61 results about "GENERATOR MALFUNCTION" patented technology

The invention discloses a low sample hydro-generator fault diagnosis method based on transfer learning. The method comprises the following steps: collecting a current signal, a vibration signal, a temperature signal and a voiceprint signal; executing preprocessing; performing label labeling, and dividing the data into a source domain data set and a target domain data set; operating parameters of the hydro-generator are collected, processed and combined into working condition feature vectors; training a source domain base model on the source domain data set to generate a pre-training parameter set; establishing a physical constraint module, and binding the physical constraint module with the parameter updating process of the source domain base model; initializing a transfer learning model, and calling a physical constraint module to apply constraint to generate a physically constrained transfer learning model; performing increment fine adjustment; and outputting a diagnosis result vector through the transfer learning model after increment fine tuning. According to the method, transfer learning and physical constraints are combined, hydro-generator fault diagnosis is optimized through increment fine tuning, and the method has the advantages of high precision, small sample adaptability and physical consistency verification.

A generator fault identification method, device, product and medium relate to the technical field of generator fault category identification. The method comprises the following steps: acquiring a vibration signal and a noise signal; performing multi-mode decomposition processing on the vibration signal and the noise signal to obtain a first sub-mode component and a second sub-mode component; determining a vibration signal sequence based on each first sub-mode component and the vibration signal; determining a noise signal sequence based on each second sub-mode component and the noise signal; determining a vibration characteristic matrix based on the vibration signal sequence; determining a noise feature matrix based on the noise signal sequence; determining a first recognition result and a first confidence coefficient based on the vibration feature matrix; determining a second recognition result and a second confidence coefficient based on the noise feature matrix; when the identification results are inconsistent, determining a fault identification result according to the confidence coefficient; and when the identification results are consistent, determining the first identification result as a fault identification result. The method is advantaged in that generator fault identification accuracy is improved.

The invention discloses a multi-machine common-bus system grounding fault arc extinction and property identification method based on machine end voltage regulation and control. The method comprises the following steps: S1, machine end arc extinction; s2, low-frequency injection; s3, low-frequency component measurement; s4, grounding transition resistance calculation: through the low-frequency components of the terminal voltage and the current of each generator at the first moment, calculating the ground impedance Zi1 during a grounding fault; calculating the ground impedance Zi2 after the fault is stable through the low-frequency components of the terminal voltage and the current at the second moment; calculating the grounding transition resistance Rgi of each generator when any generator fails by integrating the Zi1 and the Zi2; s5, fault identification: if Rgi is greater than or equal to a set threshold value, identifying the fault as an instantaneous fault, and if Rgi is less than the set threshold value, identifying the fault as a permanent fault; if a generator is identified as a permanent fault, the generator can be identified as a fault unit. According to the method, the problem of poor economical efficiency of existing fault arc extinction and fault identification in a multi-machine common-mother system can be solved.

The invention provides a permanent magnet generator fault monitoring method and device, electronic equipment and a storage medium, and relates to the technical field of motor fault detection, and the method comprises the steps: obtaining the historical operation data of a target permanent magnet generator; according to the historical operation data, establishing a motor model of the target permanent magnet generator in each working condition state; determining time domain features and frequency domain features of the target permanent magnet generator in each working condition state based on the motor model; and performing fault monitoring on the target permanent magnet generator based on the time domain feature and the frequency domain feature. Therefore, compared with the prior art that data acquisition is carried out through installation of a sensor, high-frequency voltage is injected through an inverter, then corresponding high-frequency inductance is calculated, and then identification analysis of specific harmonic waves is completed, the scheme in the invention can realize continuous monitoring and diagnosis, fault occurrence can be effectively prevented, and the fault diagnosis accuracy is improved. The dependence on sensors and monitoring conditions is small, and the operation efficiency and economic benefits of the whole wind power plant can be remarkably improved.

The invention belongs to the technical field of nuclear power, and particularly relates to a steam turbine generator fault diagnosis method, system and device. The steam turbine generator fault diagnosis method provided by the invention can realize fusion of the shaft voltage and vibration information of the steam turbine equipment, improves the comprehensiveness and accuracy of fault diagnosis of the steam turbine equipment, has important practical and economic values, can improve the safe and stable operation level of the generator, and is suitable for popularization and application. And more effective technical support is provided for operation and maintenance of a power system.

The invention provides a wind driven generator fault diagnosis method, device and equipment and a storage medium, and relates to the technical field of intelligent operation and maintenance of wind power generation equipment. The method comprises the following steps: constructing a thermal network model of the wind driven generator, and obtaining a transient heat transfer equation and a first thermal conductance matrix and a first heat source intensity vector of the wind driven generator in a healthy state based on the thermal network model; according to the first thermal conductance matrix, the first heat source intensity vector and the current temperature of each temperature measurement point of the wind driven generator, performing inversion based on a transient heat transfer equation to obtain a second thermal conductance matrix and a second heat source intensity vector; obtaining a first fault feature matrix according to the second thermal conductivity matrix and the first thermal conductivity matrix, and obtaining a second fault feature matrix according to the second heat source intensity vector and the first heat source intensity vector; and obtaining a fault diagnosis result of the wind driven generator according to the first fault feature matrix, the second fault feature matrix and a preset fault diagnosis model. According to the invention, high-precision fault diagnosis of the wind driven generator can be realized.

This invention relates to a microwave combustion exhaust gas treatment device, comprising: an ignition tube, a combustion tube, a resonant housing, and a microwave generator. The microwave generator is fixedly connected to and communicates with the resonant housing, the resonant housing is fixedly connected to and communicates with the ignition tube, and the lower part of the ignition tube is fixedly connected to and communicates with the combustion tube. When treating small-flow exhaust gas, this invention allows the microwave generator to be used alone, saving energy and reducing costs. When treating larger-flow exhaust gas, the combined microwave and combustion treatment provides greater processing capacity, and the energy consumption and cost of combined treatment are lower than using natural gas combustion alone, thus reducing the energy consumption and cost of exhaust gas treatment. When the microwave generator malfunctions or other external microwave malfunctions occur, the exhaust gas flow into the ignition tube and combustion tube increases. By increasing the amount of natural gas and air, the emergency exhaust gas treatment capacity is improved, preventing the exhaust gas from being untreated and causing harmful components to exceed emission standards.

The invention relates to the technical field of nuclear power plants, in particular to a fault monitoring method and system for an emergency diesel generator, electronic equipment and a medium. Equipment health state parameters, equipment operation condition parameters and equipment operation environment parameters need to be monitored in real time firstly; performing fault threshold analysis based on the equipment health state parameters, the equipment operation condition parameters and the equipment operation environment parameters, and determining a generator fault threshold; carrying out measuring point signal acquisition on the emergency diesel generator to obtain an equipment measuring point acquisition signal; performing signal analysis on the equipment measuring point acquisition signal based on the generator fault threshold to obtain a signal analysis result; and if the signal analysis result meets a preset fault judgment condition, executing a fault response operation. The core of the method is that an alarm threshold value is converted from a fixed value to a variable dynamically adjusted along with the equipment state, the operation condition and the environmental factor, so that the accuracy and the reliability of fault early warning are remarkably improved.

The invention provides a vehicle generator fault detection method, electronic equipment and a storage medium. The method comprises the steps that multiple types of operation state signals of a vehicle are obtained, wherein the multiple types of operation state signals at least comprise a first type of operation state signals with data missing and a second type of operation state signals with complete data; based on correlation analysis on the multiple types of operation state signals, determining correlation types among the multiple types of operation state signals; based on the second type of operation state signals, performing interpolation processing on the first type of operation state signals by using a model corresponding to the correlation type to obtain update state signals corresponding to the multiple types of operation state signals; and determining a fault detection result of the generator according to the working state of the generator in the vehicle under the condition that the update state signal meets the fault judgment condition. By means of the method, missing data in the vehicle state signal can be accurately determined, and therefore fault detection can be accurately conducted on the vehicle generator.

The invention relates to the technical field of generator fault prediction, in particular to an extreme weather power generation equipment fault prediction method based on a time sequence large model, and the method comprises the steps: synchronously collecting multiple parameters; judging an abnormal event; identifying trend types; risk level assessment; adjusting a dynamic threshold value; and risk sequence prediction and early warning. A plurality of parameters which are closely related to structural load, electrical disturbance and thermal-lubrication states are introduced through the three-axis vibration peak value in the middle of a blade, the oil viscosity of a gearbox, the current harmonic wave of a stator, the starting torque increment of a yaw system, the wind speed change rate of the top of a cabin and the temperature rise rate of a main shaft bearing; full-chain monitoring of the operation state of the wind turbine generator in the strong wind extreme weather is achieved, and the problems that due to incomplete monitoring parameter coverage and fixed threshold value setting, lubrication abnormity trend recognition lags behind in the extreme weather, fault early warning is not timely, and the unit shutdown loss is aggravated are effectively solved.

The invention belongs to the technical field of generator fault detection, and particularly discloses a generator anomaly detection method and system, and the method comprises the following steps: collecting the operation parameters of a generator and the parameters of the environment where the generator is located, fusing the characteristics of different modes, and obtaining the fault detection result of a single generator; performing cross-equipment cooperative detection to obtain operation parameters of the plurality of generators and environmental parameters of the plurality of generators, and modeling working parameter data and mutual influence among the plurality of generators; extracting shared fault mode features from the data of different generators to perform cross-device joint modeling to obtain a detection result of each generator; and synthesizing the detection result of the single cross-modal generator and the detection results of all cross-equipment generators to obtain a final detection result. By adopting the technical scheme, on the basis of cross-modal feature fusion and cross-device cooperative detection, comprehensive analysis is performed from multiple dimensions, and the accuracy and response speed of fault detection are improved.

The invention discloses a semi-direct-drive driving structure and method for a marine generator, and relates to the technical field of marine generators. The expansion sleeve connecting flange is used for being fixedly connected with a transmission main shaft of a ship main engine; the sectional type main transmission gear and the expansion sleeve connecting flange are mounted and locked; the shaft extension end of the generator is connected with the coupler; the end, away from the generator, of the coupler is connected with the coupler, and the coupler is used for achieving signal transmission and isolation and can disconnect power connection when the generator breaks down. One end of the coupler away from the coupling is connected with the pinion; the small gear is meshed with the sectional type main transmission gear in a tooth contact mode so that power can be transmitted from the sectional type main transmission gear to the coupler. All the parts are designed in a separated mode, when the whole machine is damaged or needs to be maintained, all the parts can be rapidly disassembled, the whole structure does not need to be disassembled on a large scale, fault parts can be independently replaced, manpower and material resource cost is greatly saved, and operation and maintenance efficiency is remarkably improved.

A data analysis system comprises an artificial intelligence of the machine learning type. The data analysis system collects reference data relating to measurements made and recorded during previous flights of aircraft. A training of the artificial intelligence of the machine learning type is implemented by virtue of reference data, with a first classification run for training to detect potential malfunctions suffered by integrated drive generators and a second classification run for training to determine causes of the malfunctions, where relevant. After using in production, the data analysis system collects data to be analyzed relating to measurements made and recorded during flights of the aircraft comprising an integrated drive generator to be monitored, and uses the artificial intelligence of the machine learning type to predict a potential malfunction and, where relevant, predict its cause using data to be analyzed.

The application discloses an online fault diagnosis method for a hydro-generator based on a potential interval distance of a through screw, relates to the technical field of generators, and is suitable for detecting a stator winding ground fault. Voltage collection devices are additionally arranged on a plurality of through screws of the hydro-generator, real-time induced potential is acquired, a historical data matrix is established, and a data buffer matrix is updated at interval time; the data buffer matrix is divided into a data evaluation matrix, a fundamental wave and an odd harmonic phase are extracted through Fourier decomposition, a harmonic phase matrix is constructed, a phase difference matrix is obtained by calculating normalized phase differences of adjacent columns, a quartile distance is solved after being converted into a column vector array, a dynamic quartile distance array is formed, abnormal limit values are determined by combining a box plot principle and an error level of the voltage collection device, and faults are determined by traversing the array. The method does not need additional sensing devices and traditional threshold setting, is convenient to operate, has low implementation difficulty, and can accurately and timely identify faults.

The invention discloses a hydro-generator rotor winding turn-to-turn short circuit on-line detection signal processing method, which relates to the technical field of synchronous generator fault diagnosis and comprises the following steps of: A, acquiring an induced voltage of a measuring coil; b, preprocessing the acquired induced voltage; c, determining a corresponding relation between the pulse voltage and the magnetic pole in a rotation period by using the phase discrimination signal; d, extracting the extreme value of the pulse voltage waveform corresponding to each magnetic pole; e, the fundamental wave amplitude A1 of the current waveform and the fundamental wave amplitude A0 of the historical waveform when the turn-to-turn short circuit does not occur are obtained through Fourier transform, and the current induced voltage pulse extreme value array is multiplied by A0 / A1; f, if the absolute value of a certain extreme value in the adjusted current extreme value array is smaller than the absolute value of the numerical value corresponding to the historical extreme value array, and the difference between the absolute value of the historical extreme value and the absolute value of the current extreme value exceeds a set threshold value, it is judged that the turn-to-turn short circuit fault occurs in the magnetic pole winding corresponding to the extreme value. The method is good in anti-interference performance and high in diagnosis sensitivity.

The utility model provides a thermal tripping structure of an aero-generator. The thermal tripping structure comprises an input shaft connected with a generator and a power source; the end part of the input shaft is in transmission connection with the front end of the disengagement assembly in a disengagement manner through an end surface transmission structure, the disengagement assembly is sleeved in the hollow shaft, the disengagement assembly and the hollow shaft are in transmission through a spline transmission structure, and the hollow shaft is sleeved in a generator rotor structure; the rear end of the disengaging assembly is connected into a lubricating oil way. A disengaging spring is arranged in the disengaging assembly and used for abutting against the end face transmission structure to complete transmission. When the internal temperature of the oil-cooled generator is too high, the spring is triggered to act, so that the transmission shaft is separated from the input shaft, the generator is mechanically separated from the power source, the generator can be protected from being burnt out due to faults, and the power source can be protected from being damaged due to the faults of the generator.

The invention discloses a generator fault detection method and system based on a photoelectric encoder, and relates to the technical field of generator detection, the method obtains the initial reason of generator vibration from the source through the photoelectric encoder, intercepts all single-period sinusoidal signals and carries out unified processing at the same time, and the method considers that under the condition of slight influence, the single-period sinusoidal signals are subjected to single-period sinusoidal signal processing. The optical-electricity encoder may have the phenomenon of missing detection or drifting due to slight influence of environmental interference, electrical factors and the like, the purpose of calibrating signal data in batches at the same time is achieved by removing outliers from all coordinate points based on the same abscissa, and the signal data of the same waveform serve as contrast reference. The method reduces the misjudgment probability, achieves the prediction function of signal data through the machine learning machine, enables the generator to be predicted when the generator does not generate substantial mechanical damage, and further guarantees that the service life of the generator is not affected.

The invention discloses a steam turbine generator state detection method based on a vibration signal and a related device, and relates to the technical field of generator fault detection, and the method comprises the following steps: collecting vibration information data of a steam turbine generator, and carrying out the preprocessing; performing insulation fatigue calculation according to the preprocessed vibration information data to obtain an insulation fatigue coefficient; and performing grading evaluation on the state of the steam turbine generator according to the insulation fatigue coefficient to obtain a state detection result. The problem that in the prior art, measurement needs to be carried out under shutdown or specific stable working conditions, and continuous online monitoring is difficult to achieve can be solved.

The invention discloses a safety control method and system under fault and denial service of a wind driven generator, and belongs to the technical field of wind driven generators. The method comprises the following steps: constructing a dynamical model considering an actuator fault and a sensor denial service attack; constructing a signal compensator based on a rotor speed measurement value before denial service attack, and estimating the rotor speed by combining a state observer of a neural network; on the basis of the tracking deviation of the optimal rotor rotating speed and the actual rotating speed, performance deviation is preset through finite time, and the objective function is optimized by solving a Hamilton-Jacobi-Bellman equation; and generating a torque control law based on the estimated value of the state observer and the actor-evaluator network updating law, thereby determining a safety control signal under the actuator fault and sensor denial service attack of the wind driven generator, and realizing the maximum power point tracking of the wind driven generator and the hybrid fault tolerance and toughness control of the wind driven generator.

This invention discloses a method and apparatus for identifying stator winding grounding faults in hydro-generators, relating to the field of hydro-generator fault identification technology. The method includes acquiring the zero-sequence current signal during a single-phase grounding fault in the hydro-generator; using the minimization of envelope entropy as the fitness function, the IDBO algorithm is used to adaptively optimize the number of modes and penalty factor in variational mode decomposition (VMD) to obtain the optimal parameter combination; wherein the IDBO algorithm is obtained by fusing Logistic-Tent chaotic mapping, exponential decay convergence factor, and random perturbation mechanism; based on the zero-sequence current signal and the optimal parameter combination, signal decomposition processing is performed to obtain multiple intrinsic mode functions (IMF) components; the three components with the highest correlation coefficients are selected as effective fault features, their sample entropy is calculated, and the sample entropies are combined into a feature vector representing the fault state; the feature vector is used as an input layer variable and imported into a support vector machine for fault identification.

The invention relates to the technical field of generator fault detection, and particularly discloses a wind driven generator fault tracing method and device, and the method comprises the steps: obtaining the fault type, fault position and fault time of a wind driven generator, and determining a data collection range according to the fault type, fault position and fault time; querying data in a data acquisition range in a database of the wind driven generator, and constructing and simplifying a sample set; the simplified sample sets of the same fault type are counted, the counted sample sets are compared, and feature positions are determined; constructing a mapping relation between the feature position and the fault type, and synchronously determining the mapping strength; and when the fault type is detected, reading the feature position according to the mapping strength, and generating a detection form. According to the method, when the fault type is faced, the related position can be quickly positioned, the detection sequence is determined according to the importance degree of the related position, the global traceability process is converted into the sequential traceability process, and the traceability efficiency is greatly improved.

The application discloses a kind of leading instability machine group automatic marking method, system, device and medium, belong to electric power technical field.The method includes: obtaining input sample, judge whether input sample is transient instability sample, if yes, leading instability machine group is marked;For transient instability sample, obtain the power angle curve set of all generators {δ i (t)}, according to power angle curve set {δ i (t)} Similarity matrix ρ is obtained;Based on similarity matrix ρ, after AP clustering, get the homology machine group clustering result set G AP ={G1,G2,…,G K};According to homology machine group clustering result set G AP Marking result is obtained.The application is dynamically obtained after each generator fault, and clustering mode is automatically carried out generator homology grouping, and the most leading machine group is regarded as leading instability machine group, and leading instability machine group of mass sample can be marked spontaneously and in batches, to provide fine-grained generator level stability index for data-driven stability evaluation model.

The invention belongs to the field of generator controllers, and particularly relates to a generator controller under-voltage protection fault self-detection isolation method and system. The method comprises the following steps: acquiring three-phase voltage of a voltage regulating point; when any phase voltage is lower than the under-voltage threshold value, setting an under-voltage fault mark; if the under-frequency and over-current faults do not exist and the excitation relay is switched on, time delay decreasing is carried out, and an under-voltage fault protection mark is set when time delay is finished; a fault source is then located, including determining a feeder fault, a generator fault, or a generator controller fault based on the specific three-phase voltage. Through multi-parameter fusion diagnosis and logic judgment, quick and accurate positioning of the under-voltage fault is realized, the fault is isolated to the LRU level, and the troubleshooting efficiency and the airplane attendance rate are remarkably improved.

The invention relates to a method and system for predicting a fault and / or remaining life of a generator. The method comprises the following steps: constructing a generator simulation model; for the actual vehicle with the generator fault record and the target vehicle with the generator fault to be predicted, according to the historical data of the generator rotating speed obtained through the historical data of the engine rotating speed, the historical data of the generator output current are obtained through the generator simulation model; constructing a prediction model by means of machine learning based on historical data of a generator fault, a generator output voltage, an engine speed and a generator output current of an actual vehicle; and inputting the historical data of the generator output voltage, the engine speed and the generator output current of the target vehicle into the prediction model to predict the fault and / or residual life of the generator of the target vehicle. According to the invention, the fault and / or residual life of the generator can be conveniently predicted at low cost, so that stable, safe and efficient operation of the engineering machinery is realized.

The invention relates to the field of power generation equipment maintenance, in particular to a power generation terminal emergency maintenance scheme generation method based on artificial intelligence, and the method comprises the steps: determining a load disturbance excitation value through obtaining the load characteristics of a power grid connected with a generator set, and carrying out the interference disturbance test of the generator set; according to the method, the interference disturbance characteristics of the remaining generators in the limited time are obtained, the interference state of the target generator influenced by system interference is comprehensively evaluated, and then specific dimension screening is carried out on the multi-dimensional operation data in the strong interference time period. The influence of interference between the power grid and the generator sets is effectively isolated, the influence of load change of the external power grid and complex mutual interference between multiple generator sets on generator fault judgment when the generator works in a strong interference state is reduced, the detection error and the fault misjudgment rate are remarkably reduced, the fault is shortened, and the follow-up maintenance decision accuracy is improved.

The application provides a power system operation reliability evaluation method and device considering extremely cold weather, which comprises the following steps: obtaining weather data; calculating the icing amount on the wind turbine blade according to the weather data, and determining a wind turbine operation reliability model based on the icing amount; constructing a power transmission line operation reliability model based on ice wind load; constructing a generator operation reliability model based on the functional relationship between the generator failure rate and the operating state; sampling the working states of the wind turbine, the power transmission line and the generator by using the Monte Carlo method to establish a system state set; calculating the reliability index and the wind power curtailment under each working state in the system state set by using an optimal load shedding model; evaluating the operation reliability of the power system according to the reliability index to obtain a reliability evaluation result. The application solves the problem in the prior art that there is a lack of system operation reliability model under extremely cold weather for the current high-proportion new energy system.

This application discloses a vehicle control method, a vehicle, and a computer-readable storage medium. The method includes: acquiring generator temperature, inverter temperature, and vehicle operating parameters, wherein the vehicle operating parameters include motor water pump duty cycle, cooling fan duty cycle, generator output torque, engine status, and vehicle drive mode; the engine status includes running state and stopped state; and the vehicle drive mode includes parallel drive mode and pure electric drive mode; determining a generator fault level based on the generator temperature, inverter temperature, and a preset temperature threshold, wherein the generator fault level is used to determine the degree of generator over-temperature; determining an over-temperature protection strategy based on the generator fault level and vehicle operating parameters; and performing over-temperature protection on the generator based on the over-temperature protection strategy. This application solves the technical problems of incomplete protection, unreasonable limp-out strategies, and low system reliability in related technologies when performing generator over-temperature protection.

This invention belongs to the field of generator fault detection technology, and provides a winding insulation detection method and system considering axial eccentricity faults in generators. The method includes: building a simulation system to independently simulate static eccentricity, dynamic eccentricity, and tilting eccentricity in the generator axial direction; deploying ultrasonic sensors on the generator winding insulation to collect ultrasonic signals generated by wear defects in the winding insulation; extracting the arrival time of the ultrasonic signals using an energy detection method and calculating the arrival time difference, and combining this with a genetic algorithm to locate the wear defects in the winding insulation; performing synchronous sampling and filtering preprocessing on the ultrasonic signals to achieve a quantitative assessment of the wear state of the winding insulation. The present invention is reliable and easy to implement, and the proposed detection method has the characteristics of strong anti-electromagnetic interference capability and high positioning accuracy, and can solve the axial eccentricity simulation problem of vertical hydro generators with a small-volume, low-cost, and practical device.

This invention belongs to the field of generator fault monitoring technology, and provides an online monitoring system and method for inter-turn short circuits in generator rotors based on the detection coil method. It employs a flexible substrate and a planar helical detection coil integrated on it as a flexible magnetic flux sensor, completely eliminating the safety hazards of traditional rigid magnetic flux sensors. The flexible magnetic flux sensor requires no rotor removal or drilling; it is fixed to the generator slot wedge or stator teeth using a high-temperature adhesive, significantly reducing the installation time per unit. This invention effectively filters out main magnetic flux interference and electromagnetic noise through an intelligent algorithm combining improved wavelet threshold function denoising and a BP neural network, featuring high safety, convenient installation, and high detection accuracy.

The present invention discloses an online detection signal processing method for inter-turn short circuits in a hydro-turbine generator rotor winding, relating to the technical field of synchronous generator fault diagnosis. The method comprises the following steps: A. obtaining an induced voltage from a measuring coil; B. preprocessing the obtained induced voltage; C. using a phase-locked signal to determine the correspondence between a pulse voltage and a magnetic pole within a rotation cycle; D. extracting the extreme values of the pulse voltage waveform corresponding to each magnetic pole; E. obtaining the fundamental amplitude A1 of the current waveform and the fundamental amplitude A0 of a historical waveform when no inter-turn short circuit has occurred through Fourier transform, and multiplying the array of current induced voltage pulse extreme values by A0 / A1; and F. if the absolute value of a particular extreme value in the adjusted current extreme value array is less than the absolute value of the corresponding value in the historical extreme value array, and the difference between the absolute value of the historical extreme value and the absolute value of the current extreme value exceeds a set threshold, then determining that an inter-turn short circuit fault has occurred in the magnetic pole winding corresponding to the extreme value. The method has excellent anti-interference performance and high diagnostic sensitivity.