Frequency-adaptive sampling anomaly identification and correction method, system and equipment

Abstract

The invention discloses a frequency-adaptive sampling anomaly identification and correction method, system and equipment, and the method comprises the steps: carrying out the pre-filtering processing of each channel of an original sampling pool, counting the number of zero points, calculating the system frequency in real time, dynamically tracking the change of the system frequency, and carrying out the calculation of the system frequency; updating an electrical angle between two continuous sampling points of the relay protection device under the system frequency in real time; according to the real-time system frequency, predicting peak and valley moments of a sampling signal under a normal operation condition; comparing the difference between the generation moment of the actual sampling point and the peak and valley moments of the sampling signal under the normal operation condition, and identifying a suspected dead pixel; based on the suspected dead pixels, identifying the dead pixels by using a criterion; and comparing the amplitude and the change rate of the normal sampling points in the previous period with the sampling points in the single period after the current dead pixels one by one, and recording the number of the dead pixels. According to the method, calculation errors caused by frequency deviation can be reduced, the abnormal sampling points are effectively recognized and corrected, and then the reliability of calculation and logic judgment of the relay protection device is guaranteed.

Description

Technical Field

[0001] This invention relates to the field of power grid safety technology, specifically to a frequency-adaptive sampling anomaly identification, correction method, system, and device. Background Technology

[0002] New power systems are characterized by wide frequency, weak feed, and high harmonics. After equipment failure, the DC component and transient component increase and the decay time becomes longer. The number of power electronic devices connected to the grid is increasing. The joint networking with fluctuating and intermittent power sources such as wind and solar makes relay protection devices operate in more complex abnormal conditions, increasing the probability of interference to relay protection devices. Therefore, the application of new algorithms and new architectures has become the goal.

[0003] In practical engineering, the sampling channel may be affected by communication noise, A / D conversion failure, transformer saturation, and clock synchronization loss, resulting in bad points and dropped points in the sampling points.

[0004] Current defect detection algorithms assume the sampled signal is a power frequency sine wave and calculate the phase angle value within a short data window to identify defective pixels using phase information; or they take the value from the previous cycle and compare it with the value from the next cycle; or they compare the sum of the absolute values ​​of several sample points with a threshold value. The time prediction of sampling points is usually based on a fixed power frequency condition and fails to incorporate real-time frequency offset, resulting in insufficient accuracy under low-frequency and high-frequency disturbances, making it difficult to guarantee the reliability and accuracy of sampling. Summary of the Invention

[0005] To address the aforementioned issues, this invention proposes a frequency-adaptive sampling anomaly identification and correction method, system, and device capable of accurately identifying and correcting sampling defects when the system frequency fluctuates.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solution:

[0007] The frequency-adaptive sampling anomaly identification and correction method of the present invention includes:

[0008] The original sampling pool is processed, the system frequency is calculated in real time, the changes in the system frequency are dynamically tracked, and the electrical angle between two consecutive sampling points of the relay protection device under the system frequency is updated in real time.

[0009] Based on the real-time system frequency, predict the peak and trough times of the sampled signal under normal operating conditions;

[0010] By comparing the differences between the peak and trough times of the actual sampling points and the sampling signals under normal operating conditions, suspected bad pixels can be identified.

[0011] Based on suspected bad pixels, use criteria to identify bad pixels;

[0012] The amplitude and rate of change of the normal sampling points in the previous cycle are compared one by one with the sampling points in the single cycle after the current bad point, and the number of bad points is recorded.

[0013] Based on the number of bad pixels, a corresponding scheme is adopted to correct the sampling point data.

[0014] A further improvement of this invention is that the expression for calculating the electrical angle between two consecutive sampling points at the system frequency is:

[0015] ;

[0016] in: For system frequency, This refers to the sampling time interval of the relay protection device. The electrical angle between the two sampling points.

[0017] A further improvement of the present invention is that the identification criteria for suspected bad pixels are:

[0018] ;

[0019] ;

[0020] in: For suspected bad pixel moments, This represents the peak moment of the continuously sampled signal under normal operating conditions. The valley moment of the continuously sampled signal under normal operating conditions. This is the tolerance factor. The period of the sampled signal.

[0021] A further improvement of the present invention is that: based on suspected bad pixels, bad pixels are identified using criteria, specifically including:

[0022] Calculate the rate of change of difference between any two adjacent sampling points;

[0023] Determine whether the differential change rate associated with the suspected bad pixel is greater than the maximum change rate threshold of the sinusoidal signal. If so, the suspected bad pixel is a bad pixel caused by sampling abnormality.

[0024] Determine whether the rate of change of the forward difference and the rate of change of the backward difference of the suspected bad pixel are opposite in sign. If so, the suspected bad pixel is a bad pixel caused by sampling abnormality.

[0025] The amplitude of the intermediate theoretical sampling point is calculated by taking samples from points spaced apart on either side of the suspected bad pixel. The amplitude of the suspected bad pixel is compared with that of the intermediate theoretical sampling point. If the following criteria are met, the suspected bad pixel is determined to be a bad pixel caused by sampling anomaly:

[0026] ;

[0027] in: for The amplitude at each sampling point at any given time. The sampling point at that time was suspected to be a bad pixel. for The amplitude at each sampling point at any given time. for The amplitude at each sampling point at any given time. For coefficients, The electrical angle between the two sampling points;

[0028] The second-order difference is calculated using suspected bad pixels and sampling points on both sides. Bad pixels are identified based on the degree of abrupt changes in the signal at a single point. The expression is as follows:

[0029] ;

[0030] ;

[0031] in, It is a second-order difference. For coefficients, The sampling frequency of the relay protection device. For system frequency, This represents the amplitude of a sinusoidal signal.

[0032] A further improvement of the present invention is that, based on the number of bad pixels, a corresponding scheme is used to correct the sampling point data, specifically including:

[0033] For cases with only a single bad pixel and two discontinuous bad pixels, the amplitude of the intermediate theoretical sampling point is calculated using the normal sampling points on both sides of the bad pixel, and the bad pixel is replaced with the intermediate theoretical sampling point.

[0034] In the case of two consecutive bad pixels, take the normal sampling points on both sides of the two consecutive bad pixels, use the fitting matrix to calculate the amplitude of the two consecutive theoretical sampling points, and replace the two consecutive bad pixels with the two consecutive theoretical sampling points.

[0035] For cases with three or more defective points, the sampling data of all sampling points in the following cycle are replaced with the sampling data of the normal sampling points in the previous cycle, and the corresponding protection module in the relay protection device outputs after a delay of one electrical cycle.

[0036] A further improvement of this invention is that the calculation expression for intermediate theoretical sampling points is:

[0037] ;

[0038] in: The amplitude at the intermediate theoretical sampling point. for The amplitude at each sampling point at any given time. for The amplitude at each sampling point at any given time. The electrical angle between the two sampling points.

[0039] A further improvement of this invention is that the expression for the fitting matrix is:

[0040] ;

[0041] in: , for , The amplitude of the theoretical sampling point at time . for The amplitude at each sampling point at any given time. for The amplitude at each sampling point at any given time. The electrical angle between the two sampling points.

[0042] The frequency-adaptive sampling anomaly identification and correction system of the present invention includes:

[0043] The system frequency calculation module is used to process the original sampling pool, calculate the system frequency in real time, dynamically track changes in the system frequency, and update the electrical angle between two consecutive sampling points of the relay protection device under the system frequency in real time.

[0044] The prediction module is used to predict the peak and trough times of the sampled signal under normal operating conditions based on the real-time system frequency.

[0045] The suspected bad pixel identification module is used to compare the difference between the peak and valley times of the actual sampling point generation time and the sampling signal under normal operating conditions to identify suspected bad pixels.

[0046] The defect detection module is used to identify defective pixels using criteria.

[0047] The comparison module is used to compare the amplitude and rate of change of the normal sampling points in the previous cycle with the sampling points in the single cycle after the current bad point, and record the number of bad points.

[0048] The correction module is used to correct the sampled data based on the number of bad pixels using a corresponding scheme.

[0049] The electronic device of the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the above-described frequency adaptive sampling anomaly identification and correction method.

[0050] The present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described frequency-adaptive sampling anomaly identification and correction method.

[0051] The beneficial effects of this invention are as follows: This invention improves the accuracy of bad pixel detection. By calculating intermediate theoretical values ​​from adjacent points, first-order differences, second-order differences, and comparing peak and trough times, it effectively identifies various types of bad pixels such as spikes, flying spots, and stuck spots, avoiding misjudgments of small-amplitude disturbances found in traditional methods. This invention has strong anti-interference capabilities. By tracking the real-time system frequency and adaptively adjusting according to different operating conditions and signal characteristics, it ensures stable performance under various working environments. When faced with frequent disturbances, the tolerance coefficient and bad pixel length counting can avoid misjudgments or over-repair, improving the stability and reliability of bad pixel judgment and correction. Attached Figure Description

[0052] Figure 1 This is a flowchart of the method in an embodiment of the present invention; Figure 2 This is a schematic diagram of defect identification and correction in an embodiment of the present invention. Detailed Implementation

[0053] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0054] like Figures 1 to 2 As shown, this embodiment is a frequency-adaptive sampling anomaly identification and correction method, including:

[0055] Step 1: Process the original sampling cell, calculate the system frequency in real time, dynamically track changes in the system frequency, and update the electrical angle between two consecutive sampling points of the relay protection device at the system frequency in real time.

[0056] Step 2: Based on the real-time system frequency, predict the peak and trough times of the sampled signal under normal operating conditions;

[0057] Step 3: Compare the differences between the peak and valley times of the actual sampling point generation time and the sampling signal under normal operating conditions to identify suspected bad pixels;

[0058] Step 4: Based on suspected bad pixels, identify bad pixels using criteria, specifically including:

[0059] Bad pixels are identified by calculating the rate of change between two adjacent sampling points.

[0060] Defective pixels are identified by judging the positive or negative relationship between the rate of change on both sides of a suspected defective pixel.

[0061] By calculating the amplitude of the intermediate theoretical sampling point through sampling points at intervals on both sides of the suspected bad point, the bad point is identified by comparing the amplitude relationship between the suspected bad point and the theoretical sampling point.

[0062] The second-order difference is calculated by using suspected bad points and sampling points on both sides, and bad points are identified based on the degree of abrupt change in the signal at a single point.

[0063] Step 5: Compare the amplitude and rate of change of the normal sampling points in the previous cycle with the sampling points in the single cycle after the current bad point, and record the number of bad points.

[0064] Step 6: Based on the number of bad pixels, use the corresponding scheme to correct the sampling point data.

[0065] In step 1, the system frequency is calculated in real time by zero-crossing counting, the changes in system frequency are dynamically tracked, and the electrical angle between two consecutive sampling points is updated. The expression for calculating the electrical angle is:

[0066] ;

[0067] in: For system frequency, This refers to the sampling time interval of the relay protection device. The electrical angle between the two sampling points.

[0068] This embodiment predicts the peak and trough times of the continuously sampled signal under fault-free normal operation by locating the first rising or falling zero point and combining it with the current system frequency. The predicted peak and trough times are then stored in memory allocated during relay protection device initialization. The calculation expressions for the peak and trough times of the continuously sampled signal under normal operation are as follows:

[0069] ;

[0070] ;

[0071] in: For continuously sampled signals under normal operating conditions, The valley moment of the continuously sampled signal under normal operating conditions. To reach midnight, This refers to the system frequency.

[0072] In this embodiment, suspected bad pixels are identified and located by comparing the differences between the time when a suspected bad pixel occurs and the peak and trough times of the continuously sampled signal under normal operating conditions. The identification condition for suspected bad pixels, i.e., bad pixel discrimination expression 1, is:

[0073] ;

[0074] ;

[0075] in: For suspected bad pixel moments, This represents the peak moment of the continuously sampled signal under normal operating conditions. The valley moment of the continuously sampled signal under normal operating conditions. This is the tolerance factor. The period of the sampled signal.

[0076] The expression for calculating the rate of change between two adjacent sampling points is:

[0077] ;

[0078] in, for and The rate of change between sampling points at time intervals , They are respectively and The amplitude at each sampling point at time t, where The sampling point at any given time is the current sampling point to be identified.

[0079] In sinusoidal signal Below, where t is time, the expression for calculating the maximum theoretical first-order difference is:

[0080] ;

[0081] in, The sampling frequency of the relay protection device. The amplitude of the sine wave signal. It is the first-order difference of the maximum theory.

[0082] The sampling frequency of the relay protection device is much higher than the system frequency. Approximation:

[0083] ;

[0084] The maximum theoretical first difference is:

[0085] ;

[0086] Bad pixels cause the actual rate of change to exceed the threshold; the maximum rate of change of the normal sampling curve is multiplied by a coefficient.

[0087] This is the threshold. The bad pixel detection expression 2 is:

[0088] ;

[0089] A sharp point or defect caused by a bad pixel is characterized by a change in the sign of the rate of change on both sides of the bad pixel. In this embodiment, the opposite sign of the rate of change is used as the bad pixel criterion, that is, the bad pixel discrimination expression 3 is:

[0090] ;

[0091] in: The rate of change between the sampling points at time n-1 and time n. for and The rate of change between sampling points at time intervals It is a symbolic function.

[0092] To avoid the amplitude increase process when a real fault occurs, the coefficient is taken in the most extreme short-circuit case. The bad pixel detection expression 4 is:

[0093] ;

[0094] in: for The amplitude of the sampling point at any given time.

[0095] In this embodiment, taking the suspected bad spot as the center, and selecting two adjacent points, the second-order difference expression composed of three consecutive sampling points is as follows:

[0096] ;

[0097] in, It is a second-order difference.

[0098] The second-order difference (second-order rate of change) reflects the curvature of the sampled signal near a single point. Spikes caused by sampling anomalies cause the actual second-order difference to exceed a threshold. The maximum second-order rate of change of the normal sampling curve is then multiplied by a coefficient. Given the threshold, the bad pixel detection expression 5 is:

[0099] ;

[0100] Sampling points that satisfy bad pixel discrimination expressions 1 to 5 are identified as bad pixels caused by sampling anomalies. A bad pixel counter is activated, and the sampling point data is corrected based on the number of bad pixels. The specific correction method is as follows:

[0101] For cases with only a single bad pixel or two discontinuous bad pixels, the amplitude of the intermediate theoretical sampling point is calculated using the normal sampling points on both sides of the bad pixel, and the bad pixel is replaced with the intermediate theoretical sampling point.

[0102] The formula for calculating intermediate theoretical sampling points is:

[0103] ;

[0104] in: This represents the amplitude at the intermediate theoretical sampling point.

[0105] In the case of two consecutive bad pixels, take the normal sampling points on both sides of the two bad pixels, calculate the amplitude of the two consecutive theoretical sampling points using the fitting matrix, and replace the two consecutive bad pixels with the two consecutive theoretical sampling points. The HMI board of the relay protection device displays "AD sampling abnormality" and sends a signal to the monitoring background. The expression of the fitting matrix is:

[0106] ;

[0107] in: , for , The amplitude of the theoretical sampling point at time . for The amplitude of the sampling point at any given time.

[0108] For devices with three or more defective points, the sampling data of all sampling points in the following cycle of the defective points is replaced with the sampling data of the normal sampling points in the previous cycle, and an "AD sampling abnormality" is issued. The relay protection device has several protection modules set according to the operating conditions. Each protection module is associated with a different analog channel. When the sampling point of the analog channel is judged to be abnormal, the protection module associated with the analog channel blocks the output of one electrical cycle, so as to realize the individual processing of the defective points.

[0109] The above method is implemented through the frequency-adaptive sampling anomaly identification and correction system in this embodiment, which includes:

[0110] The system frequency calculation module is used to process the original sampling pool, calculate the system frequency in real time, dynamically track changes in the system frequency, and update the electrical angle between two consecutive sampling points of the relay protection device under the system frequency in real time.

[0111] The prediction module is used to predict the peak and trough times of the sampled signal under normal operating conditions based on the real-time system frequency.

[0112] The suspected bad pixel identification module is used to compare the difference between the peak and valley times of the actual sampling point generation time and the sampling signal under normal operating conditions to identify suspected bad pixels.

[0113] The defect detection module is used to identify defective pixels using criteria.

[0114] The comparison module is used to compare the amplitude and rate of change of the normal sampling points in the previous cycle with the sampling points in the single cycle after the current bad point, and record the number of bad points.

[0115] The correction module is used to correct the sampled data based on the number of bad pixels using a corresponding scheme.

[0116] It will be understood by those skilled in the art that, unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. It should also be understood that terms such as those defined in general dictionaries should be understood to have the same meaning as in the context of the prior art, and should not be interpreted in an idealized or overly formal sense unless defined as herein.

[0117] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A frequency-adaptive sampling anomaly identification and correction method, characterized in that: include: The analog channels of the original sampling pool are pre-filtered, the number of zero points is counted and the system frequency is calculated in real time. The changes in the system frequency are dynamically tracked and the electrical angle between two consecutive sampling points of the relay protection device under the system frequency is updated in real time. Based on the real-time system frequency, predict the peak and trough times of the sampled signal under normal operating conditions; By comparing the differences between the peak and trough times of the actual sampling points and the sampling signals under normal operating conditions, suspected bad pixels can be identified. Based on suspected bad pixels, use criteria to identify bad pixels; The amplitude and rate of change of the normal sampling points in the previous cycle are compared one by one with the sampling points in the single cycle after the current bad point, and the number of bad points is recorded. Based on the number of bad pixels, a corresponding scheme is adopted to correct the sampling point data.

2. The frequency-adaptive sampling anomaly identification and correction method according to claim 1, characterized in that: The expression for calculating the electrical angle between two consecutive sampling points at the system frequency is: ； in: For system frequency, This refers to the sampling time interval of the relay protection device. The electrical angle between the two sampling points.

3. The frequency-adaptive sampling anomaly identification and correction method according to claim 1, characterized in that: The criteria for identifying suspected bad pixels are: ； ； in: For suspected bad pixel moments, This represents the peak moment of the continuously sampled signal under normal operating conditions. The valley moment of the continuously sampled signal under normal operating conditions. This is the tolerance factor. The period of the sampled signal.

4. The frequency-adaptive sampling anomaly identification and correction method according to claim 1, characterized in that: Based on suspected bad pixels, bad pixels are identified using criteria, specifically including: Calculate the rate of change of difference between any two adjacent sampling points; Determine whether the differential change rate associated with the suspected bad pixel is greater than the maximum change rate threshold of the sinusoidal signal. If so, the suspected bad pixel is a bad pixel caused by sampling abnormality. Determine whether the rate of change of the forward difference and the rate of change of the backward difference of the suspected bad pixel are opposite in sign. If so, the suspected bad pixel is a bad pixel caused by sampling abnormality. The amplitude of the intermediate theoretical sampling point is calculated by taking samples from points spaced apart on either side of the suspected bad pixel. The amplitude of the suspected bad pixel is compared with that of the intermediate theoretical sampling point. If the following criteria are met, the suspected bad pixel is determined to be a bad pixel caused by sampling anomaly: ； in: for The amplitude at each sampling point at any given time. The sampling point at that time was suspected to be a bad pixel. for The amplitude at each sampling point at any given time. for The amplitude at each sampling point at any given time. For coefficients, The electrical angle between the two sampling points; The second-order difference is calculated using suspected bad pixels and sampling points on both sides. Bad pixels are identified based on the degree of abrupt changes in the signal at a single point. The expression is as follows: ； ； in, It is a second-order difference. For coefficients, The sampling frequency of the relay protection device. For system frequency, This represents the amplitude of a sinusoidal signal.

5. The frequency-adaptive sampling anomaly identification and correction method according to claim 4, characterized in that: Based on the number of bad pixels, a corresponding scheme is adopted to correct the sampled data, specifically including: For cases with only a single bad pixel and two discontinuous bad pixels, the amplitude of the intermediate theoretical sampling point is calculated using the normal sampling points on both sides of the bad pixel, and the bad pixel is replaced with the intermediate theoretical sampling point. In the case of two consecutive bad pixels, take the normal sampling points on both sides of the two consecutive bad pixels, use the fitting matrix to calculate the amplitude of the two consecutive theoretical sampling points, and replace the two consecutive bad pixels with the two consecutive theoretical sampling points. For cases with three or more defective points, the sampling data of all sampling points in the following cycle are replaced with the sampling data of the normal sampling points in the previous cycle, and the corresponding protection module in the relay protection device outputs after a delay of one electrical cycle.

6. The frequency-adaptive sampling anomaly identification and correction method according to claim 5, characterized in that: The formula for calculating intermediate theoretical sampling points is: ； in: The amplitude at the intermediate theoretical sampling point. for The amplitude at each sampling point at any given time. for The amplitude at each sampling point at any given time. The electrical angle between the two sampling points.

7. The frequency-adaptive sampling anomaly identification and correction method according to claim 5, characterized in that: The expression for the fitting matrix is: ； in: , for , The amplitude of the theoretical sampling point at time . for The amplitude at each sampling point at any given time. for The amplitude at each sampling point at any given time. The electrical angle between the two sampling points.

8. A frequency-adaptive sampling anomaly identification and correction system based on the method described in claims 1 to 7, characterized in that: include: The system frequency calculation module is used to process the original sampling pool, calculate the system frequency in real time, dynamically track changes in the system frequency, and update the electrical angle between two consecutive sampling points of the relay protection device under the system frequency in real time. The prediction module is used to predict the peak and trough times of the sampled signal under normal operating conditions based on the real-time system frequency. The suspected bad pixel identification module is used to compare the difference between the peak and valley times of the actual sampling point generation time and the sampling signal under normal operating conditions to identify suspected bad pixels. The defect detection module is used to identify defective pixels using criteria. The comparison module is used to compare the amplitude and rate of change of the normal sampling points in the previous cycle with the sampling points in the single cycle after the current bad point, and record the number of bad points. The correction module is used to correct the sampled data based on the number of bad pixels using a corresponding scheme.

9. An electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, it implements the steps of the frequency-adaptive sampling anomaly identification and correction method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by the processor, it implements the steps of the frequency-adaptive sampling anomaly identification and correction method as described in any one of claims 1 to 7.

Images