A wideband measurement data credibility determination method and system based on a measurement value jump amplitude

Abstract

The application discloses a kind of wideband measurement data reliability determination method and system based on the jump amplitude of measured value, comprising: modeling in static reliability boundary;Error range under single disturbance is constructed;According to complex actual measurement environment, construct the reliability boundary under dynamic aliasing disturbance;Reliability boundary under dynamic aliasing disturbance of frequency jump amplitude, voltage amplitude jump amplitude, phase angle jump amplitude data superposition change is constructed;According to the reliability boundary obtained, in combination with the measurement object, the reliability weight level is constructed again;The division method based on the reliability boundary under dynamic aliasing disturbance and the new reliability weight level considering the kind of measurement object method are constructed.The application effectively solves the new power system measurement result correctness evaluation problem by constructing the reliability determination method, supports dispatching department to make reasonable decision in the link such as accident handling, power adjustment, case backtracking analysis, reduces the misoperation risk brought by data quality problem, supports power grid operation safety from basic data level.

Description

Technical Field

[0001] This invention relates to a method and system for determining the reliability of broadband measurement data, and more particularly to a method and system for determining the reliability of broadband measurement data based on the amplitude of measurement value jumps. Background Technology

[0002] In recent years, with the large-scale integration of high-proportion new energy sources and high-proportion power electronic equipment ("dual high") into my country's power grid, electrical signals in the grid are more prone to fluctuations, and dynamic and transient events occur frequently, such as rapid changes in amplitude and phase, frequency shifts, and low-frequency oscillations. This has increasingly highlighted the importance of monitoring and controlling grid stability. Therefore, timely and accurate measurement of power system signals to prepare for subsequent grid dispatching and control has become crucial. Traditional power system signal measurement methods, such as zero-crossing detection, discrete Fourier transform, and fast Fourier transform, have shown good measurement performance under steady-state grid conditions. However, when facing transient and dynamic conditions, the measured electrical signals may deviate significantly from the true values. This is not a data quality issue, but rather because traditional signal measurement methods can no longer adequately meet the development requirements of modern new power grids. Summary of the Invention

[0003] Purpose of the invention: The purpose of this invention is to provide a method and system for determining the reliability of broadband measurement data based on the amplitude of measurement value jumps, so as to determine the reliability of measurement data under different conditions based on existing measurement methods and improve the efficiency of data utilization.

[0004] Technical solution: This invention includes: modeling a static reliability boundary; constructing an error range under a single disturbance; constructing a reliability boundary under dynamic aliasing disturbance based on complex actual measurement environments: constructing a reliability boundary under dynamic aliasing disturbance with the superposition and change of frequency jump amplitude, voltage amplitude jump amplitude, and phase angle jump amplitude data; and constructing a reliability weight level again based on the obtained reliability boundary and the measurement object: constructing a new reliability weight level based on the method of dividing the reliability boundary under dynamic aliasing disturbance and considering the type of measurement object.

[0005] The modeling of the static reliability boundary specifically includes: constructing a library of existing wideband measurement algorithms, including zero-crossing detection, discrete Fourier transform, and fast Fourier transform; the zero-crossing detection, discrete Fourier transform, and fast Fourier transform algorithms will each be modeled for static reliability boundaries based on their algorithmic characteristics; the reliability classification is based on the voltage amplitude, frequency, and phase angle jump amplitude measured by the corresponding algorithms.

[0006] The construction of the error range under a single disturbance specifically involves: constructing error range values ​​for frequency jump amplitude, voltage amplitude jump amplitude, and phase angle jump amplitude under a single disturbance; the frequency jump amplitude, voltage amplitude jump amplitude, and phase angle jump amplitude data are calculated from the change amplitude of the measured frequency value, voltage amplitude, and phase angle value within a data window, and the reliability of the measurement data is determined by the magnitude of the jump amplitude of the three measured values.

[0007] The reliability boundary under dynamic aliasing disturbance is constructed based on the complex actual measurement environment. Specifically, the frequency jump amplitude, voltage amplitude jump amplitude, and phase angle jump amplitude data are superimposed. First, the jump amplitude within a data window is calculated from the measured frequency value, voltage amplitude, and phase angle value. The magnitude of the jump amplitude of two or three of the measured values ​​is selected to determine the reliability of the measurement data.

[0008] The measurement objects are categorized into measurements at the step-up bus of thermal power units, measurements at the grid connection of new energy sources, and measurements at pumped storage power stations.

[0009] The reliability of measurements taken at the pumped storage power station is higher than that taken at the new energy grid connection.

[0010] The measurement at the step-up bus of the thermal power unit uses a new reliability weighting level classification method, which adopts Equation 16-19:

[0011]

[0012]

[0013] In the formula, u1, f1, and θ1 are the electrical quantities measured in the current data window, u0, f0, and θ0 are the electrical quantities measured in the previous adjacent data window, and their values ​​represent voltage amplitude, frequency value, and phase angle value, respectively; A and B are the boundary values ​​for dividing voltage amplitude jumps, A' and B' are the boundary values ​​for dividing frequency jumps, and A” and B” are the boundary values ​​for dividing phase angle jumps.

[0014] The measurement at the new energy grid connection point uses a new reliability weighting level classification method, adopting Equation 20-23:

[0015]

[0016]

[0017] The measurement at the pumped storage power station uses a new reliability weighting level classification method, which adopts Equation 24-27:

[0018]

[0019]

[0020] A broadband measurement data reliability determination system based on the amplitude of measurement value jumps includes: a static reliability boundary modeling module, an error range construction module under a single disturbance, a reliability boundary construction module under dynamic aliasing disturbance, and a reliability weight level reconstruction module.

[0021] Beneficial effects: By constructing a reliability judgment method, this invention effectively solves the problem of judging the correctness of measurement results in new power systems, supports dispatching departments in making reasonable decisions in accident handling, power adjustment, case retrospective analysis and other aspects, reduces the risk of erroneous operation caused by data quality problems, and supports the safe operation of the power grid from the basic data level. Attached Figure Description

[0022] Figure 1 This is a flowchart of the present invention;

[0023] Figure 2 This is a schematic diagram of the confidence boundary model for the zero-crossing detection method.

[0024] Figure 3 This is a schematic diagram of the zero-crossing detection method;

[0025] Figure 4 This is a schematic diagram of the DFT transform under asynchronous sampling conditions;

[0026] Figure 5 This is a schematic diagram of the 8-point FFT time decimation algorithm;

[0027] Figure 6 This is a schematic diagram of a steady-state sampled signal;

[0028] Figure 7 This is a schematic diagram of the sampling signal during a dynamic process;

[0029] Figure 8 This is a schematic diagram of the sampling signal for a transient process. Detailed Implementation

[0030] The invention will now be further described with reference to the accompanying drawings.

[0031] This invention provides a method for determining the reliability of broadband measurement data based on the amplitude of measurement jumps. It models a static reliability boundary using existing broadband measurement algorithms. Using measurement data obtained from existing algorithms, including electrical quantities such as voltage amplitude, phase angle, and frequency in substations, it calculates the jump amplitudes of voltage amplitude, phase angle, and frequency values. Combinations of jump types in voltage amplitude, phase angle, and frequency values ​​simulate single / aliasing disturbances. The reliability boundary for measurements under single / aliasing disturbance conditions is defined by the magnitude of the voltage, frequency, and phase angle jump amplitudes. Finally, considering the type of measurement object, the reliability weight level under aliasing disturbance conditions is redefined. Figure 1As shown, the specific steps include:

[0032] S1. Modeling the static credibility boundary

[0033] A library of existing broadband measurement algorithms, including zero-crossing detection, Discrete Fourier Transform (DFT), and Fast Fourier Transform (FFT), will be constructed. Static reliability boundary modeling will be performed for the zero-crossing detection, DFT, and FFT algorithms based on their respective algorithmic characteristics. The reliability classification will be based on the voltage amplitude, frequency, and phase angle jump amplitude measured by the corresponding algorithms. Specifically, this includes:

[0034] S11. Establish a zero-crossing detection model and its confidence boundary, such as... Figure 2 and Figure 3 As shown in Formula 1-4,

[0035] Sinusoidal voltage signal:

[0036]

[0037] In the formula, t1 and t3 are the times corresponding to adjacent zero crossings of signal x(t) under the condition of frequency shift, m and n are the first sampling points after t1 and t3, and finally the actual frequency f of the signal is obtained.

[0038] The reliability boundary of the zero-crossing detection method is that when the harmonic content of a certain segment of the power frequency signal is high, such as above U%, there may be multiple zero-crossings in this segment of the signal. If the time interval between adjacent crossing points is less than the signal sampling interval, the zero-crossing detection method cannot accurately reflect the crossing action of the signal, that is, the zero-crossing detection method loses its applicability.

[0039] S12. Establish the DFT model and its confidence boundary, as shown in Equation 5. A schematic diagram of the DFT transformation under asynchronous sampling conditions is shown below. Figure 4 As shown.

[0040]

[0041] In the formula, Let x(t) be the phasor form of the signal, and N be the number of sampling points per fundamental frequency period.

[0042] The reliability boundary of the DFT method is defined when a signal undergoes a transient process (such as a step jump). At the moment the step occurs, the signal has lost its sinusoidal characteristics, and its physical meaning is no longer clear. Mathematically, this manifests as calculation errors caused by signal discontinuity. In this case, the DFT method loses its applicability.

[0043] S13. Establish the FFT model and its confidence boundary, as shown in Formula 6-7.

[0044]

[0045] A(k) and B(k) are DFTs with high composite numbers (N / 2). Continuing the decomposition as described above, let l = 0, 1, ..., N / 4-1, then:

[0046]

[0047] Continue this decomposition process until a two-point DFT is reached. Taking an eight-point FFT as an example, the process is illustrated below. Figure 5 .

[0048] The reliability boundary of the FFT method is that when asynchronous sampling occurs, spectral leakage and the picket fence effect will severely reduce the accuracy of harmonic measurements. Asynchronous sampling is unavoidable during frequency dynamics, so the reliability of the measurement results of the traditional FFT method will also decrease when frequency dynamics occur. When there are large frequency dynamics, the FFT method will lose its applicability.

[0049] S2. Construct the error range under a single perturbation.

[0050] An error range is constructed using frequency jump amplitude, voltage amplitude jump amplitude, and phase angle jump amplitude as single disturbances. These amplitudes are calculated from the changes in the measured frequency, voltage amplitude, and phase angle values ​​within a data window. The reliability of the measurement data is determined by the magnitude of the jump amplitudes of these three measurements, using formula 8-10.

[0051]

[0052] In the formula, u1, f1, and θ1 are the electrical quantities measured in the current data window, and u0, f0, and θ0 are the electrical quantities measured in the previous adjacent data window, representing voltage amplitude, frequency value, and phase angle value, respectively; A and B are the boundary values ​​for dividing the voltage amplitude jump amplitude, dividing the voltage amplitude jump amplitude into three parts, and considering the corresponding electrical states as steady-state, dynamic, and transient according to the magnitude of the jump amplitude; A' and B' are the boundary values ​​for dividing the frequency jump amplitude, dividing the frequency value jump amplitude into three parts, and considering the corresponding electrical states as steady-state, dynamic, and transient according to the magnitude of the jump amplitude; A” and B” are the boundary values ​​for dividing the phase angle jump amplitude, dividing the phase angle jump amplitude into three parts, and considering the corresponding electrical states as steady-state, dynamic, and transient according to the magnitude of the jump amplitude; The above is the error range division method under a single disturbance. The jump amplitude of voltage amplitude, frequency value, and phase angle value in one data window corresponds to three different single disturbances.

[0053] The states of the corresponding electrical signals under steady-state, dynamic, and transient conditions are shown in Equation 11. Schematic diagrams of the steady-state, dynamic, and transient signals are shown below. Figure 6-8 As shown.

[0054]

[0055] In the formula, the steady-state signal represents a sinusoidal signal with a constant or slowly changing frequency, whose amplitude and phase remain stable over a period of time. In the formula, A is the amplitude, f0 is the fundamental frequency, and Δf is the frequency offset. This is the initial phase angle.

[0056] The dynamic signal represents an unstable sine wave under disturbance, whose amplitude and phase both change with time, where the amplitude A(t) and phase angle are given by the formula. Both are functions that change over time.

[0057] Transient signals are unstable signals with brief abrupt changes and cannot be accurately described by a function. A typical transient signal is shown in Equation 11. Around time t0, the sinusoidal signal only undergoes a step change in amplitude.

[0058] S3. Based on complex real-world measurement environments, construct a confidence boundary under dynamic aliasing perturbations.

[0059] A confidence boundary is constructed under dynamic aliasing disturbances, where frequency jump amplitude, voltage amplitude jump amplitude, and phase angle jump amplitude data are superimposed and varied. The superimposed frequency jump amplitude, voltage amplitude jump amplitude, and phase angle jump amplitude data are first calculated from the measured frequency, voltage amplitude, and phase angle values ​​within a data window. The confidence of the measurement data is determined by selecting the jump amplitudes of two or three measured values, using formula 12-15.

[0060]

[0061]

[0062] In the formula, the electrical quantities u1, f1, θ1, u0, f0, θ0, and the boundary values ​​A, B, A', B', A”, B” have the same meaning as in formula 8-10, and the meanings of steady-state, dynamic, and transient electrical states are the same as in formula 11. Formulas 12-15 are the reliability boundary division method under dynamic aliasing disturbances. The jumps of voltage amplitude, frequency value, and phase angle value within a data window correspond to three different single disturbances. By combining and selecting the jumps of the three electrical quantities, four different aliasing disturbances can be formed.

[0063] S4. Based on the obtained confidence boundaries and the measurement objects, reconstruct the confidence weight levels.

[0064] A new reliability weighting level is constructed based on a reliability boundary partitioning method under dynamic aliasing disturbances and a method considering the types of measurement objects. Measurement object types are divided according to the different installation locations of the measurement equipment in the power grid, including measurements at the step-up bus of thermal power units, measurements at renewable energy grid connection points, and measurements at pumped storage power stations. Measurements at renewable energy grid connection points are more susceptible to the influence of power electronic equipment and are more likely to measure non-steady-state signals, therefore, they should be assigned lower reliability. Measurements at the step-up bus of thermal power units, being directly connected to the generation side, have better steady-state performance and higher reliability. Pumped storage power stations, due to their power supply and load operating modes, and their very rapid and frequent switching between these modes, experience non-steady-state signal influences from their frequency and phase regulation, peak shaving, and valley filling functions when acting as both power supply and load; therefore, the reliability of their measurement data is also lower, but higher than that of measurements at renewable energy grid connection points.

[0065] In addition, the matching between the type of monitoring signal and the measurement object should be considered. For example, for thermal power units, the probability of subsynchronous oscillation is relatively small. That is, when a dynamic signal or transient signal that deviates significantly from the fundamental frequency is measured from the signal of the thermal power unit, the measurement result should also be given a lower degree of confidence in terms of the matching between the measurement object and the type of signal.

[0066] For measurements at the step-up busbar of thermal power units, a new reliability weighting level classification method is adopted, using Equation 16-19.

[0067]

[0068] For measurements at new energy grid connection points, a new reliability weighting level classification method is adopted, using Equation 20-23.

[0069]

[0070] For measurements at pumped storage power stations, a new reliability weighting system is adopted, using Equation 24-27.

[0071]

[0072]

[0073] In the formula, u1, f1, and θ1 are the electrical quantities measured in the current data window, and u0, f0, and θ0 are the electrical quantities measured in the previous adjacent data window, whose values ​​represent voltage amplitude, frequency value, and phase angle value, respectively; A and B are the boundary values ​​for dividing the voltage amplitude jump amplitude, dividing the voltage amplitude jump amplitude into three parts; A' and B' are the boundary values ​​for dividing the frequency jump amplitude, dividing the frequency value jump amplitude into three parts; A” and B” are the boundary values ​​for dividing the phase angle jump amplitude, dividing the phase angle value jump amplitude into three parts. According to the magnitude of the jump amplitude, the corresponding electrical states are regarded as steady state, dynamic state, and transient state, respectively.

[0074] Based on the reliability boundary division method under dynamic aliasing disturbance, Equation 16-27 is a new reliability weight level determination method that takes into account the measurement object. The measurement objects can be classified according to their installation location into measurements at the step-up bus of thermal power units, measurements at the grid connection of new energy sources, and measurements at pumped storage power stations. Each type of measurement corresponds to four new reliability weight level determination methods under aliasing disturbance.

[0075] A system for determining the reliability of broadband measurement data based on the amplitude of measurement value jumps includes:

[0076] Static reliability boundary modeling module: Constructs a library of existing wideband measurement algorithms, including zero-crossing detection, Discrete Fourier Transform (DFT) algorithm, and Fast Fourier Transform (FFT) algorithm; the zero-crossing detection method, DFT algorithm, and FFT algorithm will be used for static reliability boundary modeling based on their respective algorithm characteristics; the reliability classification is based on the voltage amplitude, frequency, and phase angle jump amplitude obtained by the corresponding algorithm.

[0077] Error range construction module under single disturbance: Constructs error range values ​​under a single disturbance using data such as frequency jump amplitude, voltage amplitude jump amplitude, and phase angle jump amplitude; the frequency jump amplitude, voltage amplitude jump amplitude, and phase angle jump amplitude data are calculated from the change amplitude of the measured frequency value, voltage amplitude, and phase angle value within a data window, and the reliability of the measurement data is determined by the magnitude of the jump amplitude of the three measured values.

[0078] The confidence boundary construction module under dynamic aliasing disturbance constructs a confidence boundary under dynamic aliasing disturbance by superimposing and changing frequency jump amplitude, voltage jump amplitude, and phase angle jump amplitude data. The superimposed frequency jump amplitude, voltage jump amplitude, and phase angle jump amplitude data are first calculated from the measured frequency value, voltage amplitude, and phase angle value within a data window. The magnitude of the jump amplitude of two or three measured values ​​is selected to determine the confidence of the measured data.

[0079] Credibility weight level reconstruction module: Constructs a new credibility weight level based on the method of dividing credibility boundaries under dynamic aliasing disturbance and the method of considering the types of measurement objects; The types of measurement objects are divided according to the different installation locations of the measurement equipment in the power grid, which can be divided into measurements at the step-up bus of thermal power units, measurements at the grid connection of new energy sources, and measurements at pumped storage power stations.

[0080] Example

[0081] The reliability determination method for broadband measurement data based on the amplitude of measurement value jumps in this embodiment includes the following steps:

[0082] 1. Modeling at the static confidence boundary.

[0083] Construct a library of existing wideband measurement algorithms, including zero-crossing detection, Discrete Fourier Transform (DFT) algorithm, and Fast Fourier Transform (FFT) algorithm;

[0084] Zero-crossing detection, DFT algorithm, and FFT algorithm will each be modeled with static confidence boundaries based on their algorithmic characteristics; the confidence level will be determined based on the voltage amplitude, frequency, and phase angle jump amplitude measured by the corresponding algorithms.

[0085] 2. Establish a zero-crossing detection model and its confidence boundary, as shown in Formula 1-4.

[0086] Sinusoidal voltage signal:

[0087]

[0088]

[0089] In the formula, t1 and t3 are the times corresponding to adjacent zero crossings of signal x(t) under the condition of frequency shift, m and n are the first sampling points after t1 and t3, and finally the actual frequency f of the signal is obtained.

[0090] The reliability boundary of the zero-crossing detection method is that when the harmonic content of a certain segment of the power frequency signal is high, such as above U%, there may be multiple zero-crossings in this segment of the signal. If the time interval between adjacent crossing points is less than the signal sampling interval, the zero-crossing detection method cannot accurately reflect the crossing action of the signal, that is, the zero-crossing detection method loses its applicability.

[0091] 3. Establish the DFT model and its confidence boundary, as shown in Formula 5.

[0092]

[0093] In the formula, Let x(t) be the phasor form of the signal, and N be the number of sampling points per fundamental frequency period.

[0094] The reliability boundary of the DFT method is defined when a signal undergoes a transient process (such as a step jump). At the moment the step occurs, the signal has lost its sinusoidal characteristics, and its physical meaning is no longer clear. Mathematically, this manifests as calculation errors caused by signal discontinuity. In this case, the DFT method loses its applicability.

[0095] 4. Establish the FFT model and its confidence boundary, as shown in Formula 6-7.

[0096]

[0097] A(k) and B(k) are DFTs with high composite numbers (N / 2). Continuing the decomposition as described above, let l = 0, 1, ..., N / 4-1, then:

[0098]

[0099] Continue this process until you reach the DFT of two points.

[0100] The reliability boundary of the FFT method is that when asynchronous sampling occurs, spectral leakage and the picket fence effect will severely reduce the accuracy of harmonic measurements. Asynchronous sampling is unavoidable during frequency dynamics, so the reliability of the measurement results of the traditional FFT method will also decrease when frequency dynamics occur. When there are large frequency dynamics, the FFT method will lose its applicability.

[0101] 5. Construct the error range under a single disturbance.

[0102] Construct the error range value with data such as frequency jump amplitude, voltage amplitude jump amplitude, and phase angle jump amplitude as a single disturbance;

[0103] The frequency jump amplitude, voltage amplitude jump amplitude, and phase angle jump amplitude data are calculated from the changes in the measured frequency, voltage amplitude, and phase angle values ​​within a data window. The reliability of the measurement data is determined by the magnitude of the jump amplitude of the three measured values, using formula 8-10.

[0104]

[0105] A set of specific parameter values ​​is provided for reference, where,

[0106] A=10%, B=30%, α=100%, β=95%, γ=80%;

[0107] A'=2%, B'=10%, α'=100%, β'=85%, γ'=70%;

[0108] A”=3%, B”=12%, α”=100%, β”=90%, γ”=75%;

[0109] The states of the corresponding electrical signals under steady-state, dynamic, and transient conditions are shown in Equation 11.

[0110]

[0111] In the formula, the steady-state signal represents a sinusoidal signal with a constant or slowly changing frequency, whose amplitude and phase remain stable over a period of time. In the formula, A is the amplitude, f0 is the fundamental frequency, and Δf is the frequency offset. This is the initial phase angle.

[0112] The dynamic signal represents an unstable sine wave under disturbance, whose amplitude and phase both change with time, where the amplitude A(t) and phase angle are given by the formula. Both are functions that change over time.

[0113] Transient signals are unstable signals with brief abrupt changes and cannot be accurately described by a function. A typical transient signal is shown in Equation 11. Around time t0, the sinusoidal signal only undergoes a step change in amplitude.

[0114] 6. Based on the complex actual measurement environment, construct the confidence boundary under dynamic aliasing perturbation.

[0115] Construct a confidence boundary under dynamic aliasing disturbances based on the superposition and change of data of frequency jump amplitude, voltage jump amplitude, and phase angle jump amplitude;

[0116] The frequency jump amplitude, voltage amplitude jump amplitude, and phase angle jump amplitude data are superimposed. First, the jump amplitude within a data window is calculated from the measured frequency value, voltage amplitude, and phase angle value. Then, the jump amplitude of two or three measured values ​​is selected to determine the reliability of the measurement data, using formula 12-15.

[0117]

[0118]

[0119] A set of specific parameter values ​​is provided for reference, where,

[0120] A=10%, B=30%, A'=2%, B'=10%, i=100%, j=83%, k=67%;

[0121] A=10%, B=30%, A”=3%, B”=12%, i’=100%, j’=87%, k’=72%;

[0122] A'=2%, B'=10%, A"=3%, B"=12%, i"=100%, j"=80%, k"=65%;

[0123] A=10%, B=30%, A'=2%, B'=10%, A"=3%, B"=12%, i"'=100%, j"'=79%, k"'=64%;

[0124] 7. Based on the obtained confidence boundaries and the measurement objects, reconstruct the confidence weight levels.

[0125] Construct a new confidence weight level based on a confidence boundary partitioning method under dynamic aliasing perturbation and a method that considers the types of measurement objects;

[0126] The types of measurement objects can be divided according to the different installation locations of the measurement equipment in the power grid. They can be divided into measurements at the step-up bus of thermal power units, measurements at the grid connection of new energy sources, and measurements at pumped storage power stations. Measurements at the grid connection of new energy sources are more susceptible to the influence of power electronic equipment and are more likely to measure non-steady-state signals, so they should be given a lower confidence level.

[0127] Measurements taken at the step-up bus of thermal power units have good steady-state performance and high reliability because they are directly connected to the generation side. Pumped storage power stations, however, operate as both power sources and loads, and their switching between these modes is very rapid and frequent. The frequency and phase regulation, peak shaving, and valley filling functions they perform when acting as power sources or loads introduce non-steady-state signals, resulting in lower reliability of their measurement data, although still higher than that of measurements taken at renewable energy grid-connected locations.

[0128] In addition, the matching between the type of monitoring signal and the measurement object should be considered. For example, for thermal power units, the probability of subsynchronous oscillation is relatively small. That is, when a dynamic signal or transient signal that deviates significantly from the fundamental frequency is measured from the signal of the thermal power unit, the measurement result should also be given a lower degree of confidence in terms of the matching between the measurement object and the type of signal.

[0129] For measurements at the step-up busbar of thermal power units, a new reliability weighting level classification method is adopted, using Equation 16-19.

[0130]

[0131] A set of specific parameter values ​​is provided for reference, where,

[0132] A=10%, B=30%, A'=2%, B'=10%, X=100%, Y=86%, Z=70%;

[0133] A=10%, B=30%, A”=3%, B”=12%, X’=100%, Y’=90%, Z’=75%;

[0134] A'=2%, B'=10%, A"=3%, B"=12%, X"=100%, Y"=83%, Z"=68%;

[0135] A = 10%, B = 30%, A' = 2%, B' = 10%, A” = 3%, B” = 12%, X”' = 100%, Y”' = 82%, Z”' = 67%; For measurements at new energy grid connection points, a new reliability weighting level classification method is adopted, using formula 20-23.

[0136]

[0137] A set of specific parameter values ​​is provided for reference, where,

[0138] A=10%, B=30%, A'=2%, B'=10%, x=100%, y=80%, z=64%;

[0139] A=10%, B=30%, A”=3%, B”=12%, x’=100%, y’=84%, z’=69%;

[0140] A'=2%, B'=10%, A"=3%, B"=12%, x"=100%, y"=77%, z"=62%;

[0141] A=10%, B=30%, A'=2%, B'=10%, A"=3%, B"=12%, x"=100%, y"=76%, z"=61%;

[0142] For measurements at pumped storage power stations, a new reliability weighting system is adopted, using Equation 24-27.

[0143]

[0144] A set of specific parameter values ​​is provided for reference, where,

[0145] A=10%, B=30%, A'=2%, B'=10%,

[0146] A=10%, B=30%, A”=3%, B”=12%,

[0147] A'=2%, B'=10%, A"=3%, B"=12%,

[0148] A=10%, B=30%, A'=2%, B'=10%, A"=3%,

[0149] In the formula, u1, f1, and θ1 are the electrical quantities measured in the current data window, and u0, f0, and θ0 are the electrical quantities measured in the previous adjacent data window, whose values ​​represent voltage amplitude, frequency value, and phase angle value, respectively; A and B are the boundary values ​​for dividing the voltage amplitude jump amplitude, dividing the voltage amplitude jump amplitude into three parts; A' and B' are the boundary values ​​for dividing the frequency jump amplitude, dividing the frequency value jump amplitude into three parts; A” and B” are the boundary values ​​for dividing the phase angle jump amplitude, dividing the phase angle value jump amplitude into three parts. According to the magnitude of the jump amplitude, the corresponding electrical states are regarded as steady state, dynamic state, and transient state, respectively.

Claims

1. A method for determining the reliability of broadband measurement data based on the amplitude of measurement value jumps, characterized in that, include: Modeling is performed on the static confidence boundary; Construct the error range under a single perturbation; Based on the complex actual measurement environment, a confidence boundary is constructed under dynamic aliasing disturbance: a confidence boundary is constructed under dynamic aliasing disturbance with the superposition and change of frequency jump amplitude, voltage jump amplitude, and phase angle jump amplitude data. Based on the obtained confidence boundaries and combined with the measurement objects, a new confidence weight level is constructed: a new confidence weight level is constructed based on the confidence boundary division method under dynamic aliasing disturbance and the method considering the types of measurement objects; the types of measurement objects are divided into measurements at the step-up bus of thermal power units, measurements at the grid connection of new energy sources, and measurements at pumped storage power stations; for the measurements at the step-up bus of thermal power units, the new confidence weight level division method is calculated using the following formula: In the formula, , , The electrical quantities measured within the current data window. , , These are the electrical quantities measured within the previous adjacent data window, and their values ​​represent the voltage amplitude, frequency, and phase angle, respectively. , To define the threshold values ​​for voltage amplitude jumps, , To define the threshold value for frequency jump amplitude , The threshold value for defining the phase angle jump amplitude; The measurement at the new energy grid connection point, using the new reliability weighting level classification method, is calculated using the following formula: ； The new reliability weighting classification method for measurements at the pumped storage power station is calculated using the following formula: 。 2. The method for determining the reliability of broadband measurement data based on the amplitude of measurement value jumps according to claim 1, characterized in that, The modeling of the static reliability boundary specifically includes: constructing a library of existing wideband measurement algorithms, including zero-crossing detection, discrete Fourier transform, and fast Fourier transform; the zero-crossing detection, discrete Fourier transform, and fast Fourier transform algorithms will each be modeled for static reliability boundaries based on their algorithmic characteristics; the reliability classification is based on the voltage amplitude, frequency, and phase angle jump amplitude measured by the corresponding algorithms.

3. The method for determining the reliability of broadband measurement data based on the amplitude of measurement value jumps according to claim 2, characterized in that, The construction of the error range under a single disturbance specifically involves: constructing error range values ​​for frequency jump amplitude, voltage amplitude jump amplitude, and phase angle jump amplitude under a single disturbance; the frequency jump amplitude, voltage amplitude jump amplitude, and phase angle jump amplitude data are calculated from the change amplitude of the measured frequency value, voltage amplitude, and phase angle value within a data window, and the reliability of the measurement data is determined by the magnitude of the jump amplitude of the three measured values.

4. The method for determining the reliability of broadband measurement data based on the amplitude of measurement value jumps according to claim 3, characterized in that, The reliability boundary under dynamic aliasing disturbance is constructed based on the complex actual measurement environment. Specifically, the frequency jump amplitude, voltage amplitude jump amplitude, and phase angle jump amplitude data are superimposed. First, the jump amplitude within a data window is calculated from the measured frequency value, voltage amplitude, and phase angle value. The magnitude of the jump amplitude of two or three of the measured values ​​is selected to determine the reliability of the measurement data.

5. The method for determining the reliability of broadband measurement data based on the amplitude of measurement value jumps according to claim 1, characterized in that, The reliability of measurements taken at the pumped storage power station is higher than that taken at the new energy grid connection.

6. A system for determining the reliability of broadband measurement data based on the amplitude of measurement value jumps, used to implement the method for determining the reliability of broadband measurement data based on the amplitude of measurement value jumps as described in any one of claims 1 to 5, characterized in that, include: Static confidence boundary modeling module: Models the static confidence boundary. Error range construction module under a single perturbation: Constructs the error range under a single perturbation; Module for constructing credibility boundaries under dynamic aliasing disturbances: Based on complex actual measurement environments, construct credibility boundaries under dynamic aliasing disturbances: Construct credibility boundaries based on the superimposed changes of frequency jump amplitude, voltage jump amplitude, and phase angle jump amplitude data. Credibility weight level reconstruction module: Based on the obtained credibility boundary and combined with the measurement object, the credibility weight level is reconstructed: a new credibility weight level is constructed based on the method of dividing the credibility boundary under dynamic aliasing disturbance and the method of considering the type of measurement object; the types of measurement objects are divided into measurements at the step-up bus of thermal power units, measurements at the grid connection of new energy and measurements at pumped storage power stations.

Images