A method, device, equipment and medium for intelligent diagnosis of defects of hydroelectric relay protection

Abstract

The present application relates to the field of water and electricity relay protection, and specifically discloses a kind of water and electricity relay protection defect intelligent diagnosis method, device, equipment and medium, the method comprises: collecting the topology of hydropower station and equipment parameters, constructs node admittance matrix and divides fault point;Establishes the equivalent model of transient potential variation of hydroelectric generating set, introduces dynamic impedance compensation factor to solve short-circuit current and calculates contribution rate matrix;Carry out frequency domain analysis to extract characteristic frequency component, when waveform distortion or low-frequency oscillation in strong coupling area exceeds threshold value, select compensation model to correct sensitivity coefficient, draw protection characteristic curve and identify cooperation interface point.The present application solves the problem of large deviation in fault current calculation and sensitivity distortion caused by ignoring transient potential attenuation and harmonic, low-frequency oscillation in prior art through dynamic impedance compensation and frequency domain compensation model, realizes intelligent and accurate diagnosis of water and electricity relay protection defects.

Description

Technical Field

[0001] This invention belongs to the field of hydropower relay protection, specifically a method, device, equipment, and medium for intelligent diagnosis of defects in hydropower relay protection. Background Technology

[0002] Hydropower relay protection is a key technology for ensuring the safe operation of hydropower units and power grids. Currently, the setting calculations for hydropower plant relay protection are usually based on the power frequency periodic component, using the nodal admittance method or short-circuit current calculation software to solve for the fault current, and judging whether the protection meets the requirements based on the sensitivity coefficient (the ratio of fault current to protection start current).

[0003] However, during the transient fault process of hydropower units, the short-circuit current exhibits rapid decay characteristics due to factors such as the excitation system response, governor mechanical inertia, and water hammer effect, accompanied by abundant second and third harmonics and low-frequency oscillation components. Existing technologies, when calculating short-circuit current, typically treat the hydropower unit as an equivalent constant potential source or use a simplified subtransient reactance model, neglecting the dynamic process of transient potential decay over time. In sensitivity analysis, calculations are based solely on the fundamental current amplitude, ignoring the impact of harmonic distortion and low-frequency oscillations on the actual sensing quantity of the protection. This deficiency leads to a large deviation between the calculated and actual fault current values, resulting in distorted sensitivity coefficients and consequently unreasonable protection settings: failure to operate may occur when waveform distortion is severe, and maloperation may occur in the selective coordination critical region, seriously affecting the safe and stable operation of hydropower plants and the power grid. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides an intelligent diagnostic method, device, equipment, and medium for hydropower relay protection defects, thereby resolving the issues in the prior art where neglecting transient potential decay and the influence of harmonics and low-frequency oscillations leads to large deviations in fault current calculations, distorted protection sensitivity, and unreasonable setting values.

[0005] This invention provides an intelligent diagnostic method for defects in hydropower relay protection, comprising the following steps:

[0006] Step S1: System Modeling and Fault Point Identification

[0007] Collect the topology and equipment parameters of the hydropower station to construct the node admittance matrix; calibrate the electrical distance between each relay protection device and the grid connection point of the hydropower unit to generate a fault point distribution sequence; divide the protection monitoring area based on the electrical distance; collect real-time head signals and unit output signals, and determine the dynamic evaluation coefficient of each fault point according to the head and output range;

[0008] Step S2: Short-circuit current calculation

[0009] An equivalent model of transient electromotive force change of hydropower unit is established, and the transient electromotive force decay curve is obtained by solving a set of differential equations. This model is combined with the nodal admittance matrix, and a dynamic impedance compensation factor is introduced. The dynamic impedance compensation factor is based on 1, plus a correction term that is proportional to the square of the real-time short-circuit current and decays exponentially with time. The proportional coefficient and time constant of the correction term are determined according to the current head interval. The fault current is solved using a nodal algorithm. Fault points are classified according to electrical distance and fault current characteristics. A fault current contribution rate matrix is ​​established, and the actual fault current amplitude of each protection device is calculated in combination with the total short-circuit current.

[0010] Step S3: Sensitivity Analysis and Characteristic Curve Generation

[0011] The comprehensive weight of each fault point is calculated based on the fault current contribution rate matrix and dynamic evaluation coefficient, and the coupling region is divided based on the comprehensive weight. Frequency domain analysis is performed on the fault current to extract characteristic frequency components. The initial sensitivity coefficient of each protection device is calculated. For fault points located in the strong coupling region and whose waveform distortion rate or low-frequency oscillation amplitude exceeds their respective preset thresholds, a compensation model is selected according to the degree of waveform distortion, and a compensation factor is calculated to correct the sensitivity coefficient. The protection characteristic curve is plotted with electrical distance as the abscissa and the corrected sensitivity coefficient as the ordinate, and the characteristic points on the curve are identified as protection coordination interface points.

[0012] Step S4: Defect Diagnosis and Tuning Optimization

[0013] Based on the segmented results of the protection characteristic curve, perform differentiated diagnosis to identify defect types including insufficient dynamic compensation, insufficient sensitivity, and selective defects; verify the coordination relationship between adjacent protections; if the coordination requirements are not met, iteratively adjust the setting value until the requirements are met or the preset number of iterations is reached, and output a defect diagnosis report and optimized setting value.

[0014] Preferably, the criterion for dividing the head interval in step S1 is: low head is... Rated head is High water head is ,in The rated head; the standard for dividing the output range is: light load is , middle load is Overload ,in The rated output is used; the dynamic evaluation coefficient is calculated in advance using historical simulation or measured data to determine the average short-circuit current contribution rate under the combination of head interval and output interval, and is obtained by bilinear interpolation based on the current head and output during real-time diagnosis.

[0015] Preferably, the differential equations of the equivalent model of the transient electromotive force change of the hydropower unit in step S2 are as follows:

[0016]

[0017]

[0018] in, For transient potential, The excitation time constant is This is the excitation voltage. It is a direct-axis synchronous reactance. For direct-axis transient reactance, For direct-axis current, For rotational inertia, The rotor angular velocity, For synchronous angular velocity, For mechanical torque, For electromagnetic torque, The damping coefficient is used; the transient potential decay curve is obtained through numerical solution, and then the fault current is calculated.

[0019] Preferably, the dynamic impedance compensation factor in step S2 The calculation formula is:

[0020]

[0021] in, This is the compensation coefficient, and its value range is... ; To compensate for the time constant, the value range is: ; For real-time short-circuit current, The reference current is taken from the rated current of the hydropower unit; determined according to the head range. and Specific values: at low head , ; at rated head , High water head , .

[0022] Preferably, the fault point classification method in step S2 is as follows: dynamic coupling fault points are those with an electrical distance of not more than [missing information]. And the fault current decay time constant is no greater than The fault point; the weak power supply fault point is the electrical distance greater than Fault points with fault current amplitude less than 0.05 times the rated current; other fault points are those between these two values; the fault current contribution rate matrix is ​​established using the recursive least squares method, and the actual fault current amplitude of each protection device is calculated according to... Calculation, where For contribution rate, For the first The total short-circuit current at each fault point.

[0023] Preferably, the comprehensive weight in step S3 The calculation formula is: ,in Contribution rate to short-circuit current For dynamic evaluation coefficients, The total number of protection devices; the standard for dividing the coupling area is: strong coupling area. , weak coupling region transition zone The characteristic frequency components include the fundamental amplitude. Second harmonic amplitude Third harmonic amplitude and the amplitude of low-frequency oscillation components The amplitude of the low-frequency oscillation component is obtained by extracting the maximum envelope value after bandpass filtering of the fault current signal and then performing a Hilbert transform; the initial sensitivity coefficient... ,in To protect the starting current setting value.

[0024] Preferably, in step S3, the preset thresholds for waveform distortion rate and low-frequency oscillation amplitude are as follows: the preset threshold for the ratio of second harmonic amplitude to fundamental amplitude is 0.1, and the preset threshold for the ratio of low-frequency oscillation amplitude to fundamental amplitude is 0.05; waveform distortion rate is defined as... The selection rule for the compensation model is: if and Then choose the composite compensation model, if Then choose the simplified compensation model, if and Then select fundamental frequency compensation; dynamic compensation factor. Phase compensation factor ,in For fault current phase shift; corrected sensitivity coefficient The characteristic points on the protection characteristic curve include: steep change points where the absolute value of the first-order difference is greater than 0.2 and critical points where the sensitivity coefficient is equal to 1.2. The three points with the smallest electrical distance are taken as protection coordination interface points.

[0025] This invention also provides an intelligent diagnostic device for defects in hydropower relay protection, comprising:

[0026] Data acquisition module: used to acquire the topology and equipment parameters, head signal, unit output signal, and setting values ​​and action information of each protection device in step S1 above;

[0027] Topology modeling and fault point identification module: used to perform the above step S1, construct the node admittance matrix, generate the fault point distribution sequence, and determine the dynamic evaluation coefficient of each fault point;

[0028] Short-circuit current calculation module: used to perform the above step S2, establish an equivalent model of the transient electromotive force change of the hydropower unit, introduce a dynamic impedance compensation factor, use a node algorithm to calculate the fault current and short-circuit current contribution rate matrix, and calculate the actual fault current amplitude of each protection device.

[0029] Sensitivity analysis and characteristic curve generation module: used to perform the above step S3, perform frequency domain analysis, extract characteristic frequency components, select compensation model and calculate compensation factor according to whether the waveform distortion rate or low frequency oscillation amplitude exceeds the preset threshold, obtain the corrected sensitivity coefficient, draw the protection characteristic curve and identify the protection matching interface point;

[0030] Defect diagnosis and tuning optimization module: used to perform the above step S4, perform differential diagnosis on the tuning value, verify the matching relationship, and iteratively output defect diagnosis report and optimized tuning value;

[0031] Human-computer interaction module: used to display protection characteristic curves, defect location, diagnostic conclusions and recommended setting parameters, and to receive manual correction instructions.

[0032] The present invention also provides an intelligent diagnostic device for defects in hydropower relay protection, comprising: at least one processor; at least one memory, wherein the memory stores a computer-executable program; when the computer-executable program is executed by the processor, the processor implements the intelligent diagnostic method for defects in hydropower relay protection as described above.

[0033] The present invention also provides a computer-readable storage medium storing computer-executable instructions, which, when loaded and executed by a processor, implement the above-described intelligent diagnostic method for defects in hydropower relay protection.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] By constructing an equivalent model of the transient electromotive force change of hydropower units and introducing a dynamic impedance compensation factor, the attenuation characteristics of short-circuit current and the low-frequency oscillation components are accurately characterized, significantly improving the accuracy of fault current calculation and overcoming the shortcomings of traditional short-circuit calculation methods that are difficult to adapt to the complex transient processes of hydropower units.

[0036] This invention further extracts waveform distortion rate and low-frequency oscillation characteristics through frequency domain analysis, establishes a multi-level compensation model based on harmonics and oscillation intensity, realizes adaptive dynamic correction of sensitivity coefficient, and effectively solves the technical problem of protection failure or false operation caused by waveform distortion.

[0037] Based on this, the present invention plots the protection characteristic curve according to the modified sensitivity coefficient, identifies the steep change point and the critical point as the interface, and then implements differentiated defect diagnosis. It can accurately distinguish typical defect types such as insufficient dynamic compensation, insufficient sensitivity and selective mismatch, and automatically outputs the tuning value adjustment scheme with the iterative optimization mechanism.

[0038] The entire diagnostic process requires no human intervention and achieves closed-loop intelligent processing from data acquisition, fault modeling, characteristic analysis to setting optimization. This significantly improves the automation level and adaptability of hydropower relay protection defect diagnosis, ensuring the safe and stable operation of hydropower units under different head conditions and complex fault conditions. Attached Figure Description

[0039] Figure 1 This is a schematic diagram of the method flow of Embodiment 1 of the present invention. Detailed Implementation

[0040] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0041] Example 1: This example provides an intelligent diagnostic method for defects in hydropower relay protection, such as... Figure 1 As shown, it includes the following steps.

[0042] Step S1: System Modeling and Fault Point Identification

[0043] First, the topology and equipment parameters of the hydropower station are collected. Specifically, the bus connection relationships and turbine generator parameters (including rated voltage) are obtained through the power station monitoring system or offline records. Rated capacity Direct-axis subtransient reactance Direct-axis transient reactance Direct-axis synchronous reactor Excitation time constant Moment of inertia Damping coefficient Transformer parameters (rated capacity, short-circuit impedance), line parameters (unit impedance, length), and plant power load parameters; construct the nodal admittance matrix based on the above parameters. The node numbering rules follow the power flow calculation conventions of the power system.

[0044] Then, the electrical distance between each relay protection device and the grid connection point of the hydropower unit is calibrated. (in per unit value) (Representation), generating a fault point distribution sequence. The electrical distance is calculated as follows: the sum of the impedances of the lines and transformers traversed from the fault point along the shortest electrical path to the grid connection point (converted to the same voltage level); three protection monitoring zones are divided based on the electrical distance.

[0045] Near the protected area: ;

[0046] Central protected area: ;

[0047] Remote backup area: .

[0048] Acquire real-time head signals and the active power output signal of the generator unit The water head is divided into three intervals: low water head. Rated head High water head ,in The rated head; the output is divided into three ranges: light load. , in-process Heavy load ,in This is the rated output.

[0049] By using historical simulation or measured data, the average short-circuit current contribution rate under the combination of each head range and output range is calculated in advance, forming a... The lookup table is used; during real-time diagnosis, the dynamic evaluation coefficients for each fault point are obtained using bilinear interpolation based on the current head and output. The formula for bilinear interpolation is:

[0050]

[0051] in , These represent the normalized positions of the current water head and power output within the corresponding intervals. To look up the first in the table The fault point is at the first Water head section, first The average contribution rate within the output range.

[0052] Step S2: Short-circuit current calculation

[0053] (1) Equivalent model of transient electromotive force change of hydropower unit

[0054] An equivalent model of the transient electromotive force change of the hydropower unit is established, and the following set of differential equations is used to describe the excitation system and rotor motion characteristics:

[0055]

[0056]

[0057] In the formula:

[0058] — Transient potential (per unit);

[0059] — Excitation time constant (s);

[0060] — Excitation voltage (per unit);

[0061] — Direct-axis synchronous reactance (per unit);

[0062] — Direct-axis transient reactance (per unit);

[0063] — Direct-axis current (per unit);

[0064] — Moment of inertia (kg·m²);

[0065] — Rotor angular velocity (rad / s);

[0066] — Synchronous angular velocity (rad / s);

[0067] — Mechanical torque (per unit);

[0068] — Electromagnetic torque (per unit);

[0069] — Damping coefficient.

[0070] The above system of equations is solved numerically using the fourth-order Runge-Kutta method, with a time step of [value missing]. The transient potential after the fault is obtained. The attenuation curve; according to By combining the network equations, the periodic component and the decaying DC component of the fault current can be calculated, thus obtaining the fault current. The complete waveform.

[0071] (2) Introduction of dynamic impedance compensation factor

[0072] Introducing dynamic impedance compensation factor Its function is to correct the equivalent injected current of each node in the short-circuit current calculation so as to reflect the dynamic change of the equivalent internal impedance during the transient process of the hydropower unit; the dynamic impedance compensation factor is based on 1, plus a correction term that is proportional to the square of the real-time short-circuit current and decays exponentially with time, and the proportional coefficient and time constant of the correction term are determined according to the current head interval.

[0073] In this embodiment, the specific formula for calculating the dynamic impedance compensation factor is as follows:

[0074]

[0075] in:

[0076] — Compensation coefficient, range of values ;

[0077] — Compensation time constant, range of values ;

[0078] — Real-time short-circuit current (per unit);

[0079] — Reference current, the rated current of the water intake generator unit.

[0080] The compensation coefficient and time constant are determined based on the specific head interval as follows:

[0081] Low head ( ): , ;

[0082] Rated head ( ): , ;

[0083] High head ( ): , .

[0084] The above values ​​are based on the following: at low head, the turbine has large mechanical inertia, and the speed changes drastically during transient processes, resulting in significant changes in equivalent impedance. Therefore, a larger compensation range and a longer duration are required. At high head, the unit has better stability, and the compensation range is correspondingly reduced.

[0085] (3) Nodal algorithm for solving fault current

[0086] Dynamic impedance compensation factor Multiply by the equivalent injection current of each node, and then substitute into the supernode iterative equation for solving; specifically, for the For each node, the corrected injection current is:

[0087]

[0088] in For nodes The corresponding dynamic impedance compensation factor (taken as 1 if it is not a generator node); then solve the node voltage equations:

[0089]

[0090] in Here is the nodal admittance matrix. For node voltage vectors, The corrected injection current vector is used; the voltage of each node is obtained by iterative solution (such as Gauss-Seidel method or Newton-Raphson method), and then the fault current of each branch is calculated.

[0091] (4) Fault point classification

[0092] According to electrical distance Based on the fault current characteristics, fault points are divided into three categories:

[0093] Dynamic coupling fault point: electrical distance And the fault current decay time constant These types of faults are located near the generator, and the short-circuit current contains abundant aperiodic and low-frequency oscillating components, making them significantly affected by the dynamic characteristics of the hydropower unit.

[0094] Weak power supply fault point: electrical distance And the fault current amplitude Such fault points are far from the power source, the fault current amplitude is small, and the protection sensitivity is easily insufficient.

[0095] Other fault points: those falling between the two mentioned above can be handled using conventional short-circuit calculation methods.

[0096] For dynamically coupled fault points, a dynamic characteristic compensation model for the fault point is further established, and the compensation parameters are identified online by the recursive least squares (RLS) method to more accurately reflect the impact of unit dynamics on short-circuit current.

[0097] (5) Fault current contribution rate matrix

[0098] The fault current contribution rate matrix is ​​established using the recursive least squares method. The input to RLS is the current sampling value at each protection installation point for the first 5 cycles after the fault occurs, and the output is the short-circuit current contribution rate of each protection device at each fault point. (satisfies after normalization) The RLS recursive formula is:

[0099]

[0100] in Here is the gain matrix. The input vector.

[0101] Finally, the total short-circuit current at each fault point is considered. (Calculated using metallic short circuit at the fault point), calculate the actual fault current amplitude of each protection device:

[0102]

[0103] Step S3: Sensitivity Analysis and Characteristic Curve Generation

[0104] (1) Comprehensive weight and coupling region division

[0105] Calculate each fault point Overall weight :

[0106]

[0107] in Contribution rate to short-circuit current For dynamic evaluation coefficients, The total number of protection devices; this weight comprehensively reflects the degree of impact of the fault point on the protection devices and the time-varying nature of the operating conditions.

[0108] Delineating coupling regions based on comprehensive weights:

[0109] Strongly coupled region: ;

[0110] Weak coupling zone: ;

[0111] Transition zone: .

[0112] (2) Frequency domain analysis and extraction of characteristic frequency components

[0113] Fault current sampling sequence for each fault point Perform a Fast Fourier Transform (FFT) at a sampling frequency of 4000 Hz, and take 10 cycles after the fault in the analysis window; extract the following characteristic frequency components:

[0114] Fundamental amplitude (50 Hz or 60 Hz);

[0115] Second harmonic amplitude (100 Hz or 120 Hz);

[0116] Third harmonic amplitude (150 Hz or 180 Hz).

[0117] low-frequency oscillation component amplitude Extraction method: The fault current signal is zero-mean processed, and a passband frequency of is applied. A bandpass filter (such as a fourth-order Butterworth filter) is used, and then a Hilbert transform is performed on the filtered signal to obtain the instantaneous amplitude envelope. The maximum value of the envelope is taken as the amplitude. This component reflects the current fluctuations in hydropower units caused by governor response or low-frequency oscillations in the power system.

[0118] (3) Calculation of initial sensitivity coefficient

[0119] No. The first protection device in the The initial sensitivity coefficient at each fault point is defined as:

[0120]

[0121] in This is the starting current setting value (or operating threshold) for the protection device; for distance protection, the current corresponding to the minimum operating impedance can be used.

[0122] (4) Compensation triggering conditions and model selection

[0123] Dynamic sensitivity compensation is activated when the fault point is located in a strongly coupled region and meets any of the following conditions:

[0124] Second harmonic amplitude ratio to fundamental frequency ;

[0125] Low-frequency oscillation to fundamental amplitude ratio .

[0126] The thresholds mentioned above (0.1 and 0.05) are empirical values ​​obtained through extensive simulation and field fault data analysis. They can effectively distinguish between harmonics caused by normal excitation inrush current or load fluctuations and waveform distortions associated with faults.

[0127] If compensation is triggered, the waveform distortion rate (THD) is calculated:

[0128]

[0129] The compensation model is selected based on THD and the intensity of low-frequency oscillations:

[0130] Composite compensation model: when and At the same time, the effects of harmonics and low-frequency oscillations should be considered simultaneously;

[0131] Simplified compensation model: when At that time, only the harmonic effects are considered, and the phase compensation of low-frequency oscillations is ignored;

[0132] Fundamental wave compensation model: When and At this time, the waveform distortion is very small and no additional compensation is required.

[0133] (5) Compensation factor calculation and sensitivity correction

[0134] Dynamic compensation factor and phase compensation factor The calculation formula is:

[0135]

[0136]

[0137] in The phase shift of the fault current can be calculated using the real and imaginary parts of the fundamental component: .

[0138] For the simplified compensation model, take For the fundamental frequency compensation model, take , .

[0139] The corrected sensitivity coefficient is:

[0140]

[0141] (6) Plotting protection characteristic curves and identifying feature points

[0142] electrical distance (x-axis) and the corrected sensitivity coefficient (Vertical axis) Plot the protection characteristic curve Since the fault points are arranged in ascending order of electrical distance, the spacing between adjacent points is... (Typically, 0.01 pu is used); Calculate the first difference of the curve:

[0143]

[0144] Will The point is marked as a steep change point; The point (or its closest approximation point) is marked as the critical point; the three points with the smallest electrical distance among all abrupt change points and critical points are taken as the protection coordination interface points; these three points correspond to the protection coordination boundaries of the near zone, middle zone, and far zone, respectively, and are important references for the differentiated adjustment of the setting value.

[0145] Step S4: Defect Diagnosis and Tuning Optimization

[0146] (1) Differentiated defect diagnosis

[0147] Based on the segmentation results of the protection characteristic curve, perform the following differential diagnosis:

[0148] Insufficient dynamic compensation: This is particularly relevant for strongly coupled regions and dynamic compensation factors. The section indicating that the existing setting value cannot adapt to waveform distortion is judged to be a defect of insufficient dynamic compensation; the output suggestion is to reduce the starting current setting value to its original value. times.

[0149] Insufficient sensitivity: Regarding the sensitivity coefficient The section is deemed to have insufficient sensitivity; output recommendations: adopt adaptive tuning (such as automatically adjusting the threshold according to the load current) or add auxiliary protection (such as adding a directional element).

[0150] Selectivity defects: regarding sensitivity coefficient And the time difference between adjacent protection actions The section in question is classified as having a selective defect; recommended output: increase the starting current setting value back to its original value. Double, or increase latency .

[0151] The diagnostic thresholds (1.2, 2.5, 0.1s) were determined based on the Technical Specification for Relay Protection and Automatic Safety Devices (GB / T14285) and the actual operating experience of hydropower plants.

[0152] (2) Verification and iterative optimization of cooperation relationship

[0153] Verify whether the coordination relationship between adjacent protection systems meets the following conditions:

[0154] Action time difference ;

[0155] Corrected sensitivity coefficient (For primary protection) or (For backup protection, this embodiment uniformly takes 1.2).

[0156] If the conditions are not met, the tuning value is adjusted iteratively. Each adjustment is 5% of the current tuning value, and the adjustment direction is determined according to the defect type (the tuning value is lowered for defects with insufficient dynamic compensation and insufficient sensitivity, and the tuning value is increased for selective defects). After each iteration, steps S2 and S3 are re-executed to calculate the new sensitivity coefficient and matching relationship. When the matching condition is met or the number of iterations reaches 10, the iteration is stopped, and the final defect diagnosis report and optimized tuning value are output.

[0157] The report includes: defect type, affected area (electrical distance range), current sensitivity coefficient, recommended setting value, and suggested operating time.

[0158] Example 2: This example provides an intelligent diagnostic device for defects in hydropower relay protection, used to execute the above method; the device includes the following modules:

[0159] Data acquisition module: Used to acquire the topology and equipment parameters, head signal, unit output signal, and setting values ​​and action information of each protection device in step S1 above; this module connects to the power plant monitoring system via the IEC 61850 protocol and also has an analog input interface ( or It is used to connect to pressure transmitters, power transmitters, etc.

[0160] Topology modeling and fault point identification module: This module constructs the node admittance matrix, generates a fault point distribution sequence, and determines the dynamic evaluation coefficients of each fault point using bilinear interpolation based on head and output. Internally, this module maintains a... The lookup table is used to implement the linear interpolation algorithm.

[0161] The short-circuit current calculation module is used to establish an equivalent model of the transient electromotive force change of the hydropower unit, introduce a dynamic impedance compensation factor, use a node algorithm to calculate the fault current and short-circuit current contribution rate matrix at each fault point, and calculate the actual fault current amplitude of each protection device based on the total short-circuit current. This module uses the fourth-order Runge-Kutta method to solve the differential equation and implements the recursive least squares method to identify the contribution rate.

[0162] Sensitivity analysis and characteristic curve generation module: used to perform frequency domain analysis (FFT, bandpass filtering, Hilbert transform), extract characteristic frequency components, calculate waveform distortion rate, select compensation model according to preset threshold and calculate dynamic compensation factor and phase compensation factor, obtain the corrected sensitivity coefficient, draw protection characteristic curve and identify protection interface points.

[0163] Defect Diagnosis and Tuning Optimization Module: Used to perform differential diagnosis on tuning values ​​(insufficient dynamic compensation, insufficient sensitivity, selective defects), verify the matching relationship, and iteratively output defect diagnosis reports and optimized tuning values.

[0164] Human-machine interaction module: used to display the protection characteristic curve (horizontal axis is electrical distance, vertical axis is sensitivity coefficient) in a graphical interface, and highlight the location of the defect point, the diagnosis conclusion and the recommended setting parameters; supports touch screen or keyboard and mouse operation, and can receive manual correction instructions (such as forcibly setting the setting value range).

[0165] The above modules can be integrated into the intelligent relay protection operation and maintenance terminal or edge computing node of the hydropower station, and have real-time data interfaces with fault recorders, protection devices, and synchronous phasor measurement units (PMUs), and support data interaction with the power station monitoring system through the IEC61850 protocol.

[0166] Example 3: This example provides an intelligent diagnostic device for defects in hydropower relay protection, including at least one processor and at least one memory; the memory stores a computer-executable program, and when the program is executed by the processor, the processor implements the method described in Example 1.

[0167] Specifically, the device can be a hydropower station relay protection operation and maintenance server (deployed in the central control room) or a portable tester (used for on-site debugging); the device also includes communication interfaces for connecting to fieldbus or Ethernet (such as RJ45, fiber optic interfaces), as well as a database (SQLite or MySQL) for storing historical defect diagnosis cases.

[0168] The processor can be an ARM Cortex-A series, Intel Xeon, or a domestic Phytium processor. Memory includes DDR4 RAM and NAND Flash or SSD; the device runs a Linux or Windows operating system.

[0169] Example 4: This example provides a computer-readable storage medium storing computer-executable instructions; when these instructions are loaded and executed by a processor, the method described in Example 1 is implemented; the storage medium may be a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0170] The embodiments of the present invention are given for the purposes of illustration and description. Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Any changes, modifications, substitutions and variations made by those skilled in the art to the above embodiments within the scope of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for intelligent diagnosis of defects in hydropower relay protection, characterized in that, Includes the following steps: Step S1: System Modeling and Fault Point Identification Collect the topology and equipment parameters of the hydropower station and construct the node admittance matrix; The electrical distance between each relay protection device and the grid connection point of the hydropower unit is calibrated to generate a fault point distribution sequence; the protection monitoring area is divided based on the electrical distance; real-time head signal and unit output signal are collected, and the dynamic evaluation coefficient of each fault point is determined according to the head and output range; Step S2: Short-circuit current calculation An equivalent model of the transient electromotive force change of a hydropower unit was established, and the transient electromotive force decay curve was obtained by solving a set of differential equations. The model is combined with the node admittance matrix, and a dynamic impedance compensation factor is introduced. The dynamic impedance compensation factor is based on 1, plus a correction term that is proportional to the square of the real-time short-circuit current and decays exponentially with time. The proportional coefficient and time constant of the correction term are determined according to the current head interval. The fault current is solved using a node algorithm. Fault points are classified according to electrical distance and fault current characteristics. A fault current contribution rate matrix is ​​established, and the actual fault current amplitude of each protection device is calculated in combination with the total short-circuit current. Step S3: Sensitivity Analysis and Characteristic Curve Generation The comprehensive weight of each fault point is calculated based on the fault current contribution rate matrix and dynamic evaluation coefficients, and the coupling region is divided based on the comprehensive weight. Frequency domain analysis is performed on the fault current to extract characteristic frequency components; Calculate the initial sensitivity coefficient of each protection device. For fault points located in strong coupling areas and whose waveform distortion rate or low-frequency oscillation amplitude exceeds their respective preset thresholds, select a compensation model according to the degree of waveform distortion and calculate the compensation factor to correct the sensitivity coefficient. Plot the protection characteristic curve with electrical distance as the abscissa and the corrected sensitivity coefficient as the ordinate, and identify the feature points on the curve as protection coordination interface points. Step S4: Defect Diagnosis and Tuning Optimization Based on the segmented results of the protection characteristic curves, perform differentiated diagnosis to identify defect types, including insufficient dynamic compensation, insufficient sensitivity, and selective defects; and verify the coordination relationship between adjacent protection systems. If the matching requirements are not met, the setting value is iteratively adjusted until the requirements are met or the preset number of iterations is reached, and a defect diagnosis report and optimized setting value are output.

2. The intelligent diagnostic method for defects in hydropower relay protection according to claim 1, characterized in that, The criteria for dividing the head interval in step S1 are as follows: low head is... Rated head is High water head is ,in The rated head; the standard for dividing the output range is: light load is , middle load is Overload ,in The rated output is used; the dynamic evaluation coefficient is calculated in advance using historical simulation or measured data to determine the average short-circuit current contribution rate under the combination of head interval and output interval, and is obtained by bilinear interpolation based on the current head and output during real-time diagnosis.

3. The intelligent diagnostic method for defects in hydropower relay protection according to claim 1, characterized in that, The differential equations of the equivalent model of the transient electromotive force change of the hydropower unit in step S2 are as follows: in, For transient potential, The excitation time constant is This is the excitation voltage. It is a direct-axis synchronous reactance. For direct-axis transient reactance, For direct-axis current, For rotational inertia, The rotor angular velocity, For synchronous angular velocity, For mechanical torque, For electromagnetic torque, The damping coefficient is used; the transient potential decay curve is obtained through numerical solution, and then the fault current is calculated.

4. The intelligent diagnostic method for defects in hydropower relay protection according to claim 1, characterized in that, Dynamic impedance compensation factor in step S2 The calculation formula is: in, This is the compensation coefficient, and its value range is... ; To compensate for the time constant, the value range is: ; For real-time short-circuit current, The reference current is taken from the rated current of the hydropower unit; determined according to the head range. and Specific values: at low head , ; at rated head , High water head , .

5. The intelligent diagnostic method for defects in hydropower relay protection according to claim 1, characterized in that, The fault point classification method in step S2 is as follows: dynamic coupling fault points are those with an electrical distance of not more than [missing information]. And the fault current decay time constant is no greater than The point of failure; The weak power supply fault point is the electrical distance greater than Fault points with fault current amplitude less than 0.05 times the rated current; other fault points are those between these two values; the fault current contribution rate matrix is ​​established using the recursive least squares method, and the actual fault current amplitude of each protection device is calculated according to... Calculation, where For contribution rate, For the first The total short-circuit current at each fault point.

6. The intelligent diagnostic method for defects in hydropower relay protection according to claim 1, characterized in that, The comprehensive weighting in step S3 The calculation formula is: ,in Contribution rate to short-circuit current For dynamic evaluation coefficients, The total number of protection devices; the standard for dividing the coupling area is: strong coupling area. , weak coupling region transition zone ; The characteristic frequency components include the fundamental frequency amplitude. Second harmonic amplitude Third harmonic amplitude and the amplitude of low-frequency oscillation components The amplitude of the low-frequency oscillation component is obtained by extracting the maximum envelope value after bandpass filtering of the fault current signal and then performing a Hilbert transform; the initial sensitivity coefficient... ,in To protect the starting current setting value.

7. The intelligent diagnostic method for defects in hydropower relay protection according to claim 1, characterized in that, In step S3, the preset thresholds for waveform distortion rate and low-frequency oscillation amplitude are as follows: the preset threshold for the ratio of second harmonic amplitude to fundamental amplitude is 0.1, and the preset threshold for the ratio of low-frequency oscillation amplitude to fundamental amplitude is 0.05; waveform distortion rate is defined as... The selection rule for the compensation model is: if and Then choose the composite compensation model, if Then choose the simplified compensation model, if and Then select fundamental frequency compensation; dynamic compensation factor. Phase compensation factor ,in This is due to the phase shift of the fault current; Corrected sensitivity coefficient The characteristic points on the protection characteristic curve include: steep change points where the absolute value of the first-order difference is greater than 0.2 and critical points where the sensitivity coefficient is equal to 1.

2. The three points with the smallest electrical distance are taken as protection coordination interface points.

8. A smart diagnostic device for defects in hydropower relay protection, characterized in that, include: Data acquisition module: used to acquire the topology and equipment parameters, head signal, unit output signal, and setting values ​​and action information of each protection device in step S1 of claim 1; Topology modeling and fault point identification module: used to perform step S1 as described in claim 1, construct the node admittance matrix, generate the fault point distribution sequence, and determine the dynamic evaluation coefficient of each fault point; Short-circuit current calculation module: used to execute step S2 as described in claim 1, establish an equivalent model of transient electromotive force change of hydropower unit, introduce dynamic impedance compensation factor, use node algorithm to calculate fault current and short-circuit current contribution rate matrix, and calculate the actual fault current amplitude of each protection device. Sensitivity analysis and characteristic curve generation module: used to perform step S3 as described in claim 1, perform frequency domain analysis, extract characteristic frequency components, select a compensation model and calculate a compensation factor according to whether the waveform distortion rate or low frequency oscillation amplitude exceeds a preset threshold, obtain the corrected sensitivity coefficient, draw the protection characteristic curve and identify the protection matching interface point; Defect diagnosis and tuning optimization module: used to execute step S4 as described in claim 1, perform differential diagnosis on tuning values, verify the matching relationship, and iteratively output defect diagnosis report and optimized tuning values; Human-computer interaction module: used to display protection characteristic curves, defect location, diagnostic conclusions and recommended setting parameters, and to receive manual correction instructions.

9. A smart diagnostic device for defects in hydropower relay protection, characterized in that, include: At least one processor; At least one memory storing a computer-executable program; when the computer-executable program is executed by the processor, the processor implements the intelligent diagnostic method for hydropower relay protection defects as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The storage medium stores computer-executable instructions, which, when loaded and executed by a processor, implement the intelligent diagnostic method for hydropower relay protection defects as described in any one of claims 1 to 7.

Images