. The hydropower station humid environment will cause the sensors in the data acquisition module to be exposed to a high-humidity mist environment for a long time, causing the hygroscopicity of the insulating material, and thus causing the instability of the electrical parameters. The external strong electromagnetic interference will affect the air gap magnetic field strength of the motor and the measurement signal of the sensor, causing distortion of the electrical parameter signal, making it difficult to guarantee the extraction accuracy of the real fault and weak fault during the operation of the hydropower station, and reducing the reliability of the hydropower fault identification and diagnosis.

[0049] Specifically, during the operation of the digital small hydropower station, the humid environment interference will cause the hydropower generator rotor insulation resistance to absorb moisture and decay, causing partial discharge to cause current leakage, and then superimposing the high-frequency noise of the rotor current through inductive coupling. In addition, the humid environment will also aggravate the dielectric loss of the hydropower generator stator winding, causing the generator reactive power compensation leakage loss phenomenon to be obvious, that is, the generator reactive power produces obvious mean value rising oscillation. At the same time, the external strong electromagnetic interference will interfere with the speed signal of the hydropower station governor, causing the degree of low-frequency oscillation of the active power to increase, and causing the voltage of the excitation system of the hydropower station to produce obvious peak conditions.

[0050] Based on the above analysis, the application divides each 1h as a monitoring interval during the operation of the small hydropower station, constructs the environmental noise interference condition, and uses it to represent the parameter distortion degree caused by the external humid environment and strong electromagnetic interference during the operation of the digital small hydropower station. The generator rotor current data sequence and the active power data sequence in each monitoring interval during the operation of the digital small hydropower station are taken as inputs, and the FFT Fourier transform is used to obtain the representation of the generator rotor current data and the active power data in the frequency domain in each monitoring interval. The frequency domain of the generator rotor current data and the active power data is >0Hz and <5Hz, Each frequency component of 50Hz is recorded as a low frequency component and a high frequency component of the rotor current data and the active power data respectively.

[0051] The generator reactive power data sequence and the excitation voltage data sequence of each monitoring interval during the operation of the digital small hydropower station are taken as inputs, the AMPD (Automatic Multiscale-based Peak Detection) peak detection algorithm is used to obtain all the peaks in the generator reactive power data sequence and the excitation voltage data sequence in each monitoring interval, and the time range between each peak data and the next peak data in the generator reactive power data sequence and the excitation voltage data sequence is recorded as the fluctuation interval of the reactive power data and the excitation voltage data respectively.

[0052]

[0053] In the above formula, is the environmental noise interference condition of the i th monitoring interval during the operation of the digital small hydropower station; is the moisture interference of the i th monitoring interval during the operation of the digital small hydropower station, wherein is the variance of the maximum energy value of the generator rotor current data corresponding to all high frequency components in the i th monitoring interval, is the difference between the maximum energy value and the minimum energy value of the generator active power data corresponding to all low frequency components in the i th monitoring interval; is the electromagnetic interference of the i th monitoring interval during the operation of the digital small hydropower station, wherein is the absolute value accumulation result of the difference between the average values of the generator reactive power data corresponding to all fluctuation intervals in the i th monitoring interval, is the ratio accumulation result between the difference between the maximum excitation voltage and the minimum excitation voltage of all excitation voltage data in the i th monitoring interval in all fluctuation intervals and the time interval of the corresponding fluctuation interval; norm() is a normalization function, so that the value of is within the range of [0, 1].

[0054] The environmental noise interference condition reflects the parameter distortion degree caused by external humid environment interference and strong electromagnetic interference in each monitoring interval during the operation of the digital small hydropower station; the humid interference reflects the energy difference of the high-frequency component of the generator rotor current and the highest oscillation degree of the low-frequency component of the active power caused by the influence of the external humid environment during the operation of the digital small hydropower station; and the electromagnetic interference reflects the fluctuation drift condition of the generator reactive power mean value and the rapid change condition of the excitation voltage reaching the peak value caused by the influence of the external strong electromagnetic interference during the operation of the digital small hydropower station; when the external humid environment and strong electromagnetic interference are more serious, the hydropower station generator rotor insulation resistance absorbs moisture more seriously, the energy difference caused by the superposition of the high-frequency component of the generator rotor current is higher, and the oscillation amplitude of the low-frequency component of the active power is larger, that is, the calculation index is larger; at the same time, the generator reactive power mean value drift condition caused by the disturbance of the speed signal of the speed regulator of the hydropower station is more obvious, and the speed of the excitation voltage reaching the peak value is faster, that is, the calculation index is larger.

[0055] At this point, the environmental noise interference condition of any monitoring interval during the operation of the digital small hydropower station can be obtained by the above method.

[0056] c) Based on the trend item difference of all adjacent moments of the generator active power data in the monitoring interval and the similarity between the active power and the reactive power, and combined with the periodic item abnormality degree of the generator rotor current data and the fitting straight line slope of the excitation voltage data, the energy magnetic field fault condition of the monitoring interval is constructed.

[0057] Only with the environmental noise interference condition of each monitoring interval during the operation of the digital small hydropower station, the operation fault of the hydropower station cannot be accurately evaluated, that is, the accurate fault feature extraction of the key parameters of the digital small hydropower station is lacking, the early fault of the digital small hydropower station is ignored or the lag diagnosis phenomenon is ignored, which may lead to continuous deterioration of the hydropower station fault and exceed the critical threshold, causing a sudden accident, and seriously threatening the safe operation and flexible adjustment ability of the hydropower station. The wear of the water turbine impeller and the short circuit of the generator winding can effectively reflect the energy conversion efficiency fault condition of the hydropower station, and the electromagnetic energy conversion fault is closely related to the rotor current and excitation voltage data of the hydropower station.

[0058] Specifically, in the process of the digital small hydropower station, when the water energy conversion and the generator winding short circuit fault condition of the hydropower station are more obvious, the linear downward trend of the generator active power data caused by the decline of the energy conversion efficiency of the water turbine-generator system is more obvious, and the weak positive correlation between the active power and the reactive power data during the normal operation of the generator is broken, that is, the correlation between the active power data and the reactive power data of the generator presents a negative correlation; at the same time, the imbalance phenomenon of the rotor current data trend caused by the rotor winding short circuit and the silicon tube breakdown fault is more obvious, and the excitation voltage keeps a continuous upward trend to maintain the stability of the rotor current.

[0059] Based on the above analysis, the energy magnetic field fault condition is constructed to represent the energy conversion efficiency and the abnormality degree of the electromagnetic energy conversion process in the process of the digital small hydropower station; the active power data sequence and the rotor current data sequence of the generator in each monitoring interval in the process of the digital small hydropower station are taken as inputs, and the STL (seasonal and Trend decomposition using Loess) sequence decomposition algorithm is used to obtain the trend item of the active power data and the periodic item of the rotor current data at each moment in each monitoring interval.

[0060] The excitation voltage data sequence of the digital small hydropower station in each monitoring interval is taken as input, and the least square method is used to obtain the fitting straight line of the excitation voltage data sequence corresponding to each monitoring interval, and the slope of the fitting straight line of the excitation voltage data of each monitoring interval is calculated. When the slope of the excitation voltage data fitting straight line is higher, it means that the excitation voltage of the hydropower station in the monitoring interval is rising to maintain the stability of the rotor current of the hydropower station, and the fault caused by the electromagnetic conversion process of the rotor winding short circuit or the silicon tube breakdown fault is more serious.

[0061]

[0062] In the above formula, is the energy magnetic field fault condition of the i th monitoring interval in the process of the digital small hydropower station; is the energy conversion abnormality of the i th monitoring interval in the process of the digital small hydropower station, , wherein is the absolute value of the difference between the trend item of the active power data of each moment and the next moment in the i th monitoring interval except the last moment, is the cosine similarity between the active power data sequence and the reactive power data sequence of the i th monitoring interval; For the electromagnetic conversion abnormality of the i-th monitoring interval in the operation process of the digital small hydropower station, in one processing case of the present application, the product of the variation coefficient of the sequence composed of all the rotor current data corresponding to the period in the monitoring interval and the slope of the fitted straight line between the data sequence corresponding to the excitation voltage is taken as the electromagnetic conversion abnormality; norm() is a normalization function, so that the value range is between [0, 1].

[0063] The energy magnetic field failure condition reflects the energy conversion abnormality and the imbalance degree of electromagnetic energy conversion in the operation process of the digital small hydropower station; the energy conversion abnormality reflects the strength of the linear downward trend of active power and the imbalance degree of the correlation between active power and reactive power caused by water energy conversion and generator winding short circuit failure in the operation process of the hydropower station; and the electromagnetic conversion abnormality reflects the imbalance of the rotor current data change trend and the excitation voltage rising trend caused by rotor winding short circuit or silicon tube breakdown failure in the operation process of the hydropower station; in the operation process of the digital small hydropower station, the more serious the energy conversion and electromagnetic conversion failure conditions of the hydropower station, the more obvious the linear downward trend of the active power data of the generator of the hydropower station, and the lower the correlation between the active power data and the reactive power data of the generator of the hydropower station, that is, the calculation index becomes larger; at the same time, the more serious the period fluctuation condition of the rotor current data change trend of the hydropower station, the more obvious the excitation voltage data rising trend.

[0064] d) obtaining the hydropower station failure factor by using the ratio of the environmental noise interference condition and the energy magnetic field failure condition, and judging whether the digital small generator is faulty according to whether the hydropower station failure factor is greater than a preset threshold.

[0065] In the operation process of the digital small hydropower station, the more serious the water energy conversion and generator winding short circuit failure of the digital small hydropower station, and the more obvious the rotor winding short circuit and silicon tube breakdown phenomenon, the more significant the operation failure condition of the hydropower station, when the key parameters are less affected by the humid environment and electromagnetic interference of the hydropower station, and the higher the energy conversion efficiency abnormality and the imbalance degree of electromagnetic energy conversion of the hydropower station.

[0066] Based on the above analysis, the present application constructs a hydropower station failure factor for representing the failure severity in each monitoring interval in the operation process of the digital small hydropower station; which can be obtained through the environmental noise interference condition and the energy magnetic field failure condition.

[0067] Specifically, in one processing case of the present application, the ratio of the energy magnetic field failure condition and the environmental noise interference condition of each monitoring interval in the operation process of the digital small hydropower station is taken as the hydropower station failure factor.

[0068] When the hydropower station failure factor is greater, it indicates that the digital small hydropower station is slightly disturbed by environmental noise in the monitoring interval, and the energy conversion efficiency of the hydropower station operation and the abnormal degree of electromagnetic energy conversion are higher.

[0069] At this point, the hydropower station failure factor of any monitoring interval in the operation process of the digital small hydropower station can be obtained by the above method.

[0070] The hydropower station failure factors of all the generators in each monitoring interval in the operation process of the digital small hydropower station are taken as inputs, the maximum inter-class variance method is used to obtain the segmentation threshold R, and the generators in each monitoring interval in the operation process of the digital small hydropower station whose hydropower station failure factors are greater than or equal to the segmentation threshold R are recorded as failure generators. Since the maximum inter-class variance method is a known technology, the specific obtaining process will not be described in detail. In other embodiments, a preset threshold can also be used as a failure judgment threshold for judging the digital small hydropower station.

[0071] Other embodiments of the application will be apparent to those skilled in the art from consideration of the specification and practice of the application disclosed herein. It is intended that the application cover any variations, uses or adaptations of the application following, in general, the principles of the application and including such departures from the present disclosure as come within known or customary practice in the art to which the application pertains.

[0072] It should be understood that the application is not limited to the precise construction that has been described above and shown in the accompanying drawings, and that various modifications and changes can be made by those skilled in the art without departing from the scope of the application.

Claims

1. A digital small hydropower station fault identification method, characterized in that, The method comprises the following steps: a) Collect and pre-process the multi-source electrical time series data of the digital small hydropower station during operation, including generator rotor current, generator active power, reactive power, and excitation voltage data; b) Determine the damp interference of the monitoring interval based on the high-frequency frequency components of the generator rotor current data and the low-frequency frequency components of the active power data in the monitoring interval; determine the electromagnetic interference of the monitoring interval based on the difference between the average values of the reactive power data in the fluctuation interval and the speed of the range of excitation voltage data in the fluctuation interval under the change of time; and take the normalized value of the product of the damp interference and the electromagnetic interference as the environmental noise interference condition of the monitoring interval; The determination method of the damp interference of the monitoring interval is to take the product of the variance of the maximum energy value of all high-frequency frequency components of the generator rotor current data in the monitoring interval and the difference between the maximum energy value and the minimum energy value of all low-frequency frequency components of the active power data as the damp interference of the monitoring interval; The determination method of the electromagnetic interference of the monitoring interval is to take the product of the absolute value of the difference between the average values of the reactive power data in all fluctuation intervals in the monitoring interval and the ratio between the difference between the maximum excitation voltage and the minimum excitation voltage of the excitation voltage data in all fluctuation intervals in the monitoring interval and the corresponding time interval of the fluctuation interval as the electromagnetic interference of the monitoring interval; c) Determine the energy conversion abnormality of the monitoring interval based on the trend item difference of all adjacent time points of the generator active power data and the similarity between the active power and the reactive power in the monitoring interval; determine the electromagnetic conversion abnormality of the monitoring interval based on the periodic item abnormality degree of the generator rotor current data and the fitting straight line slope of the excitation voltage data; and take the normalized value of the product of the energy conversion abnormality and the electromagnetic conversion abnormality as the energy magnetic field fault condition of the monitoring interval; The determination method of the energy conversion abnormality of the monitoring interval is to take the absolute value of the difference between the trend items of all adjacent time points of the generator active power data in the monitoring interval as the index of the exponential function with the natural constant as the base, and take the ratio between the calculation result of the exponential function and the cosine similarity between the active power and the reactive power as the energy conversion abnormality of the monitoring interval; The determination method of the electromagnetic conversion abnormality of the monitoring interval is to take the product of the coefficient of variation of the periodic item sequence of the generator rotor current data and the fitting straight line slope of the excitation voltage data as the electromagnetic conversion abnormality of the monitoring interval; d) Obtain the hydropower station fault factor by using the ratio of the environmental noise interference condition and the energy magnetic field fault condition, and determine whether the digital small generator is faulty according to whether the hydropower station fault factor is greater than a preset threshold.

2. A digital small hydropower plant fault identification method according to claim 1, characterized in that, The high-frequency frequency component is a frequency component greater than or equal to 50 Hz after the data is converted to the frequency domain; the low-frequency frequency component is a frequency component greater than 0 Hz and less than 5 Hz after the data is converted to the frequency domain.

3. A digital small hydropower plant fault identification method according to claim 1, characterized in that, The fluctuation interval is the time range between adjacent peak data in the data in the monitoring interval.

4. A digital small hydropower station fault identification device, characterized in that, The device comprises: a data acquisition module for acquiring and preprocessing multi-source electrical time series data of the digital small hydropower station during operation, to realize step a) as claimed in claim 1; a data analysis module connected with the data acquisition module, for executing the method as claimed in claim 1 to output a fault analysis result; a data transmission module for transmitting data between modules; a human-computer interaction module for displaying the fault analysis result.

Images