A method for calculating voltage sag characteristics based on goodness of fit test

Abstract

The present invention provides a method for calculating voltage sag characteristics based on a goodness-of-fit test, comprising the following steps: sliding a reference window and a detection window to the right simultaneously with a step size of one sampling point, calculating a goodness-of-fit test statistic between the two windows, thereby detecting transition segments of the voltage sag; calculating the voltage sag type based on the number of detected transition segments; calculating the starting and ending points of the voltage sag based on the transition segments to obtain the voltage sag duration; calculating the phase angles of the starting and ending points of the obtained voltage sag based on the sampling points where the fundamental voltage crosses zero and the derivative is positive, thereby determining the voltage sag waveform points; excluding the transition segments from the phase angle calculation process to avoid computational artifacts caused by them, thereby obtaining the phase jump of the voltage sag. Compared with the prior art, the present invention can calculate all single-event characteristics of a voltage sag in one go, thereby improving the computational efficiency of power quality monitoring equipment.

Description

Technical Field

[0001] The present invention relates to the field of power systems, and in particular to a method for calculating voltage sag characteristics based on goodness of fit test. Background Art

[0002] Residual voltage, duration, sag type, waveform point, and phase jump are the five typical single-event characteristics of voltage sag, which can affect the normal operation of many sensitive devices. Accurately calculating the single-event characteristics of voltage sag is the key to accurately compensating for voltage sags in power systems or users. Among them, residual voltage is considered to be the lowest value of the effective value of the three-phase voltage during the voltage sag process, which is easy to calculate and is therefore not discussed in this invention. In addition, with the development of new power systems, distributed power sources may be disconnected from the grid due to short-circuit faults, resulting in more complex multi-stage voltage sags. Compared with traditional rectangular voltage sags, multi-stage voltage sags present some new basic characteristics, which affect the calculation of single-event characteristics such as sag types.

[0003] Currently, there are several studies both domestically and internationally on methods for calculating various voltage sag characteristics. These include calculating voltage sag type based on three-phase voltage symmetry; calculating duration based on the effective value threshold method; calculating waveform points based on time domain analysis, frequency domain analysis, and combined time-frequency analysis; and calculating phase jumps based on incremental voltage measurements. However, there is no unified, systematic method for calculating the characteristics of a single voltage sag event, and existing methods struggle to calculate the characteristics of multi-stage voltage sags in the context of new power systems. In power systems, power quality monitoring devices must employ multiple different methods to calculate different voltage sag characteristics, resulting in multiple calculations for each voltage sag, which increases memory usage and computation time. Therefore, in practical engineering applications, a method is needed that can calculate all these characteristics and is compatible with multi-stage voltage sags in the context of new power systems, thereby improving the computational efficiency of power quality monitoring equipment. Summary of the Invention

[0004] The purpose of the present invention is to overcome the deficiencies of the above-mentioned prior art and to propose a method for calculating voltage sag characteristics based on goodness of fit test.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] A method for calculating voltage sag characteristics based on goodness of fit test comprises the following steps:

[0007] Slide the reference window and the detection window to the right at the same time with a sampling point as the step size, and calculate the goodness of fit test statistics between the two windows Thus, the transition section of voltage sag is detected;

[0008] Calculating the voltage sag type based on the number of detected transition sections;

[0009] Calculate the starting point and ending point of the voltage sag based on the transition section to obtain the voltage sag duration;

[0010] According to the obtained starting point and ending point of the voltage sag, the phase angles of the starting point and ending point are calculated with reference to the sampling point where the fundamental voltage crosses zero and the derivative is positive, and the voltage sag waveform point is determined;

[0011] During the phase angle calculation, the transition section is excluded to avoid the calculation artifacts caused by it, and the phase jump of the voltage sag is obtained.

[0012] Furthermore, the definitions of the reference window and the detection window specifically include:

[0013] The data in the reference window serves as a reference for the transition period of the impending voltage sag. The probability distribution corresponding to the composed vector is the steady-state voltage or the voltage fluctuation caused by noise during the sag process, which is regarded as the target probability distribution. The detection window is adjacent to the reference window and has the same length as the reference window. It serves as the working window for the goodness-of-fit test.

[0014] Furthermore, the goodness of fit test statistic between the two windows is calculated Thus, the transition section of the voltage sag is detected, including: using the chi-square test in the goodness of fit test to calculate the goodness of fit test statistic between the two windows When the detection window slides across the voltage sag transition period, the test statistic will exceed the critical value

[0015] Furthermore, the voltage sag type is calculated based on the number of detected transition segments, specifically including: when only one transition segment is detected, the voltage sag is classified as a slow recovery type voltage sag; when two transition segments are detected, the voltage sag is classified as a rectangular voltage sag; when a voltage sag has three or more transition segments, the voltage sag is considered to be a multi-stage voltage sag.

[0016] Furthermore, the calculation of the starting point and ending point of the voltage sag based on the transition segment to obtain the voltage sag duration specifically includes: the actual duration of the voltage sag is the time interval from the starting point to the ending point; after the transition segment is detected, the starting point of the rectangular voltage sag is the first sampling point of the first transition segment, and the ending point is the first sampling point of the second transition segment; the starting point of the multi-stage voltage sag is the first sampling point of the first transition segment, and the ending point is the first sampling point of the last transition segment; the starting point of the slow recovery voltage sag is the first sampling point of the transition segment, and the ending time is the time when the effective value voltage recovers to 0.9 pu.

[0017] Furthermore, the phase angles of the starting and ending points are calculated with reference to the sampling points where the fundamental voltage crosses zero and the derivative is positive. For rectangular voltage sags and multi-stage voltage sags, the moments corresponding to the starting and ending points can be calculated from the detected transition segments. For slow-recovery voltage sags, there are no waveform ending point parameters, only waveform starting point parameters, because there is no fault-clearing moment.

[0018] Furthermore, in the phase angle calculation process, the transition section is excluded to avoid the calculation artifacts it produces, and the phase jump of the voltage sag is obtained, specifically including: for the voltage sag caused by a common fault, the transition section is excluded, and the phase jump is the maximum absolute value of the phase angle measured for any phase of the three-phase voltage during the sag event; for the multi-stage voltage sag caused by a multi-stage fault, since there is no evidence that the phase jump in the second stage will affect sensitive equipment, the phase jump in the first stage of the multi-stage voltage sag is used as its phase jump parameter.

[0019] Compared with the prior art, the present invention has the following beneficial effects:

[0020] 1. The present invention proposes a method for calculating voltage sag characteristics based on goodness-of-fit test, which solves the problem that power quality monitoring devices need to use multiple different methods to calculate different characteristics of voltage sags, resulting in multiple calculations for a single voltage sag, which consumes a lot of memory and time. It can calculate all single-event characteristics of voltage sags in one go, thereby improving the computing efficiency of power quality monitoring equipment.

[0021] 2. The present invention proposes a voltage sag characteristic calculation method based on goodness of fit test, which can accurately calculate the single event characteristics of traditional rectangular voltage sag and is compatible with multi-stage voltage sag in the context of new power systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 Schematic diagram of a flow chart of a method for calculating voltage sag characteristics based on goodness of fit test according to an embodiment of the present invention;

[0023] Figure 2 Schematic diagram of reference window and detection window in goodness of fit test;

[0024] Figure 3 This is a schematic diagram of a certain measured voltage sag transition section detection;

[0025] Figure 4 This is a schematic diagram of a measured multi-stage voltage sag waveform;

[0026] Figure 5 This is a schematic diagram of a measured slow recovery voltage sag waveform;

[0027] Figure 6 This is a schematic diagram for calculating the duration of voltage sag;

[0028] Figure 7 Schematic diagram of the phase angle calculation method;

[0029] Figure 8 Schematic diagram of phase jump calculation method;

[0030] Figure 9 Schematic diagram of phase jump of multi-stage voltage sag. DETAILED DESCRIPTION

[0031] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0032] The embodiment of the present invention provides a synchronized phasor measurement method based on dynamic window length, such as Figure 1 As shown, the following steps are included:

[0033] Step S1: Slide the reference window and the detection window to the right at the same time with a sampling point as the step size, and calculate the goodness of fit test statistic between the two windows. Thus, the transition section of voltage sag is detected.

[0034] In this embodiment, the data in the reference window serves as a reference for the transition period of an impending voltage sag. The probability distribution corresponding to the vector formed is the steady-state voltage or the voltage fluctuation caused by noise during a sag, which is considered the target probability distribution. The detection window is adjacent to the reference window and has the same length as the reference window. It serves as the working window for the goodness-of-fit test and is specifically defined as follows.

[0035] W ref ={v(n)|(1≤n≤m)} (1)

[0036] Where W ref is the reference window, v(n) is the instantaneous voltage value, m is the reference window length, and in order to make the elements in the reference window correspond one-to-one with the elements in the detection window defined below, m is the number of sampling points contained in one cycle of the instantaneous value sampling signal.

[0037] W det ={v(n)|(m+1≤n≤2m)} (2)

[0038] Where W detis the detection window, v(n) is the instantaneous voltage value, and m is the detection window length. The schematic diagram of the reference window and the detection window is as follows Figure 2 shown.

[0039] In the detection of voltage sag, the detection window W det The i-th element W in det (i) is equivalent to the observation frequency of the i-th sampling point in the goodness of fit test, and the reference window W ref W in ref (i) is equivalent to the expected frequency of the i-th sampling point in the goodness-of-fit test, so:

[0040]

[0041] in, is the goodness-of-fit test statistic in voltage sag detection, m is the number of sampling points in one sampling cycle of the sampling waveform, and W det (i) is the detection window W det The i-th data in W ref (i) is the reference window W ref The i-th data in .

[0042] Slide the reference window and the detection window to the right at the same time with a sampling point as the step size, and calculate the goodness of fit test statistics between the two windows Thus, the transition section of voltage sag is detected. When the detection window slides over the transition section of voltage sag, the test statistic will exceed the critical value To achieve the detection of transition segments, such as Figure 3 shown.

[0043] Figure 3 (a) shows the instantaneous voltage waveform of a measured voltage sag data, where the solid line represents the instantaneous voltage in the steady-state stage and the dashed line represents the instantaneous voltage during the sag period; Figure 3 (b) shows the trend of the goodness-of-fit test statistic for the measured voltage sag, where the solid line represents the voltage below the critical value. The test statistic The dashed line in the shaded area represents the value exceeding the critical value. The test statistic Test statistic Not exceeding the critical value The part corresponds to the steady-state voltage or the voltage during the voltage sag, and the test statistic Exceeding the critical value The part corresponds to the transition voltage, respectively Figure 3 (c) The solid line part of the voltage RMS waveform and the dashed line part in the shadow.

[0044] Step S2: Calculate the voltage sag type based on the number of detected transition sections.

[0045] When a short circuit occurs in the power system, in most cases, the fault will be cleared in a short time due to the action of the relay protection device, forming a rectangular voltage sag, such as Figure 3 (a) and Figure 3 As shown in (c), for this type of voltage sag, the effective value voltage has two transition sections, corresponding to the occurrence and clearance of the fault respectively.

[0046] When power system parameters change during a short circuit fault, such as when a protective relay isolates the fault or the fault progresses to another stage before being cleared, it may cause multi-stage voltage sags, e.g. Figure 4 shown. Figure 4 (a) is a waveform of the instantaneous voltage value of a measured multi-stage voltage sag, where the steady-state voltage is marked as a solid line, the voltage in the first stage of the voltage sag event is marked as a dotted line, and the voltage in the second stage of the voltage sag is marked as a dashed line. Figure 4 (b) shows the RMS voltage of the measured multi-stage voltage sag, which has three transition segments. For this type of voltage sag, the RMS voltage has two or more transition segments. The first transition segment corresponds to the occurrence of a short-circuit fault, and the last transition segment corresponds to the clearing of the fault. The remaining transition segments correspond to changes in power system parameters or the development of a fault.

[0047] Slow recovery voltage sag refers to a sudden drop in voltage caused by the high current generated by the excitation of an unloaded transformer or the starting of a large induction motor. For this type of sag, the voltage recovery process is a slow process rather than a step change. Therefore, a slow recovery voltage sag has only one transition period, such as Figure 5 shown. Figure 5 The RMS voltage of the measured slow-recovery voltage sag is consistent with the analysis above. This voltage sag has only a single transition segment. Because this type of voltage sag has no clear fault-clearing moment, its waveform parameters only include a starting point, not an end point. Furthermore, the occurrence of a slow-recovery voltage sag is independent of changes in the grid structure, so a slow-recovery voltage sag lacks phase jump characteristics.

[0048] According to the above analysis, the voltage sag type can be calculated according to the detected transition section, as follows:

[0049] As shown:

[0050]

[0051] where N Tis the number of voltage drop transitions. When only one transition is detected, the voltage drop is classified as a slow recovery voltage sag; when two transitions are detected, the voltage sag is classified as a rectangular voltage sag; when a voltage sag has three or more transitions, it is considered a multi-stage voltage sag.

[0052] Step S3: Calculate the starting point and ending point of the voltage sag based on the transition section to obtain the duration of the voltage sag.

[0053] The actual duration of a voltage sag is the time interval from the starting point to the end point. After the transition section is detected, the duration of different types of voltage sags can be calculated as follows:

[0054] 1) Rectangular voltage sag (N T =2). For a rectangular voltage sag, the starting point is the first sampling point of the first transition segment, and the ending point is the first sampling point of the second transition segment, such as Figure 6 As shown in (a) in the figure. The duration is the time interval between the start point and the end point, as shown in the following formula:

[0055] T=t r -t i (5)

[0056] Where, T is the duration of voltage sag, t r is the time corresponding to the end point, t i is the time corresponding to the starting point.

[0057] 2) Multi-stage voltage sag (N T >2). For multi-stage voltage sag, the starting point is the

[0058] The first sampling point of a transition segment, the end point is the first sampling point of the last transition segment, such as Figure 6 As shown in (b) in the figure, duration is the time interval between the start point and the end point, which is calculated by formula (5).

[0059] 3) Slow recovery type (N T =1). For slow recovery voltage sag, since there is no fault clearing moment, the present invention defines the end time of this type of voltage sag as the moment when the effective value voltage recovers to 0.9 pu, and the starting point is the first sampling point of the transition period, such as Figure 6 As shown in (c) in the figure, duration is the time interval between the start point and the end point, which is calculated by formula (5).

[0060] Step S4: Based on the obtained starting point and ending point of the voltage sag, the phase angles of the starting point and ending point are calculated with reference to the sampling points where the fundamental voltage crosses zero and the derivative is positive, and the voltage sag waveform points are determined.

[0061] The starting point and ending point of the waveform point are two parameters related to the occurrence and clearance of the fault. Determining the moment of fault occurrence (the sampling point number corresponding to the starting point) and the moment of fault clearance (the sampling point number corresponding to the ending point) is the key to calculating the waveform point. After obtaining these two moments, the phase angles of the starting point and the ending point are calculated with reference to the sampling point where the fundamental voltage passes through the zero crossing point and the derivative is positive.

[0062] First pass Figure 7 This section describes how to calculate the phase angles at the start and end points using upward zero crossings. The leftmost black dot in the figure is the reference point for the start point; the phase angle at this point is considered to be 0°. Based on this reference point, the phase angle at the start point is calculated to be 90°. Similarly, the rightmost black dot in the figure is the reference point for the end point; the phase angle at this point is considered to be 0°. Based on this reference point, the phase angle at the end point is calculated to be 180°.

[0063] Next, for different types of sags, the waveform point feature calculation steps are as follows:

[0064] 1) Rectangular voltage sag / multi-stage voltage sag (N T ≥2). For rectangular voltage sags and multi-stage voltage sags, the time corresponding to the start and end points can be calculated from the detected transition segments. Therefore, for these two types of sags, the waveform start and end points in the waveform points can be calculated according to equations (6) and (7).

[0065] POW-I=Angle(t i ) (6)

[0066] POW-R=Angle(t r ) (7)

[0067] Where POW-I is the waveform starting point in the waveform point, POW-R is the waveform ending point in the waveform point, and Angle is Figure 7 The phase angle calculation method shown, t i is the time corresponding to the starting point, t r is the time corresponding to the end point.

[0068] 2) Slow recovery voltage sag (N T =1). For a slow recovery voltage sag, there is no waveform end point parameter because there is no fault clearing moment. The waveform start point parameter can be calculated according to equation (6). As mentioned above, the start time can be calculated based on the detected transition segment.

[0069] Step S5: During the phase angle calculation process, the transition section is excluded to avoid calculation artifacts caused by the transition section, and the phase jump of the voltage sag is obtained.

[0070] The most important task in phase jump calculation is to exclude the transition section to avoid calculation artifacts generated in the transition section during phase angle calculation. Figure 8 The figure shows the possible calculation artifacts that may appear during the phase jump calculation process, where the true value of the phase jump and the maximum calculation artifact are marked with circles. The true value of the phase jump is the stable value during the voltage sag. Figure 8 In the process of calculating the voltage phase angle, if the transition section is not removed, the calculation artifacts will lead to calculation errors of the phase jump. Figure 8 The calculation artifact with the largest absolute value is -23.92°, which will lead to a phase jump calculation error of 10.98°, which is unacceptable in practical engineering.

[0071] To this end, for voltage sags caused by common faults, after detecting the transition section using the method proposed in the present invention, the transition section is excluded, and the phase jump can be calculated as the maximum absolute value of the phase angle measured for any phase of the three-phase voltage during the sag event.

[0072] In addition, for multi-stage voltage sags caused by multi-stage faults, the development of the fault will cause the power system parameters to change again. When the voltage sag stage changes, the phase jump parameters will also change once, such as Figure 9 As shown in the figure, the phase jump in the first stage is -22.13°, and the phase jump in the second stage is -14.28°. Since there is no evidence that the phase jump in the second stage will affect sensitive equipment, in order to calculate the phase jump single event characteristics, the present invention uses the phase jump in the first stage of the multi-stage voltage sag as its phase jump parameter. Figure 9 It can be seen that if the phase jump in the second stage needs to be calculated, the method proposed in the present invention only needs to calculate the phase jump between the last two transition segments to achieve it.

[0073] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A method for calculating voltage sag characteristics based on goodness of fit test, characterized in that: The following steps are involved: Slide the reference window and the detection window to the right at the same time with a sampling point as the step size, and calculate the goodness of fit test statistics between the two windows , thereby detecting the transition section of voltage sag; Calculating the voltage sag type based on the number of detected transition sections; Calculate the starting point and ending point of the voltage sag based on the transition section to obtain the voltage sag duration; According to the obtained starting point and ending point of the voltage sag, the phase angles of the starting point and ending point are calculated with reference to the sampling point where the fundamental voltage crosses zero and the derivative is positive, and the voltage sag waveform point is determined; During the phase angle calculation, the transition section is excluded to avoid the calculation artifacts caused by it, and the phase jump of the voltage sag is obtained; The calculation of the goodness-of-fit test statistic between two windows , thereby detecting the transition section of the voltage sag, including: using the chi-square test in the goodness of fit test to calculate the goodness of fit test statistic between the two windows , when the detection window slides across the voltage sag transition period, the test statistic will exceed the critical value ; The voltage sag type is calculated based on the number of detected transition segments, specifically including: when only one transition segment is detected, the voltage sag is classified as a slow recovery type voltage sag; when two transition segments are detected, the voltage sag is classified as a rectangular voltage sag; when a voltage sag has three or more transition segments, the voltage sag is considered a multi-stage voltage sag; In the phase angle calculation process, the transition section is excluded to avoid calculation artifacts caused by it, and the phase jump of the voltage sag is obtained. Specifically, for the voltage sag caused by a common fault, the transition section is excluded, and the phase jump is the maximum absolute value of the phase angle measured for any phase of the three-phase voltage during the sag event; for the multi-stage voltage sag caused by a multi-stage fault, since there is no evidence that the phase jump in the second stage will affect sensitive equipment, the phase jump in the first stage of the multi-stage voltage sag is used as its phase jump parameter.

2. The method for calculating voltage sag characteristics based on goodness of fit test according to claim 1, characterized in that: The definitions of the reference window and detection window include: The data in the reference window serves as a reference for the transition period of the impending voltage sag. The probability distribution corresponding to the composed vector is the steady-state voltage or the voltage fluctuation caused by noise during the sag process, which is regarded as the target probability distribution. The detection window is adjacent to the reference window and has the same length as the reference window. It serves as the working window for the goodness-of-fit test.

3. The method for calculating voltage sag characteristics based on goodness of fit test according to claim 1, characterized in that: The calculation of the starting point and ending point of the voltage sag based on the transition segment to obtain the voltage sag duration specifically includes: the actual duration of the voltage sag is the time interval from the starting point to the ending point; after the transition segment is detected, the starting point of the rectangular voltage sag is the first sampling point of the first transition segment, and the ending point is the first sampling point of the second transition segment; the starting point of the multi-stage voltage sag is the first sampling point of the first transition segment, and the ending point is the first sampling point of the last transition segment; the starting point of the slow recovery voltage sag is the first sampling point of the transition segment, and the ending time is the time when the effective value voltage recovers to 0.9 pu.

4. The method for calculating voltage sag characteristics based on goodness of fit test according to claim 1, wherein: The phase angles of the starting and ending points are calculated with reference to the sampling points where the fundamental voltage crosses zero and the derivative is positive. For rectangular voltage sags and multi-stage voltage sags, the times corresponding to the starting and ending points are calculated through the detected transition segments. For slow recovery voltage sags, there are no waveform ending point parameters, only waveform starting point parameters, because there is no fault clearing moment.

Images