A method and apparatus for real-time calculation of short-time flicker

Abstract

This invention relates to a method and apparatus for real-time calculation of short-time flicker, belonging to the field of power system measurement technology. This invention only requires converting the cumulative probability needed for short-time flicker calculation into a probability value adapted to an adaptive method, using this probability value as the quantile probability, and then using the obtained quantile probability, the range of the instantaneous flicker value, and the instantaneous flicker value to calculate the quantile value corresponding to each quantile probability; finally, based on the obtained quantile probability and the corresponding quantile value, real-time short-time flicker calculation can be performed. This invention does not require statistical analysis of the cumulative probability distribution of data within the detection time, nor does it require piecewise linear interpolation. It only needs to use the range of the instantaneous flicker value and the instantaneous flicker value to calculate the quantile value corresponding to each quantile probability. The calculation method is simple and does not require storing a large amount of data; at the same time, it can adaptively adjust the estimated value according to the dynamic changes of the instantaneous flicker value, ensuring estimation accuracy.

Description

Technical Field

[0001] This invention relates to a method and apparatus for real-time calculation of short-time flicker, belonging to the field of power system measurement technology. Background Technology

[0002] The visual perception of flicker caused by voltage fluctuations resulting in unstable light intensity (light flicker) is called flicker. In other words, flicker reflects the impact of light flicker caused by voltage fluctuations on human visual perception. Short-term flicker (Pst) is a statistical measure of the intensity of flicker over a short period of time (several minutes). Typically, Pst = 1 is the limit for the visual stimulation caused by flicker. As an important technical indicator in power quality monitoring, flicker reflects real-time voltage fluctuations. Research on flicker is increasing, but it generally requires caching instantaneous visual perception data for a period of time before statistical calculation. For example, if the short-term flicker value needs to be calculated every 10 minutes, at a sampling rate of 400 Hz, 240,000 data points need to be cached, which not only requires a large amount of storage space but also has a large computational load, affecting computational efficiency.

[0003] To address this, a method for calculating short-term flicker that reduces storage requirements has been proposed. Chinese patent application CN102928703A discloses a method and system for measuring short-term flicker. This method classifies instantaneous flicker values ​​and records the number of flicker values ​​at each level. This allows the data storage space to only store the number of flicker values ​​at each level, reducing the amount of data that needs to be stored during calculation. This scheme reduces the number of data points required for calculating short-term flicker from 240,000 to the number of levels (at least 64 levels according to IEC standards). However, this method requires pre-setting a classification threshold. All instantaneous flicker values ​​within a preset time period are classified according to the preset threshold, requiring the storage of the number of values ​​at each level. This involves real-time calculation of the instantaneous flicker value distribution histogram. However, flicker values ​​in real-world environments are often unpredictable. If the preset threshold does not conform to the distribution pattern of instantaneous flicker values, the calculated distribution histogram will be distorted, making it impossible to obtain accurate short-term flicker values. Summary of the Invention

[0004] The purpose of this invention is to provide a method and apparatus for real-time calculation of short-time flicker, so as to solve the problem of poor calculation accuracy caused by the need to pre-set hierarchical thresholds in the current real-time calculation of short-time flicker.

[0005] To solve the above-mentioned technical problems, the present invention provides a method for real-time calculation of short-time flicker, the method comprising:

[0006] 1) Obtain the instantaneous flicker value within the detection time;

[0007] 2) The cumulative probability required for short-time flicker calculation is converted into a probability value that is compatible with the adaptive method, called the quantile probability. Each quantile probability is equal to 1 minus the corresponding cumulative probability.

[0008] 3) Calculate the quantile value corresponding to each quantile probability using the obtained quantile probabilities, the range of instantaneous flicker values, and the instantaneous flicker values;

[0009] 4) Multiply each quantile probability by its corresponding quantile value and sum them to obtain the short-time flicker value within the detection time.

[0010] Furthermore, the formula used to calculate the quantile values ​​corresponding to each quantile probability in step 3) is as follows:

[0011]

[0012] in, Let be the quantile value corresponding to the probability of the m-th quantile. Let m be the probability of the m-th quantile. This refers to the range of instantaneous flicker values ​​corresponding to the nth sample value. This is the instantaneous flicker value corresponding to the nth sample value.

[0013] Furthermore, the range of instantaneous flicker values ​​is obtained through real-time updates, and the calculation formula used is as follows:

[0014]

[0015]

[0016]

[0017] in These are the upper and lower bounds of the range to which the instantaneous flicker value belongs. The range of instantaneous flicker values ​​corresponding to the nth sample value. This represents the instantaneous flicker value corresponding to the nth sample value. and All coefficients are greater than 0 and less than 1.

[0018] Furthermore, the calculation formula used for the short-time flicker value within the detection time in step 4) is as follows:

[0019]

[0020] in , , , and The quantile probabilities corresponding to cumulative probabilities of 0.1%, 1%, 3%, 10%, and 50% within the respective detection time periods. , , , and These are the quantile probabilities during the detection time. , , , and The corresponding quantile value.

[0021] Furthermore, the instantaneous flicker value within the detection time is obtained by squaring, bandpass filtering, visual sensitivity weighted filtering, and smoothing weighted filtering of the voltage sampling signal within the detection time.

[0022] Furthermore, the bandpass filter used in the bandpass filtering consists of a high-pass filter with a cutoff frequency of 0.05 Hz and a low-pass filter with a cutoff frequency of 35 Hz.

[0023] Furthermore, the transfer function of the perception-weighted filter used in the perception-weighted filtering is:

[0024]

[0025] in, , , , , , .

[0026] Furthermore, the aforementioned visual perception weighted filter is implemented by converting it into a digital filter using the bilinear function in MATLAB.

[0027] Furthermore, the transfer function used in the smoothing weighted filtering is:

[0028]

[0029] in 300ms was selected.

[0030] The present invention also provides a short-time flicker real-time computing device, including a processor, the processor being used to execute computer program instructions to implement the short-time flicker real-time computing method of the present invention.

[0031] The beneficial effects of this invention are as follows: As an improved invention, this invention only needs to convert the cumulative probability required for short-time flicker calculation into a probability value adapted to the adaptive method, use this probability value as the quantile probability, and then use the obtained quantile probability, the range of the instantaneous flicker value, and the instantaneous flicker value to calculate the quantile value corresponding to each quantile probability; finally, based on the obtained quantile probability and the corresponding quantile value, real-time short-time flicker calculation can be performed. This invention does not require statistical analysis of the cumulative probability distribution of data within the detection time, nor does it require piecewise linear interpolation. It only needs to use the range of the instantaneous flicker value and the instantaneous flicker value to calculate the quantile value corresponding to each quantile probability. The calculation method is simple and does not require storing a large amount of data. At the same time, it can adaptively adjust the estimated value according to the dynamic changes of the instantaneous flicker value, ensuring the estimation accuracy. Attached Figure Description

[0032] Figure 1 This is a flowchart of the short-time flicker real-time calculation method of the present invention;

[0033] Figure 2 This is a schematic diagram of the quantile estimation curve obtained in an embodiment of the present invention. Detailed Implementation

[0034] The specific embodiments of the present invention will be further described below with reference to the accompanying drawings.

[0035] This invention only requires converting the cumulative probability needed for short-time flicker calculation into a probability value adapted to the adaptive method. This probability value is then used as the quantile probability. The obtained quantile probabilities, the range of the instantaneous flicker value, and the instantaneous flicker value are then used to calculate the quantile value corresponding to each quantile probability. Finally, based on the obtained quantile probabilities and corresponding quantile values, real-time short-time flicker calculation can be performed. This invention does not require statistical analysis of the cumulative probability distribution of data within the detection time, nor does it require piecewise linear interpolation. It only needs to estimate the five probability quantiles and flicker range required for short-time flicker calculation in real time based on the input instantaneous flicker value, dynamically update and store these six data points, and retrieve the five quantile values ​​at the preset time to calculate the short-time flicker value. This method further reduces the number of cached data levels from six and considers the range changes of the instantaneous flicker value during estimation. It can adaptively adjust the estimated value according to the dynamic changes of the instantaneous flicker value, ensuring estimation accuracy.

[0036] Implementation Examples of Short-Time Flicker Real-Time Calculation Methods

[0037] This invention first converts the cumulative probability required for short-time flicker calculation into a probability value adapted to the adaptive method, called the quantile probability. Then, using the obtained quantile probabilities, the range of the instantaneous flicker value, and the instantaneous flicker value, the quantile value corresponding to each quantile probability is calculated. Finally, each quantile probability is multiplied by its corresponding quantile value and summed to obtain the short-time flicker value within the detection time. The calculation process of this method is as follows: Figure 1 As shown below, a detailed explanation will follow.

[0038] 1. Acquire voltage sampling signals containing flicker and preprocess them.

[0039] In this embodiment, the flicker signal is a square wave signal with a fluctuation of 0.722% and a frequency of 0.916, with a sampling rate of fs=400h. The preprocessing of the acquired voltage sampling signal containing flicker in this invention includes squaring, bandpass filtering, and weighted filtering. The acquired voltage sampling signal containing flicker is as follows:

[0040]

[0041] The squared voltage signal obtained by squaring the above voltage sampled signal is: ,

[0042] in:

[0043]

[0044] in w is the modulation index of the flicker signal. f The frequency of the flicker signal is defined as U, the rated voltage of the fundamental wave is w, the fundamental frequency is n, and the sampling point is n.

[0045] The voltage square signal is subjected to a 0.05Hz-35Hz bandpass filter to remove out-of-band signals, yielding f(n). In this embodiment, the bandpass filter consists of a high-pass filter with a cutoff frequency of 0.05Hz and a low-pass filter with a cutoff frequency of 35Hz. The resulting f(n) is:

[0046]

[0047]

[0048] The obtained passband voltage squared signal f(n) is subjected to visual sensitivity weighted filtering to convert all frequency fluctuations within the band into fluctuations at a frequency of 8.8 Hz, thus obtaining the signal. The transfer function of the visual perception weighted filter used in this embodiment is:

[0049]

[0050] in , , , , , It is implemented by converting the filter into a digital filter using the bilinear function in MATLAB.

[0051] The signal is obtained by using the above-mentioned visual perception weighted filter. for:

[0052]

[0053]

[0054] 2. Calculate the instantaneous flicker value based on the visual perception weighted and filtered signal. And then normalize it.

[0055] For the obtained signal The instantaneous flicker value is obtained by performing squared and smoothed weighted filtering. The smoothing weighted filter used in this embodiment is:

[0056]

[0057] in 300ms was selected. The instantaneous flicker value was obtained using the aforementioned smoothing weighted filter. for:

[0058]

[0059] Will Normalization yields S(n).

[0060] 3. Calculate the instantaneous flicker value based on the obtained normalized instantaneous flicker degree value. Then, the short-time flicker value is calculated according to the IEC standard formula.

[0061] The IEC marking formula is as follows:

[0062]

[0063] , , , and These are the flicker visual sensitivity values ​​corresponding to probabilities exceeding 0.1%, 1%, 3%, 10%, and 50% respectively within the detection time (10 minutes in this embodiment). This invention does not require cumulative probability distribution statistics on the data within the detection time (e.g., 10 minutes), nor does it require piecewise linear interpolation. Instead, it uses quantile adaptive estimation to update the values ​​in real time. The specific steps are as follows:

[0064] 1) Convert the cumulative probability values ​​required in the IEC standard calculation formula into probability values ​​that are compatible with the adaptive estimation method. The cumulative probability value Pcdf given by the IEC standard m =[0.1% 1% 3% 10% 50%] is converted into the probability values ​​required for subsequent estimation. Based on the cumulative probability value in the calculation formula of the IEC recommended calculation scheme, this invention obtains 5... The initial estimates are calculated using the first 2400 data points within the detection time, with values ​​of 0.999, 0.99, 0.97, 0.9, and 0.5 respectively. These initial estimates are obtained by performing probability statistics on the 2400 instantaneous flicker values ​​to obtain five quantile values, which accelerates the estimation convergence speed.

[0065] 2) Based on the input instantaneous flicker value S(n), estimate and update in real time. , These are the upper and lower bounds of the estimated value, respectively. This is the quantile probability.

[0066]

[0067]

[0068] in , In this embodiment , .

[0069] 3) Real-time probability estimation Corresponding quantile values The specific calculation formula used is as follows:

[0070]

[0071] The current input S(n) is compared to When the value is large, then increase the increment. , The value is added to the increment value; the input S(n) is compared to... Hours, increase , Value minus increment value, The range is (0, 1), selected based on debugging results. Other feasible values ​​can be chosen, but the value should comprehensively consider the estimated convergence speed and the fluctuation range of the estimated value. Here, we select... The estimated quantile curve is as follows: Figure 2 As shown.

[0072] 4) The 5 estimates obtained at 10 minutes Substitute The calculation formula yields the short-time flicker. The present invention adopts The calculation formula is:

[0073]

[0074] in This represents the quantile probability corresponding to a cumulative probability of 0.1% within the detection period. This represents the quantile probability corresponding to a cumulative probability of 1% within the detection time. This represents the quantile probability corresponding to a cumulative probability of 3% within the detection period. This represents the quantile probability corresponding to a cumulative probability of 10% within the detection period. This represents the quantile probability corresponding to a cumulative probability of 50% within the detection period. , , , and These are the quantile probabilities during the detection time. , , , and The corresponding quantile value.

[0075] According to the test standard modulated square wave flicker signal in Table 1, the calculation results of the present invention are compared with the calculation results of the traditional method, and the results are shown in Table 1.

[0076] Table 1

[0077]

[0078] As can be seen from Table 1, the calculation results of the two methods are basically the same. However, this invention only needs to use the range of the instantaneous flicker value and the instantaneous flicker value to calculate the quantile value corresponding to each quantile probability. During the calculation process, only 6 data points need to be cached, and it can dynamically track the changes in the instantaneous flicker probability distribution in real time, adaptively adjusting the estimated value to meet different P values. st The value flicker test scenario is simple to calculate, has high computational efficiency, and does not require storing a large amount of data.

[0079] Example of a short-time flicker real-time computing device

[0080] The short-time flicker real-time computing device of the present invention can be used to execute computer program instructions to implement the short-time flicker real-time computing method in the above-described method embodiments. The specific implementation process of this method has been described in detail in the embodiments of the short-time flicker real-time computing method, and will not be repeated here.

Claims

1. A method for real-time calculation of short-time flicker, characterized in that, The method includes: 1) Obtain the instantaneous flicker value within the detection time; 2) The cumulative probability required for short-time flicker calculation is converted into a probability value that is compatible with the adaptive method, called the quantile probability. Each quantile probability is equal to 1 minus the corresponding cumulative probability. 3) Calculate the quantile value corresponding to each quantile probability using the obtained quantile probabilities, the range of the instantaneous flicker value, and the instantaneous flicker value; the calculation formula is: in, Let be the quantile value corresponding to the probability of the m-th quantile at the n-th sampling point. Let m be the quantile value corresponding to the probability of the m-th quantile at the (n-1)-th sampling point. Let m be the probability of the m-th quantile. This refers to the range of instantaneous flicker values ​​corresponding to the nth sample value. This represents the instantaneous flicker value corresponding to the nth sample value. is a coefficient, and its value range is (0, 1); 4) Multiply each quantile probability by its corresponding quantile value and sum them to obtain the short-time flicker value within the detection time.

2. The short-time flicker real-time calculation method according to claim 1, characterized in that, In step 3) The value of is determined by considering the estimated convergence rate and the range of fluctuation of the estimated value.

3. The short-time flicker real-time calculation method according to claim 1 or 2, characterized in that, The range of instantaneous flicker values ​​is obtained through real-time updates, and the calculation formula used is: in These are the upper and lower bounds of the range to which the instantaneous flicker value belongs. The range of instantaneous flicker values ​​corresponding to the nth sample value. This represents the instantaneous flicker value corresponding to the nth sample value. and All coefficients are greater than 0 and less than 1.

4. The short-time flicker real-time calculation method according to claim 1 or 2, characterized in that, The calculation formula used for the short-time flicker value within the detection time in step 4) is: in , , , and These represent the quantile probabilities corresponding to cumulative probabilities of 0.1%, 1%, 3%, 10%, and 50% during the detection period. , , , and These are the quantile probabilities during the detection time. , , , and The corresponding quantile value.

5. The short-time flicker real-time calculation method according to claim 1 or 2, characterized in that, The instantaneous flicker value within the detection time is obtained by squaring, bandpass filtering, visual sensitivity weighted filtering, and smoothing weighted filtering of the voltage sampling signal within the detection time.

6. The short-time flicker real-time calculation method according to claim 5, characterized in that, The bandpass filter used in the above-mentioned bandpass filtering consists of a high-pass filter with a cutoff frequency of 0.05 Hz and a low-pass filter with a cutoff frequency of 35 Hz.

7. The short-time flicker real-time calculation method according to claim 5, characterized in that, The transfer function of the visual perception weighted filter used in visual perception weighted filtering is: in, , , , , , .

8. The short-time flicker real-time calculation method according to claim 7, characterized in that, The aforementioned visual perception weighted filter is implemented by converting it into a digital filter using the bilinear function in MATLAB.

9. The short-time flicker real-time calculation method according to claim 5, characterized in that, The transfer function used in the smoothing weighted filtering is: in 300ms was selected.

10. A short-time flicker real-time computing device, comprising a processor, characterized in that, The processor is used to execute computer program instructions to implement the short-time flash real-time calculation method as described below: 1) Obtain the instantaneous flicker value within the detection time; 2) The cumulative probability required for short-time flicker calculation is converted into a probability value that is compatible with the adaptive method, called the quantile probability. Each quantile probability is equal to 1 minus the corresponding cumulative probability. 3) Calculate the quantile value corresponding to each quantile probability using the obtained quantile probabilities, the range of the instantaneous flicker value, and the instantaneous flicker value; the calculation formula is: in, Let be the quantile value corresponding to the probability of the m-th quantile at the n-th sampling point. Let m be the quantile value corresponding to the probability of the m-th quantile at the (n-1)-th sampling point. Let m be the probability of the m-th quantile. This refers to the range of instantaneous flicker values ​​corresponding to the nth sample value. This represents the instantaneous flicker value corresponding to the nth sample value. is a coefficient, and its value range is (0, 1); 4) Multiply each quantile probability by its corresponding quantile value and sum them to obtain the short-time flicker value within the detection time.

11. The short-time flicker real-time computing device according to claim 10, characterized in that, In step 3) The value of is determined by considering the estimated convergence rate and the range of fluctuation of the estimated value.

12. The short-time flicker real-time computing device according to claim 10 or 11, characterized in that, The range of instantaneous flicker values ​​is obtained through real-time updates, and the calculation formula used is: in These are the upper and lower bounds of the range to which the instantaneous flicker value belongs. The range of instantaneous flicker values ​​corresponding to the nth sample value. This represents the instantaneous flicker value corresponding to the nth sample value. and All coefficients are greater than 0 and less than 1.

13. The short-time flicker real-time computing device according to claim 10 or 11, characterized in that, The calculation formula used for the short-time flicker value within the detection time in step 4) is: in , , , and The quantile probabilities corresponding to cumulative probabilities of 0.1%, 1%, 3%, 10%, and 50% within the respective detection time periods. , , , and These are the quantile probabilities during the detection time. , , , and The corresponding quantile value.

14. The short-time flicker real-time computing device according to claim 10 or 11, characterized in that, The instantaneous flicker value within the detection time is obtained by squaring, bandpass filtering, visual sensitivity weighted filtering, and smoothing weighted filtering of the voltage sampling signal within the detection time.

15. The short-time flicker real-time computing device according to claim 14, characterized in that, The bandpass filter used in the above-mentioned bandpass filtering consists of a high-pass filter with a cutoff frequency of 0.05 Hz and a low-pass filter with a cutoff frequency of 35 Hz.

16. The short-time flicker real-time computing device according to claim 14, characterized in that, The transfer function of the visual perception weighted filter used in visual perception weighted filtering is: in, , , , , , .

17. The short-time flicker real-time computing device according to claim 16, characterized in that, The aforementioned visual perception weighted filter is implemented by converting it into a digital filter using the bilinear function in MATLAB.

18. The short-time flicker real-time computing device according to claim 14, characterized in that, The transfer function used in the smoothing weighted filtering is: in 300ms was selected.

Images