Filtering method and system for PMU power oscillation data based on tracking differentiator

Abstract

The present invention discloses a filtering method and system for PMU power oscillation data based on a tracking differentiator, and relates to the technical field of power oscillation data identification. Active power data is measured by using the voltage phasor of the PMU measurement node and the current phasor of the branch. According to the working principle of the PMU, the numerical observability of the system and the system topology observability are analyzed, and the configuration is combined with the genetic algorithm. The configured PMU is used to obtain the low-frequency oscillation signal, and the error of the obtained signal is regarded as a white noise signal with a power spectrum density distributed in the entire frequency domain, and a tracking differentiator is designed for filtering. The self-disturbance rejection control is combined with the parameter identification of the low-frequency oscillation signal, and a tracking differentiator is designed to filter the PMU power oscillation data, so as to speed up the approximation efficiency and reduce the distortion, thereby improving the accuracy of the algorithm in identifying the oscillation parameters.

Description

Technical Field

[0001] The present invention relates to the technical field of power oscillation data identification, and in particular to a filtering method and system for PMU power oscillation data based on a tracking differentiator. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] In recent years, with the development of society and technology, the structure of the power system has been constantly changing, and people have higher requirements for the operation of the power system. While the scale of the power grid is constantly expanding, it also brings some safety hazards. The interconnected power grid technology can greatly improve the reliability of power supply, more reasonably allocate electricity, and improve the economic efficiency of power plant operation. However, various oscillations and oscillation accidents have a great impact on the safe operation of the power system. Therefore, oscillation suppression is necessary, and the identification of oscillation parameters is the primary condition for suppressing oscillations, which has far-reaching significance. At present, there are many algorithms that can directly calculate the oscillation parameters, which can make a better approximation to the power oscillation signal.

[0004] The development of the GPS-based synchronous vector measurement unit (Phasor Measurement Unit, PMU) has realized the measurement of voltage and current vectors, laying the foundation for the implementation of wide-area measurement systems. It can also collect real-time information on the power grid, providing conditions for power grid operation, analysis, and monitoring. Therefore, it is widely used in power systems.

[0005] However, the inventors found that the power oscillation data obtained by the PMU will contain a certain amount of noise, which will make the results of the oscillation parameter identification inaccurate, and bring considerable difficulties to the elimination of oscillation. Chinese patent CN13162044A, published on July 23, 2021, discloses a method and device for optimizing FIR filters for frequency measurement by DN-PMU, and characterizes the differential passband performance, harmonic stopband performance and full-band anti-noise performance of the filter to be optimized with technical indicators; establishes a cost function based on the characterized technical indicators; determines the order, fundamental bandwidth and harmonic bandwidth of the filter to be optimized according to the PMU standard; and brings the cost function and the order, fundamental bandwidth and harmonic bandwidth of the filter to be optimized into the calculation formula to generate an optimization problem with constraints. Solve the optimization problem with constraints, and optimize the filter to be optimized based on the solution results. However, the technical solution does not consider the optimal configuration of the PMU, nor does it consider whether the data measured by the PMU can be used to identify low-frequency oscillation parameters, which causes the PMU measurement to be easily contaminated by noise, resulting in inaccurate results, and the calculation will be more complicated and inefficient.

[0006] In addition, many traditional filters usually use various filters, but there will be different degrees of distortion. Therefore, how to improve the accuracy of the algorithm in identifying the oscillation parameters obtained by the PMU has become a problem that needs to be solved urgently in the prior art. Summary of the invention

[0007] In view of the shortcomings of the prior art, the purpose of the present invention is to provide a filtering method and system for PMU power oscillation data based on a tracking differentiator, combining the self-disturbance rejection control with the parameter identification of the low-frequency oscillation signal, designing a tracking differentiator to filter the PMU power oscillation data, accelerating the approximation efficiency, reducing distortion, and thus improving the accuracy of the algorithm in identifying the oscillation parameters.

[0008] In order to achieve the above object, the present invention is implemented through the following technical solutions:

[0009] The first aspect of the present disclosure provides a method for filtering PMU power oscillation data based on a tracking differentiator, comprising the following steps:

[0010] The PMU device is used to obtain the voltage phasor of each node in the system and the current phasor of the branch connected to the node, and the PMU is used to measure the voltage phasor of the node and the current phasor of the branch to measure the active power data;

[0011] Use numerical observability and topological observability to determine the observability of the system;

[0012] After the determination is completed, the PMU is optimized and configured, and the configured PMU is used to obtain data. Based on the obtained data, the properties of the low-frequency oscillation signal are analyzed to obtain the sampling error and the truncation error;

[0013] The sampling error and truncation error are regarded as Gaussian white noise signals with power spectrum density uniformly distributed in the whole frequency domain;

[0014] A low-frequency oscillation signal is selected and added to a Gaussian white noise signal, and a tracking differentiator is constructed according to the Gaussian white noise;

[0015] Add the low-frequency oscillation signal to the tracking differentiator and adjust the parameters to achieve the best filtering effect.

[0016] Furthermore, the specific steps of using the PMU device to perform synchronous measurement of phasors are: using the PMU to measure active power data, filtering the obtained low-frequency oscillation signal through the PMU built-in filter to obtain a signal with interference suppression, converting the AC signal into a digital quantity under the control of a clock pulse to obtain a digital quantity signal with higher precision, and processing the digital quantity signal to obtain the phasor value to realize synchronous measurement.

[0017] Furthermore, during the process of optimizing the configuration of the PMU, a feasible solution in the optimization problem is used as a chromosome and represented by decimal or binary.

[0018] Furthermore, multiple chromosomes are initially selected as the initial population for the optimization problem, and crossover and mutation are performed using the laws of nature to obtain the next population; a fitness function is established when the selection begins, which serves as a condition for population optimization.

[0019] Furthermore, for the configuration of PMU, a fitness function that conforms to the genetic algorithm is designed, and the chromosomes are encoded in binary to configure the PMU.

[0020] Furthermore, the tracking differentiator is as follows:

[0021] r1(k+1)=r1(k)+hr(k)

[0022] r2(k+1)=r2(k)+hfhan(r(k)-u0,r2(k),r,h), where r1 and r2 are output signals; k represents the discretized sampling point; h is called the filter factor, which filters the generated noise; r can determine the tracking speed, so it is called the "speed factor", which is an adjustable parameter; fhan() is a fastest comprehensive nonlinear function.

[0023] Furthermore, the specific structure of fhan() is as follows:

[0024]

[0025] Among them, fhan() represents the fastest control comprehensive function; r1, r2 represent the output signal, h represents the filter factor; r represents the speed factor; a0, d represent the variables of the saturation function; y represents the output of the system; u represents the reference signal of the system; Sign represents the sign function; a1, a2 represent the intermediate transition variables; s y The equivalent formula for the symbolic function; s a Represents an equivalent formula for a symbolic function.

[0026] A second aspect of the present disclosure provides a filtering system for PMU power oscillation data based on a tracking differentiator, comprising:

[0027] The synchronous measurement module is configured to obtain the voltage phasor of each node in the system and the current phasor of the branch connected to the node by using the PMU device, measure the voltage phasor of the node and the current phasor of the branch by using the PMU, and measure the active power data;

[0028] An observability determination module is configured to determine the observability of the system using numerical observability and topological observability;

[0029] The optimization configuration module is configured to optimize the configuration of the PMU after the determination is completed, and use the configured PMU to obtain data, and analyze the properties of the low-frequency oscillation signal based on the obtained data to obtain a sampling error and a truncation error; the sampling error and the truncation error are regarded as a Gaussian white noise signal with a power spectrum density uniformly distributed in the entire frequency domain;

[0030] A differentiator construction module is configured to select a low-frequency oscillation signal and add a Gaussian white noise signal, and construct a tracking differentiator according to the Gaussian white noise;

[0031] The filtering module is configured to add the low-frequency oscillation signal to the tracking differentiator and perform parameter debugging to achieve the best filtering effect.

[0032] A third aspect of the present disclosure provides a medium on which a program is stored. When the program is executed by a processor, the steps in the filtering method for PMU power oscillation data based on a tracking differentiator as described in the first aspect of the present disclosure are implemented.

[0033] The fourth aspect of the present disclosure provides a device, including a memory, a processor, and a program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps in the filtering method for PMU power oscillation data based on a tracking differentiator as described in the first aspect of the present disclosure are implemented.

[0034] The beneficial effects of the above embodiments of the present invention are as follows:

[0035] The present invention proposes a filtering method for PMU power oscillation data based on a tracking differentiator, which combines the self-disturbance rejection control with the parameter identification of the low-frequency oscillation signal. First, the internal structure of the PMU is analyzed, and the voltage phasor of the PMU measurement node and the current phasor of the branch are used to measure the active power data. According to the working principle of the PMU, the system characteristics are analyzed, and the observability of the system is analyzed, and the numerical observability of the system and the system topology observability are analyzed. According to the analysis, the genetic algorithm is combined for configuration, and the configured algorithm is used to obtain the low-frequency oscillation signal using the PMU. In practical applications, the obtained low-frequency oscillation power may have errors. Considering the error and the truncation error, these errors are regarded as white noise signals with power spectrum density distributed in the entire frequency domain, the low-frequency oscillation signal is analyzed to enter the white noise, and the tracking differentiator is used for filtering, thereby improving the accuracy of the algorithm in identifying the oscillation parameters.

[0036] The present invention combines the anti-disturbance control with the parameter identification of the low-frequency oscillation signal, pre-processes the PMU measurement data, and filters the different degrees of noise contained in the oscillation data obtained by the PMU, which can improve the accuracy of the subsequent algorithm for oscillation parameter identification and has important significance in the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0038] Figure 1 An internal structure diagram of a PMU of a filtering system for PMU power oscillation data based on a tracking differentiator in a second embodiment of the present invention;

[0039] Figure 2 A schematic diagram showing a flow chart of a method for filtering PMU power oscillation data based on a tracking differentiator according to a first embodiment of the present invention;

[0040] Figure 3 A diagram showing an ideal low-frequency oscillation signal in a filtering method for PMU power oscillation data based on a tracking differentiator according to a first embodiment of the present invention;

[0041] Figure 4 A diagram showing a low-frequency oscillation signal after white noise is added to a filtering method for PMU power oscillation data based on a tracking differentiator according to a first embodiment of the present invention;

[0042] Figure 5 A diagram showing an ideal low-frequency oscillation signal after adding a tracking differentiator to a filtering method for PMU power oscillation data based on a tracking differentiator according to a first embodiment of the present invention;

[0043] Figure 6 A tracking signal diagram is shown in a filtering method for PMU power oscillation data based on a tracking differentiator according to a first embodiment of the present invention. DETAILED DESCRIPTION

[0044] It should be noted that the following detailed descriptions are illustrative and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present application belongs.

[0045] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0046] PMU provides conditions for power grid operation, analysis and monitoring, and is increasingly attracting widespread attention. In the analysis of power system oscillations, PMU measurement data is mainly used in early warning, online monitoring, disturbance location and anti-interference. However, the power oscillation data obtained by the PMU will contain certain noise, which will make the results of oscillation parameter identification inaccurate, and bring considerable difficulties to eliminating oscillations. Many traditional filters usually use various filters, but there will be different degrees of distortion. The present invention can improve the accuracy of subsequent algorithm for oscillation parameter identification by preprocessing the PMU measurement data and filtering the oscillation data obtained by the PMU that contains different degrees of noise, which is of great significance in the power system.

[0047] The development of GPS technology has provided a very important condition for the realization of system vector measurement. The vector measurement device achieves time synchronization through the second pulse signal of GPS. For traditional measurements, the PMU device can obtain the voltage and current vectors of each point in the system, which plays an important role in the dynamic monitoring and real-time evaluation of the power grid, and plays an important role in the safe and reliable operation of the power grid.

[0048] Embodiment 1:

[0049] The first embodiment of the present disclosure provides a filtering method for PMU power oscillation data based on a tracking differentiator, such as Figure 2 As shown, the following steps are included:

[0050] S1: Analyze the internal structure of PMU, use PMU device to obtain the voltage phasor of each node in the system and the current phasor of the branch connected to the node, use PMU to measure the voltage phasor of the node and the current phasor of the branch, and measure the active power data.

[0051] The PMU device is used to obtain the voltage phasor of each node in the system and the current phasor of the branch connected to the node. For a system, if the voltage vector of a node can be obtained by direct measurement or calculation, the node is called observable.

[0052] S1.1: The specific steps of using the PMU device to perform synchronous measurement of phasors are to use the PMU to measure active power data, filter the obtained low-frequency oscillation signal through the PMU built-in filter to obtain an interference-suppressed signal, convert the AC signal into a digital quantity under the control of the clock pulse to obtain a digital quantity signal with higher precision, and process the digital quantity signal to obtain the phasor value to realize synchronous measurement.

[0053] Preferably, the PMU can complete synchronous measurement of data. The clocks sent by the GPS are generally synchronized, so measurement data with the same time scale can be obtained at different locations, and the consistency of the measurement time section can be achieved.

[0054] S2: Use numerical observability and topological observability to determine the observability of the system. For a system, if the voltage vector of a node can be directly measured or calculated, then the node is observable.

[0055] S2.1: Use numerical observability to determine the observability of the system. If the system has n state quantities and m quantity measurements, then the equation between the state quantities and quantity measurements can be established as:

[0056] z=Hx+v (1)

[0057] Among them, z is the m-dimensional measurement, x is the state quantity, H is the Jacobian matrix between the m×(2n-1)-dimensional measurement and the state quantity, and v is the m-dimensional measurement error vector. The least squares method can be used to obtain the gain matrix, and the observability of the system can be judged by judging the rank of the gain matrix or directly judging the rank of H.

[0058] S2.2: Starting from topology, we use topological observability to determine the observability of the system. The topological graph consists of n nodes and m edges, forming a graph G = (V, E). In the topological graph, V and E represent the set of vertices and edges of the topological graph. In the power system, it is the set of buses and branches. The measurement network constitutes a subgraph of the system G′ = (V′, E′), with If satisfied Then all vertices are contained in the subgraph G′, and the system is observable.

[0059] S3: After the determination is completed, the PMU is optimized and configured, and data is acquired using the configured PMU. Based on the acquired data, the properties of the low-frequency oscillation signal are analyzed to obtain a sampling error and a truncation error.

[0060] Specifically, based on the analysis of the low-frequency oscillation signal based on the acquired data and combined with practical applications, there may be errors, such as PMU measurement errors, errors caused by changes in system load, errors caused by switching operations, errors caused by data transmission delays, etc. The errors caused by data measurement, calculation, and transmission are considered to be sampling errors and truncation errors.

[0061] S3.1: After the observability of the system is determined, the PMU is optimized using the traditional genetic algorithm. For the configuration of the PMU, a fitness function that conforms to the genetic algorithm is designed, and the chromosome is encoded in binary.

[0062] Specifically, the adaptation function is obtained by adding a penalty term, and the function is expressed as follows:

[0063] h=C max -c2m-c1m1

[0064] The fitness function h is a positive number, C max represents a larger constant,

[0065] c1 represents the penalty coefficient, c2 represents the constant adaptation coefficient, m represents the number of installation points, and m1 represents the bus that cannot be observed

[0066] Preferably, if there are N nodes in a system, then N binary codes are needed to represent it. If the binary code is 1, it means that the PMU needs to be installed on this node, and if it is 0, then it does not need to be installed.

[0067] After the configuration is completed, PMU is used to collect data signals.

[0068] S3.2: The data signal acquisition process is as follows: For a system, the power system can be simplified into a set of nonlinear equations as shown below:

[0069]

[0070] x is the n-order state vector, u is the r-order input vector, and y is the m-order output vector. Expressed as the time derivative of the state vector.

[0071] Let x0,u0 be the values ​​of the state vector and input vector at the system equilibrium point, we can get,

[0072] f(x0,u0)=0 (3)

[0073] Apply Δx and Δu near the system equilibrium point, where Δx represents the disturbance applied to the state and Δu represents the disturbance applied to the output; x and u can be rewritten as follows.

[0074] x=x0+Δx

[0075] u=u0+Δu (4)

[0076] When the disturbance is relatively small, the nonlinear functions f(x,u) and y=g(x,u) can be expressed by Taylor series. Ignoring the second-order and higher-order terms of Δx and Δu, and assuming that y is not a direct function of u, the linearized representation of (1) can be written.

[0077]

[0078] In the formula,

[0079]

[0080] Among them, after linearization, A represents the system matrix, B represents the input matrix, and C represents the output matrix. f, represents (x,u) and g represents the corresponding relationship between (x,u) and y; n represents the order is nth order; x represents the system state, and u represents the system input.

[0081] Performing Laplace transform on (5), the transfer function of the system is shown in (7).

[0082]

[0083] Where G(s) represents the transfer function after Laplace transformation of (5); Δy represents the output of the system, s represents the Laplace operator; I represents the identity matrix; R i represents the i-th mode, and μ is always the eigenvalue of A.

[0084] The disturbance near the equilibrium point is equivalent to an impulse or step response, and the corresponding time domain output is shown in (7).

[0085]

[0086] In the formula, A i , are the amplitude and phase of the i-th mode, μ i =σ i +j2πf i , σ i , f i are the attenuation factor and frequency of the ith mode respectively; t represents time; y(t) represents the time domain output.

[0087] According to the obtained y(t), sampling it gives (9):

[0088]

[0089] According to y(k), some methods can be used to solve the oscillation parameters. In the parameter identification of low-frequency oscillation, because of the noise, it will bring errors to the parameter determination process. The tracking differentiator is designed to filter it.

[0090] S4: The sampling error and truncation error are regarded as Gaussian white noise signals whose power spectrum density is uniformly distributed in the entire frequency domain.

[0091] In actual operations such as random changes in system loads, switching operations, and saturation of current transformers (CTs), errors may exist in the measurement, calculation, and transmission of the low-frequency oscillation active power data obtained. The sampling error and truncation error are first regarded as Gaussian white noise with a power spectrum density uniformly distributed in the entire frequency domain. An ideal low-frequency oscillation signal is selected, and a filter is designed based on the properties of Gaussian white noise.

[0092] S4.1: The selected ideal low-frequency oscillation signal is as follows:

[0093]

[0094] Low frequency ideal signal such as Figure 3 As shown, Gaussian white noise is mixed in, and the low-frequency oscillation signal is filtered as follows Figure 4 shown.

[0095] S5: Select a low-frequency oscillation signal and add a Gaussian white noise signal. According to the Gaussian white noise, a tracking differentiator is constructed to realize self-disturbance rejection control. Since the low-frequency oscillation signal is contaminated by white noise, the parameter identification process will lead to inaccurate parameter identification. Therefore, TD (tracking differentiator) is first used to track the input signal. The TD input signal is u(t), and the output signals are r1(t) and r2(t). Among them, r1(t) can extract high-quality input signal u(t), and r2(t) can approximately extract the differential signal of u(t).

[0096] Preferably, when constructing a tracking differentiator, different functions are selected as the tracking differentiator acceleration function to improve the response speed of the tracking differentiator to the input signal, the noise suppression capability and the convergence speed near the equilibrium point.

[0097] S5.1: The tracking differentiator is as follows:

[0098] r1(k+1)=r1(k)+hr(k)

[0099] r2(k+1)=r2(k)+hfhan(r(k)-u0,r2(k),r,h),

[0100] Among them, r1 and r2 are output signals; k represents the discretized sampling point; h is called the filter factor, which filters the generated noise; r can determine the tracking speed, so it is called the "speed factor" and is an adjustable parameter; fhan() is a fastest comprehensive nonlinear function. The specific structure is as follows:

[0101]

[0102] Among them, fhan() represents the fastest control comprehensive function; r1, r2 represent the output signal, h represents the filter factor; r represents the speed factor; a0, d represent the variables of the saturation function; y represents the output of the system; u represents the reference signal of the system; Sign represents the sign function; a1, a2 represent the intermediate transition variables; s y The equivalent formula for the symbolic function; s a Represents an equivalent formula for a symbolic function.

[0103] Take y(t) as the input signal u(t) and send it to the tracking differentiator to obtain the output signals r1 and r2. The subsequent parameter calculation is performed based on the obtained output signals.

[0104] S6: Add the low-frequency oscillation signal to the tracking differentiator and adjust the parameters to achieve the best filtering effect. Figure 5 and Figure 6 The image of the low-frequency oscillation signal with white noise and its tracking signal r1(t) is shown. It can be seen that with the change of time, r1(t) basically tracks the y(t) signal with Gaussian noise and plays a good filtering role. In the subsequent parameter identification of the low-frequency oscillation signal, r1(t) can be used for parameter identification to achieve good results and make the results more accurate.

[0105] Embodiment 2:

[0106] Embodiment 2 of the present disclosure provides a filtering system for PMU power oscillation data based on a tracking differentiator, including:

[0107] The synchronous measurement module is configured to obtain the voltage phasor of each node in the system and the current phasor of the branch connected to the node by using the PMU device, measure the voltage phasor of the node and the current phasor of the branch by using the PMU, and measure the active power data;

[0108] Preferably, the PMU is composed of a current sensor, a filter, an analog-to-digital converter, a data processing unit, and a GPS clock receiver. Figure 1 As shown. The PMU is used to measure the active power data, and the low-frequency oscillation signal is filtered through the built-in filter of the PMU to obtain a signal that suppresses interference. The AC signal is converted into a digital quantity under the control of the clock pulse to obtain a digital quantity signal with high precision. These signals are sent to the data processing unit for data processing to obtain the phasor value. Finally, these data are sent to the monitoring center through the communication structure.

[0109] An observability determination module is configured to determine the observability of the system using numerical observability and topological observability;

[0110] The optimization configuration module is configured to optimize the configuration of the PMU after the determination is completed, and use the configured PMU to obtain data, and analyze the properties of the low-frequency oscillation signal based on the obtained data to obtain a sampling error and a truncation error; the sampling error and the truncation error are regarded as a Gaussian white noise signal with a power spectrum density uniformly distributed in the entire frequency domain;

[0111] A differentiator construction module is configured to select a low-frequency oscillation signal and add a Gaussian white noise signal, and construct a tracking differentiator according to the Gaussian white noise;

[0112] The filtering module is configured to add the low-frequency oscillation signal to the tracking differentiator and perform parameter debugging to achieve the best filtering effect.

[0113] Embodiment three:

[0114] Embodiment 3 of the present disclosure provides a medium on which a program is stored. When the program is executed by a processor, the steps in the filtering method of PMU power oscillation data based on a tracking differentiator as described in Embodiment 1 of the present disclosure are implemented. The steps are:

[0115] S1: Use the PMU device to obtain the voltage phasor of each node in the system and the current phasor of the branch connected to the node, use the PMU to measure the voltage phasor of the node and the current phasor of the branch, and measure the active power data;

[0116] S2: Use numerical observability and topological observability to determine the observability of the system;

[0117] S3: After the determination is completed, the PMU is optimized and configured, and data is acquired using the configured PMU. Based on the acquired data, the properties of the low-frequency oscillation signal are analyzed to obtain a sampling error and a truncation error;

[0118] S4: The sampling error and truncation error are regarded as Gaussian white noise signals with power spectrum density uniformly distributed in the entire frequency domain;

[0119] S5: Select a low-frequency oscillation signal and add a Gaussian white noise signal, and construct a tracking differentiator based on the Gaussian white noise;

[0120] S6: Add the low-frequency oscillation signal to the tracking differentiator and adjust the parameters to achieve the best filtering effect.

[0121] The detailed steps are the same as the filtering method for PMU power oscillation data based on the tracking differentiator provided in the first embodiment, and will not be repeated here.

[0122] Embodiment 4:

[0123] Embodiment 4 of the present disclosure provides a device, including a memory, a processor, and a program stored in the memory and executable on the processor. When the processor executes the program, the steps in the method for filtering PMU power oscillation data based on a tracking differentiator as described in Embodiment 1 of the present disclosure are implemented, and the steps are:

[0124] S1: Use the PMU device to obtain the voltage phasor of each node in the system and the current phasor of the branch connected to the node, use the PMU to measure the voltage phasor of the node and the current phasor of the branch, and measure the active power data;

[0125] S2: Use numerical observability and topological observability to determine the observability of the system;

[0126] S3: After the determination is completed, the PMU is optimized and configured, and data is acquired using the configured PMU. Based on the acquired data, the properties of the low-frequency oscillation signal are analyzed to obtain a sampling error and a truncation error;

[0127] S4: The sampling error and truncation error are regarded as Gaussian white noise signals with power spectrum density uniformly distributed in the entire frequency domain;

[0128] S5: Select a low-frequency oscillation signal and add a Gaussian white noise signal, and construct a tracking differentiator based on the Gaussian white noise;

[0129] S6: Add the low-frequency oscillation signal to the tracking differentiator and adjust the parameters to achieve the best filtering effect.

[0130] The detailed steps are the same as the filtering method for PMU power oscillation data based on the tracking differentiator provided in the first embodiment, and will not be repeated here.

[0131] Those skilled in the art will appreciate that the embodiments of the present disclosure may be provided as methods, systems, or computer program products. Therefore, the present disclosure may take the form of hardware embodiments, software embodiments, or embodiments combining software and hardware. Furthermore, the present disclosure may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) containing computer-usable program code.

[0132] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A filtering method for PMU power oscillation data based on a tracking differentiator, characterized in that: The following steps are involved: The PMU device is used to obtain the voltage phasor of each node in the system and the current phasor of the branch connected to the node, and the PMU is used to measure the voltage phasor of the node and the current phasor of the branch to measure the active power data; Use numerical observability and topological observability to determine the observability of the system; After the determination is completed, the PMU is optimized and configured, and the configured PMU is used to obtain data. Based on the obtained data, the properties of the low-frequency oscillation signal are analyzed to obtain the sampling error and the truncation error; The sampling error and truncation error are regarded as Gaussian white noise signals with power spectrum density uniformly distributed in the whole frequency domain; A low-frequency oscillation signal is selected and added to a Gaussian white noise signal, and a tracking differentiator is constructed according to the Gaussian white noise; Add the low-frequency oscillation signal to the tracking differentiator and adjust the parameters to achieve the best filtering effect; The tracking differentiator takes the following form: r1(k+1)=r1(k)+hr(k) r2(k+1)=r2(k)+hfhan(r(k)-u0,r2(k),r,h), Where r1 and r2 are output signals; k represents the discretized sampling point; h is called the filter factor, which filters the generated noise; r can determine the tracking speed, so it is called the "speed factor" and is an adjustable parameter; fhan() is a fastest comprehensive nonlinear function; The specific structure of fhan() is as follows: Among them, fhan() represents the fastest control comprehensive function; r1, r2 represent the output signal, h represents the filter factor; r represents the speed factor; a0, d represent the variables of the saturation function; y represents the output of the system; u represents the reference signal of the system; Sign represents the sign function; a1, a2 represent the intermediate transition variables; s y The equivalent formula for the symbolic function; s a Represents an equivalent formula for a symbolic function.

2. The method for filtering PMU power oscillation data based on a tracking differentiator as claimed in claim 1, characterized in that: The specific steps of using the PMU device to perform synchronous measurement of phasors are: using the PMU to measure active power data, filtering the obtained low-frequency oscillation signal through the PMU's built-in filter to obtain an interference-suppressed signal, converting the signal into digital quantity under the control of the clock pulse to obtain a digital quantity signal with higher precision, and processing the digital quantity signal to obtain the phasor value to achieve synchronous measurement.

3. The method for filtering PMU power oscillation data based on a tracking differentiator as claimed in claim 1, characterized in that: During the optimization configuration process of PMU, a feasible solution to the optimization problem is taken as a chromosome and represented by decimal or binary.

4. The method for filtering PMU power oscillation data based on a tracking differentiator as claimed in claim 3, characterized in that: Initially select multiple chromosomes as the initial population for the optimization problem, use the laws of nature to perform crossover and mutation, and obtain the next population; establish a fitness function when starting the selection as a condition for population optimization.

5. The method for filtering PMU power oscillation data based on a tracking differentiator as claimed in claim 4, characterized in that: For the configuration of PMU, a fitness function that conforms to the genetic algorithm is designed, and the chromosomes are encoded in binary to configure the PMU.

6. A filtering system for the filtering method of PMU power oscillation data based on tracking differentiator according to any one of claims 1 to 5, characterized in that: include: The synchronous measurement module is configured to obtain the voltage phasor of each node in the system and the current phasor of the branch connected to the node by using the PMU device, measure the voltage phasor of the node and the current phasor of the branch by using the PMU, and measure the active power data; An observability determination module is configured to determine the observability of the system using numerical observability and topological observability; The optimization configuration module is configured to optimize the configuration of the PMU after the determination is completed, and use the configured PMU to obtain data, and analyze the properties of the low-frequency oscillation signal based on the obtained data to obtain a sampling error and a truncation error; the sampling error and the truncation error are regarded as a Gaussian white noise signal with a power spectrum density uniformly distributed in the entire frequency domain; A differentiator construction module is configured to select a low-frequency oscillation signal and add a Gaussian white noise signal, and construct a tracking differentiator according to the Gaussian white noise; The filtering module is configured to add the low-frequency oscillation signal to the tracking differentiator and perform parameter debugging to achieve the best filtering effect.

7. A computer-readable storage medium, characterized in that: A plurality of instructions are stored therein, and the instructions are suitable for being loaded by a processor of a terminal device and executing the filtering method for PMU power oscillation data based on a tracking differentiator as described in any one of claims 1-5.

8. A terminal device, characterized in that: It includes a processor and a computer-readable storage medium, the processor is used to implement various instructions; the computer-readable storage medium is used to store multiple instructions, and the instructions are suitable for being loaded by the processor and executing the filtering method for PMU power oscillation data based on a tracking differentiator according to any one of claims 1-5.

Images