A smart detection system for power quality in residential areas

Abstract

The present invention relates to the field of data analysis technology, and in particular to an intelligent detection system for electric energy quality in residential areas, the system comprising a processor and a memory, the processor executing a computer program in the memory to implement the following steps: obtaining quality index scores corresponding to each sampling moment of N kinds of electric energy quality indicators in the residential area within a preset time period; detecting whether there are scores to be revised in the quality index scores, if scores to be revised are detected, then revising each score to be revised to obtain a corresponding revised score, and obtaining a final electric energy quality score at each sampling moment according to the revised score at each sampling moment; and evaluating the electric energy quality of the residential area within the preset time period according to the final electric energy quality score at each sampling moment within the preset time period. The present invention obtains the final electric energy quality score by revising the score, evaluates the electric energy quality of the residential area, and improves the accuracy and reliability of the electric energy quality evaluation of the residential area.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and in particular to an intelligent detection system for power quality in residential areas. Background Art

[0002] With the rapid advancement of urbanization and the significant improvement of residents' living standards, the stability and reliability of the power quality of residential areas, as the basic units in the power network, play a vital role in ensuring the daily life of residents and promoting the harmonious development of the community. However, in the current power supply system, residential areas face many power quality challenges, such as voltage fluctuations, harmonic pollution, and three-phase imbalance. These problems not only affect the normal operation of residents' electrical equipment, but also may pose a potential threat to the safety of the power grid. Therefore, it is necessary to adopt an effective method to comprehensively evaluate the power quality of residential areas, and feedback the current power grid stability and reliability through the power quality score of residential areas.

[0003] Traditional evaluation methods generally use subjective or objective weighting methods to set weights for power quality indicators or scores of each indicator, and use the weighted sum of scores of each indicator as the overall evaluation of power quality of the power grid. This weighting method directly adds up the scores of each power quality after obtaining the scores, regardless of whether the data is abnormal or not, to get the overall evaluation, which will cause the overall evaluation of power quality of the power grid to be biased due to abnormal data of some power quality indicators, resulting in errors in the evaluation of power quality.

[0004] Therefore, how to reduce the impact of abnormal data in the power quality assessment process and improve the accuracy and reliability of power quality assessment in residential areas has become an urgent problem to be solved. Summary of the invention

[0005] In view of this, an embodiment of the present invention provides a smart power quality detection system for residential areas to solve the problem of how to reduce the impact of abnormal data in the power quality assessment process and improve the accuracy and reliability of power quality assessment in residential areas.

[0006] An embodiment of the present invention provides a residential area power quality intelligent detection system, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the following method is implemented:

[0007] Obtain multi-dimensional electric energy data of the residential area at any sampling time, obtain parameter values ​​corresponding to N preset power quality indicators at the sampling time, obtain quality indicator scores corresponding to the N preset power quality indicators at the sampling time according to all parameter values, and obtain quality indicator scores corresponding to each sampling time of the N preset power quality indicators within a preset time period;

[0008] For any sampling moment within the preset time period, a time window of preset length is established with the sampling moment as the center, and according to the quality indicator score corresponding to each sampling moment within the time window, it is detected whether there is a score to be corrected in the sampling moment;

[0009] If it is detected that there is at least one score to be corrected at the sampling moment, then according to all quality indicator scores at each sampling moment in the time window, each score to be corrected at the sampling moment is corrected to obtain a corresponding corrected score, and all non-corrected scores and all corrected scores at the sampling moment are accumulated to obtain a final power quality score at the sampling moment;

[0010] The final power quality score at each sampling moment in the preset time period is obtained respectively, and the power quality of the residential area in the preset time period is evaluated according to the final power quality score at each sampling moment in the preset time period.

[0011] Preferably, after detecting whether there is a score to be corrected at the sampling moment, the method further includes:

[0012] If no score to be corrected is detected at the sampling moment, the sum of all quality indicator scores at the sampling moment is accumulated as the final power quality score at the sampling moment.

[0013] Preferably, obtaining the quality indicator scores corresponding to the N types of power quality indicators at the sampling time respectively according to all parameter values ​​includes:

[0014] For any power quality indicator among the N power quality indicators, setting a first parameter threshold and a second parameter threshold of the power quality indicator, wherein the first parameter threshold is less than the second parameter threshold;

[0015] If the parameter value corresponding to the power quality indicator at the sampling time is less than or equal to the first parameter threshold, obtain the ratio of the parameter value of a first preset multiple to the first parameter threshold as a first ratio, and calculate the difference between a constant 1 and the first ratio as the quality indicator score corresponding to the power quality indicator at the sampling time;

[0016] If the parameter value corresponding to the power quality indicator at the sampling time is greater than the first parameter threshold value and less than or equal to the second parameter threshold value, obtain the difference between the first parameter threshold value and the second parameter threshold value as the first difference value, calculate the ratio of the parameter value of the second preset multiple to the first difference value as the second ratio, calculate the ratio of the second parameter threshold value of the second preset multiple to the first difference value as the third ratio, and obtain the difference between the second ratio and the third ratio as the quality indicator score corresponding to the power quality indicator at the sampling time;

[0017] If the parameter value corresponding to the power quality indicator at the sampling time is greater than the second parameter threshold, the quality indicator score corresponding to the power quality indicator at the sampling time is set to a preset value.

[0018] Preferably, the step of detecting whether there is a score to be corrected at each sampling moment according to the quality indicator score corresponding to each sampling moment in the time window comprises:

[0019] For any quality indicator score at the sampling moment, all quality indicator scores belonging to the same power quality indicator as the quality indicator score are obtained in the time window, and form a quality indicator score sequence with the quality indicator score;

[0020] Fitting the quality indicator score sequence to obtain a fitting function, using the fitting function to predict the quality indicator score to obtain a predicted value, obtaining an absolute value of a difference between the quality indicator score and the predicted value, normalizing the absolute value of the difference to obtain an abnormal coefficient of the quality indicator score;

[0021] Obtaining an abnormal coefficient of each quality indicator score in the quality indicator score sequence, and when the abnormal coefficient of the quality indicator score is greater than or equal to a preset abnormal coefficient threshold, and the abnormal coefficient of each quality indicator score in the quality indicator score sequence except the quality indicator score is less than the preset abnormal coefficient threshold, confirming that the quality indicator score is a score to be revised;

[0022] When the abnormal coefficient of the quality indicator score is greater than or equal to the preset abnormal coefficient threshold, and in the quality indicator score sequence, there is at least one quality indicator score other than the quality indicator score whose abnormal coefficient is greater than or equal to the preset abnormal coefficient threshold, confirming that the quality indicator score is a normal score;

[0023] When the abnormal coefficient of the quality indicator score is less than a preset abnormal coefficient threshold, the quality indicator score is confirmed to be a normal score.

[0024] Preferably, the step of correcting each score to be corrected at the sampling moment according to all quality indicator scores at each sampling moment within the time window to obtain a corresponding corrected score includes:

[0025] For any score to be corrected at the sampling moment, obtaining a first related indicator and a second related indicator related to the power quality indicator corresponding to the score to be corrected from the N types of power quality indicators;

[0026] According to the quality indicator score corresponding to the first relevant indicator at the sampling time, obtaining a first relevant score of the score to be corrected, and obtaining a correlation coefficient between the score to be corrected and the first relevant score, recorded as a first correlation coefficient;

[0027] According to the quality indicator score corresponding to the second related indicator at the sampling time, obtaining a second related score of the score to be corrected, and obtaining a correlation coefficient between the score to be corrected and the second related score, recorded as a second correlation coefficient;

[0028] Respectively obtaining the instantaneous change rates of the first correlation coefficient and the second correlation coefficient within the time window, obtaining the absolute value of the difference between the instantaneous change rate of the first correlation coefficient and the instantaneous change rate of the second correlation coefficient, and performing normalization processing on the absolute value of the difference to obtain an error coefficient;

[0029] The first correlation coefficient is used as the first weight of the first correlation score, the error coefficient is used as the second weight of the score to be revised, and the first correlation score and the score to be revised are weightedly summed to obtain a revised score of the score to be revised.

[0030] Preferably, obtaining the first relevant score of the score to be corrected according to the quality indicator score corresponding to the first relevant indicator at the sampling time includes:

[0031] When the quality indicator score corresponding to the first relevant indicator at the sampling time is a normal score, taking the quality indicator score corresponding to the first relevant indicator at the sampling time as the first relevant score of the score to be corrected;

[0032] When the quality indicator score corresponding to the first related indicator at the sampling moment is a score to be revised, detecting whether there is a normal score at the sampling moment, and if it is detected that there is at least one normal score at the sampling moment, taking any normal score at the sampling moment as an available score at the sampling moment; if it is detected that there is no normal score at the sampling moment, obtaining a preset target power quality indicator, and obtaining an average of the quality indicator scores corresponding to the target power quality indicator in the first preset sub-period and the second preset sub-period as the available score at the sampling moment;

[0033] The expected score of the first relevant indicator at the sampling moment is obtained according to the available score at the sampling moment, and the expected score of the first relevant indicator at the sampling moment is used as the first relevant score of the score to be revised.

[0034] Preferably, the obtaining of the correlation coefficient between the score to be corrected and the first related score includes:

[0035] In a first preset sub-period, all quality indicator scores that belong to the same power quality indicator as the score to be corrected are obtained to form a first scoring sequence. In a first preset sub-period, all quality indicator scores that belong to the same power quality indicator as the first related score are obtained to form a second scoring sequence. The absolute value of the correlation coefficient between the first scoring sequence and the second scoring sequence is obtained and recorded as a first coefficient.

[0036] In the second preset sub-period, all quality indicator scores that belong to the same power quality indicator as the score to be corrected are obtained to form a third scoring sequence. In the second preset sub-period, all quality indicator scores that belong to the same power quality indicator as the first related score are obtained to form a fourth scoring sequence. The absolute value of the correlation coefficient between the third scoring sequence and the fourth scoring sequence is obtained and recorded as the second coefficient.

[0037] An average value of the first coefficient and the second coefficient is obtained as a correlation coefficient between the score to be corrected and the first relevant score.

[0038] Preferably, obtaining the expected score of the first related indicator at the sampling moment according to the available score at the sampling moment includes:

[0039] Record the quality index score corresponding to the first related indicator at the sampling moment as the abnormal score; if the first related indicator or the second related indicator related to the power quality indicator corresponding to the available score is the power quality indicator corresponding to the abnormal score, obtain the correlation coefficient between the available score and the abnormal score, calculate the product of the correlation coefficient between the available score and the abnormal score and the available score as the first product, obtain the mean of the quality index scores of the power quality indicator corresponding to the available score in the first preset sub-period and the second preset sub-period as the first mean, calculate the product of the correlation coefficient between the available score and the abnormal score and the first mean as the second product, obtain the mean of the quality index scores of the power quality indicator corresponding to the abnormal score in the first preset sub-period and the second preset sub-period as the second mean, calculate the addition result of the first product and the second mean, and obtain the difference between the addition result and the second product as the expected score of the first related indicator at the sampling moment;

[0040] If the first correlation indicator and the second correlation indicator related to the power quality indicator corresponding to the available score are not the power quality indicators corresponding to the abnormal score, the first correlation indicator related to the power quality indicator corresponding to the available score is marked as the first indicator, and the second correlation indicator related to the power quality indicator corresponding to the available score is marked as the second indicator. According to the correlation coefficients between the quality indicator score corresponding to the first indicator and the quality indicator score corresponding to the second indicator and the available score, respectively, the expected scores of the first indicator and the second indicator at the sampling time are obtained, which are recorded as candidate expected scores. According to the correlations between the power quality indicator corresponding to the abnormal score and the first indicator and the second indicator, respectively, a target expected score is obtained from the candidate expected scores. The target expected score is used as a new available score. According to the method for obtaining the expected score of the first correlation indicator at the sampling time according to the available score at the sampling time, the expected score of the first correlation indicator at the sampling time is obtained using the new available score.

[0041] Preferably, obtaining a target expected score from the candidate expected scores according to the correlation between the power quality indicator corresponding to the abnormal score and the first indicator and the second indicator respectively includes:

[0042] If the first related indicator or the second related indicator related to the first indicator is the power quality indicator corresponding to the abnormal score, the expected score of the first indicator at the sampling time is used as the target expected score;

[0043] If neither the first correlation indicator and the second correlation indicator related to the first indicator are the power quality indicators corresponding to the abnormal score, detect whether any of the first correlation indicator and the second correlation indicator related to the second indicator is the power quality indicator corresponding to the abnormal score; if it is detected that the first correlation indicator or the second correlation indicator related to the second indicator is the power quality indicator corresponding to the abnormal score, take the expected score of the second indicator at the sampling time as the target expected score.

[0044] Preferably, the evaluating the power quality of the residential area within the preset period of time according to the final power quality score at each sampling moment within the preset period of time includes:

[0045] Setting a power quality threshold, if the final power quality score at any sampling moment within the preset time period is greater than or equal to the power quality threshold, then determining that the power quality at the sampling moment is qualified;

[0046] The qualified rate of the power quality in the preset time period is obtained, and when the qualified rate of the power quality in the preset time period is greater than a preset qualified rate threshold, it is confirmed that the power quality of the residential area in the preset time period is qualified.

[0047] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0048] The present invention obtains multi-dimensional electric energy data of a residential area at any sampling moment, obtains parameter values ​​corresponding to N preset electric energy quality indicators at the sampling moments, obtains quality indicator scores corresponding to the N preset electric energy quality indicators at the sampling moments according to all parameter values, and obtains quality indicator scores corresponding to each sampling moment of the N preset electric energy quality indicators at a preset time period; for any sampling moment in the preset time period, a time window of preset length is established with the sampling moment as the center, and according to the quality indicator score corresponding to each sampling moment in the time window, it is detected whether there is a score to be corrected in the sampling moment; if it is detected that there is at least one score to be corrected in the sampling moment, then according to all quality indicator scores of each sampling moment in the time window, each score to be corrected at the sampling moment is corrected to obtain a corresponding corrected score, and all non-corrected scores and all corrected scores at the sampling moment are accumulated to obtain a final electric energy quality score at the sampling moment; the final electric energy quality score of each sampling moment in the preset time period is respectively obtained, and according to the final electric energy quality score of each sampling moment in the preset time period, the electric energy quality of the residential area in the preset time period is evaluated. The present invention detects the scores of various quality indicators of residential substations, obtains the scores to be corrected due to instantaneous fluctuations and corrects them, and evaluates the power quality of the residential substations through the final power quality scores obtained by correcting the scores, thereby reducing the problem of inaccurate final power quality scores caused by abnormal data generated by instantaneous fluctuations, thereby improving the accuracy and reliability of power quality evaluation of residential substations. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0050] Figure 1 It is a flow chart of a method for intelligent detection of power quality in residential areas provided in Embodiment 1 of the present invention. DETAILED DESCRIPTION

[0051] Embodiments of the present disclosure are described in detail below, and examples of the embodiments are shown in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present disclosure, but should not be understood as limiting the present disclosure.

[0052] It should be noted that the terms "first", "second", etc. in the specification of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure.

[0053] In order to illustrate the technical solution of the present invention, specific embodiments are provided below for illustration.

[0054] The embodiment of the present invention provides a residential area power quality intelligent detection system, including a processor and a memory, wherein the processor executes a computer program of the memory to implement a residential area power quality intelligent detection method, such as Figure 1 As shown, the intelligent detection method of power quality in residential areas includes the following steps:

[0055] Step S101, obtain multi-dimensional electric energy data of the residential area at any sampling time, obtain the parameter values ​​corresponding to the preset N kinds of electric energy quality indicators at the said sampling time, obtain the quality indicator scores corresponding to the said N kinds of electric energy quality indicators at the said sampling time according to all the parameter values, and obtain the quality indicator scores corresponding to the said N kinds of electric energy quality indicators at each sampling time within the preset time period.

[0056] A monitoring point is set up in the residential area to monitor the multi-dimensional power data of the residential area every 0.1 second, and the multi-dimensional power data within two minutes is obtained. There is no restriction here and it can be set according to the specific implementation scenario. The multi-dimensional power data includes actual load, actual voltage, harmonic current, maximum three-phase voltage, minimum three-phase voltage, average three-phase voltage, standard voltage, fundamental current and rated load.

[0057] According to the multi-dimensional electric energy data of the residential substation at any sampling time, the parameter value corresponding to each electric energy quality indicator of the residential substation at the sampling time is calculated. In this embodiment, four electric energy quality indicators are used, including but not limited to overload rate, voltage deviation, harmonic impact index and three-phase voltage imbalance rate. Each electric energy quality indicator belongs to the prior art and will not be described in detail here.

[0058] After obtaining the parameter value corresponding to each power quality indicator in the residential area at the sampling time, the quality indicator score corresponding to each power quality indicator at the sampling time can be obtained according to the parameter value corresponding to each power quality indicator at the sampling time.

[0059] Among them, according to the corresponding parameter value of each power quality indicator at the sampling time, the method for obtaining the quality indicator score corresponding to each power quality indicator at the sampling time is as follows:

[0060] (1) For any power quality indicator among the N power quality indicators, a first parameter threshold and a second parameter threshold of the power quality indicator are set, wherein the first parameter threshold is less than the second parameter threshold.

[0061] Set the maximum parameter threshold allowed for each power quality indicator, where the maximum parameter threshold allowed for the overload rate is set to 100%, the maximum parameter threshold allowed for the voltage deviation is set to 10%, the maximum parameter threshold allowed for the harmonic impact indicator is set to 5%, and the maximum parameter threshold allowed for the three-phase voltage unbalance rate is set to 5%. There is no restriction here and it can be set according to the specific implementation scenario.

[0062] For any power quality indicator among the N power quality indicators, the first parameter threshold of the power quality indicator is set to one tenth of the maximum parameter threshold, and the second parameter threshold of the power quality indicator is set to the maximum parameter threshold allowed for the power quality indicator to be qualified. There is no restriction here and it can be set according to the specific implementation scenario.

[0063] (2) If the parameter value corresponding to the power quality indicator at the sampling time is less than or equal to the first parameter threshold, obtain the ratio of the parameter value of a first preset multiple to the first parameter threshold as the first ratio, and calculate the difference between the constant 1 and the first ratio as the quality indicator score corresponding to the power quality indicator at the sampling time.

[0064] In one implementation, taking the nth power quality indicator as an example, a parameter value corresponding to the nth power quality indicator at the i-th sampling time is obtained. If the parameter value corresponding to the nth power quality indicator at the i-th sampling time is less than or equal to the first parameter threshold of the nth power quality indicator, the formula for calculating the quality indicator score corresponding to the nth power quality indicator at the i-th sampling time is:

[0065]

[0066] in, Score the quality index corresponding to the nth power quality index at the i-th sampling moment; is the parameter value corresponding to the nth power quality indicator at the i-th sampling moment; i is the sequence number of the sampling moment in the preset time period; is the first parameter threshold of the nth power quality indicator; n is the power quality indicator sequence number among the N power quality indicators; 0.2 represents the first preset multiple, and 1 represents a constant.

[0067] It should be noted that if the parameter value corresponding to the nth power quality indicator at the i-th sampling time is , determine that the quality indicator score corresponding to the nth power quality indicator at the i-th sampling moment is excellent, and the range of the quality indicator score is 0.8-1. There is no restriction here, and it can be set according to the specific implementation scenario. The larger the parameter value corresponding to the nth power quality indicator at the i-th sampling moment, the smaller the quality indicator score corresponding to the nth power quality indicator at the i-th sampling moment.

[0068] (3) If the parameter value corresponding to the power quality indicator at the sampling time is greater than the first parameter threshold and less than or equal to the second parameter threshold, obtain the difference between the first parameter threshold and the second parameter threshold as the first difference, calculate the ratio of the parameter value of the second preset multiple to the first difference as the second ratio, calculate the ratio of the second parameter threshold of the second preset multiple to the first difference as the third ratio, and obtain the difference between the second ratio and the third ratio as the quality indicator score corresponding to the power quality indicator at the sampling time.

[0069] In one implementation, taking the nth power quality indicator as an example, a parameter value corresponding to the nth power quality indicator at the i-th sampling time is obtained. If the parameter value corresponding to the nth power quality indicator at the i-th sampling time is greater than the first parameter threshold of the nth power quality indicator and is less than or equal to the second parameter threshold of the nth power quality indicator, the formula for calculating the quality indicator score corresponding to the nth power quality indicator at the i-th sampling time is:

[0070]

[0071] in, Score the quality index corresponding to the nth power quality index at the i-th sampling moment; is the parameter value corresponding to the nth power quality indicator at the i-th sampling moment; i is the sequence number of the sampling moment in the preset time period; is the first parameter threshold of the nth power quality indicator; is the second parameter threshold of the nth power quality indicator; n is the power quality indicator sequence number among the N power quality indicators; and 0.8 represents the second preset multiple.

[0072] It should be noted that if , determine that the quality indicator score corresponding to the nth power quality indicator at the i-th sampling moment is qualified, and the range of the quality indicator score is 0-0.8. There is no restriction here, and it can be set according to the specific implementation scenario. The larger the parameter value corresponding to the nth power quality indicator at the i-th sampling moment, the smaller the quality indicator score corresponding to the nth power quality indicator at the i-th sampling moment.

[0073] (4) If the parameter value corresponding to the power quality indicator at the sampling time is greater than the second parameter threshold, setting the quality indicator score corresponding to the power quality indicator at the sampling time to a preset value.

[0074] In one embodiment, taking the nth power quality indicator as an example, the parameter value corresponding to the nth power quality indicator at the i-th sampling time is obtained. If the parameter value corresponding to the nth power quality indicator at the i-th sampling time is , then the quality index score corresponding to the nth power quality index at the i-th sampling moment is set to a preset value of 0. There is no restriction here and it can be set according to the specific implementation scenario.

[0075] According to the method for obtaining the quality indicator score corresponding to the nth power quality indicator at the i-th sampling moment, the quality indicator scores corresponding to each sampling moment of the N types of power quality indicators within a preset time period are obtained.

[0076] At this point, the quality index score corresponding to each power quality index at each sampling moment within the preset time period is obtained.

[0077] Step S102, for any sampling moment within the preset time period, a time window of preset length is established with the sampling moment as the center, and according to the quality indicator score corresponding to each sampling moment in the time window, it is detected whether there is a score to be corrected in the sampling moment.

[0078] In the process of obtaining the quality index score, each electric energy data in the obtained multidimensional electric energy data will be affected by many factors such as load changes, power supply fluctuations, environmental factors, etc. and produce instantaneous fluctuations. When the electric energy data in the multidimensional electric energy data has instantaneous fluctuations, it will cause the parameter value of the electric energy quality index corresponding to the sampling moment to be abnormal, and then cause the quality index score corresponding to the sampling moment to be abnormal, thereby affecting the evaluation of the electric energy quality of the residential area. Therefore, for any sampling moment within the preset time period, it is necessary to detect whether there is an abnormal quality index score caused by the instantaneous fluctuation of the electric energy data at the sampling moment, that is, the score to be corrected, and correct the score to be corrected.

[0079] Since the score to be corrected is caused by the instantaneous fluctuation of the power data in the multi-dimensional power data, it is necessary to judge the quality index score of the sampling moment according to the quality index scores of the same power quality index within a certain range before and after the sampling moment. Therefore, with the sampling moment as the center, a time window including the 10 sampling moments before and after the sampling moment is established. There is no restriction here. It can be set according to the specific implementation scenario. According to the quality index score corresponding to each sampling moment in the time window, it is detected whether there is a score to be corrected in the sampling moment.

[0080] According to the quality indicator score corresponding to each sampling moment in the time window, the method for detecting whether there is a score to be corrected in the sampling moment is as follows:

[0081] For any quality indicator score at the sampling moment, all quality indicator scores belonging to the same power quality indicator as the quality indicator score are obtained in the time window, and form a quality indicator score sequence with the quality indicator score;

[0082] Fitting the quality indicator score sequence to obtain a fitting function, using the fitting function to predict the quality indicator score to obtain a predicted value, obtaining an absolute value of a difference between the quality indicator score and the predicted value, normalizing the absolute value of the difference to obtain an abnormal coefficient of the quality indicator score;

[0083] Obtaining an abnormal coefficient of each quality indicator score in the quality indicator score sequence, and when the abnormal coefficient of the quality indicator score is greater than or equal to a preset abnormal coefficient threshold, and the abnormal coefficient of each quality indicator score in the quality indicator score sequence except the quality indicator score is less than the preset abnormal coefficient threshold, confirming that the quality indicator score is a score to be revised;

[0084] When the abnormal coefficient of the quality indicator score is greater than or equal to the preset abnormal coefficient threshold, and in the quality indicator score sequence, there is at least one quality indicator score other than the quality indicator score whose abnormal coefficient is greater than or equal to the preset abnormal coefficient threshold, confirming that the quality indicator score is a normal score;

[0085] When the abnormal coefficient of the quality indicator score is less than a preset abnormal coefficient threshold, the quality indicator score is confirmed to be a normal score.

[0086] In one embodiment, taking the quality indicator score corresponding to the nth power quality indicator at the ith sampling moment as an example, in the time window of the ith sampling moment, all quality indicator scores of the nth power quality indicator are obtained to form a quality indicator score sequence. The least squares method is used to perform data fitting on all quality indicator scores in the quality indicator score sequence to obtain a score fitting function, wherein the least squares method belongs to the prior art and is not described here in detail. The score fitting function is used to predict the quality indicator score corresponding to the nth power quality indicator at the ith sampling moment to obtain a predicted value, and the abnormal coefficient of the quality indicator score corresponding to the nth power quality indicator at the ith sampling moment is calculated:

[0087]

[0088] in, is the abnormal coefficient of the quality index score corresponding to the nth power quality index at the i-th sampling moment; is the predicted value of the quality index score corresponding to the nth power quality index at the i-th sampling moment; Score the quality index corresponding to the nth power quality index at the i-th sampling moment; is the parameter value corresponding to the nth power quality indicator at the i-th sampling moment; i is the sequence number of the sampling moment in the preset time period; n is the sequence number of the power quality indicator among the N power quality indicators; norm() is the normalization function; | | is the absolute value symbol.

[0089] It should be noted that the greater the difference between the quality indicator score corresponding to the n-th power quality indicator at the ith sampling moment and the predicted value, the greater the abnormality coefficient of the quality indicator score corresponding to the n-th power quality indicator at the ith sampling moment, indicating that the quality indicator score corresponding to the n-th power quality indicator at the ith sampling moment is more likely to be a score to be corrected.

[0090] The preset abnormal coefficient threshold is set to 0.4, which is not limited here. It can be set according to the specific implementation scenario to obtain the abnormal coefficient of each quality indicator score in the quality indicator scoring sequence. When the abnormal coefficient of the quality indicator score corresponding to the n-th power quality indicator at the ith sampling moment is greater than or equal to 0.4, and the abnormal coefficient of each quality indicator score in the quality indicator scoring sequence except the quality indicator score corresponding to the n-th power quality indicator at the ith sampling moment is less than 0.4, it means that the abnormality of the quality indicator score corresponding to the n-th power quality indicator at the ith sampling moment is caused by instantaneous fluctuations in power data, and the quality indicator score corresponding to the n-th power quality indicator at the ith sampling moment is determined to be the score to be corrected; when the quality indicator score corresponding to the n-th power quality indicator at the ith sampling moment is When the abnormal coefficient of the quality indicator score is greater than or equal to 0.4, and in the quality indicator score sequence, except for the quality indicator score corresponding to the nth power quality indicator at the ith sampling moment, there is at least one quality indicator score with an abnormal coefficient greater than or equal to 0.4, it means that the abnormality of the quality indicator score corresponding to the nth power quality indicator at the ith sampling moment is not caused by the instantaneous fluctuation of the power data, and the quality indicator score corresponding to the nth power quality indicator at the ith sampling moment is determined to be a normal score; when the abnormal coefficient of the quality indicator score corresponding to the nth power quality indicator at the ith sampling moment is less than 0.4, it means that the quality indicator score corresponding to the nth power quality indicator at the ith sampling moment is normal, and the quality indicator score corresponding to the nth power quality indicator at the ith sampling moment is determined to be a normal score.

[0091] If there are not 10 sampling moments before the i-th sampling moment, each sampling moment before the i-th sampling moment and the 10 sampling moments after the i-th sampling moment will form the time window of the i-th sampling moment; if there are not 10 sampling moments after the i-th sampling moment, each sampling moment after the i-th sampling moment and the 10 sampling moments before the i-th sampling moment will form the time window of the i-th sampling moment.

[0092] According to the above-mentioned method for detecting the quality indicator score corresponding to the nth power quality indicator at the i-th sampling moment, it is detected whether the quality indicator score corresponding to each power quality indicator at the i-th sampling moment is a score to be corrected.

[0093] At this point, it is determined whether there is a score to be corrected in the quality indicator scores corresponding to all power quality indicators at the i-th sampling moment.

[0094] Step S103: If it is detected that there is at least one score to be corrected at the sampling moment, then according to all quality indicator scores at each sampling moment in the time window, each score to be corrected at the sampling moment is corrected to obtain a corresponding corrected score, and all non-corrected scores and all corrected scores at the sampling moment are accumulated to obtain a final power quality score at the sampling moment.

[0095] When no score to be corrected is detected at the i-th sampling moment, it means that each quality indicator score at the i-th sampling moment is normal, so no correction is required, and the final power quality score at the i-th sampling moment can be directly obtained.

[0096] Specifically, if no score to be corrected is detected at the sampling moment, the sum of all quality indicator scores at the sampling moment is accumulated as the final power quality score at the sampling moment.

[0097] In one implementation, taking the i-th sampling moment as an example, the final power quality score at the i-th sampling moment is calculated:

[0098]

[0099] in, is the final power quality score at the i-th sampling moment; Score the quality index corresponding to the nth power quality index at the i-th sampling moment; is the parameter value corresponding to the nth power quality indicator at the i-th sampling moment; i is the sequence number of the sampling moment in the preset time period; n is the sequence number of the power quality indicator among the N types of power quality indicators; and N is the total number of power quality indicators.

[0100] It should be noted that, the higher the quality index score of each power quality index at the i-th sampling moment, the higher the final power quality score at the i-th sampling moment, indicating that the power quality at the i-th sampling moment is better.

[0101] When a score to be revised is detected at the i-th sampling moment, since the score to be revised will affect the evaluation of the power quality of the residential area, it is necessary to revise the score to be revised to reduce the impact of the score to be revised on the evaluation of the power quality of the residential area. For any score to be revised at the i-th sampling moment, since in the time window of the i-th sampling moment, each quality indicator score except the quality indicator score corresponding to the n-th power quality indicator at the i-th sampling moment is normal, and the power quality indicators are calculated from multi-dimensional power data, and the N power quality indicators have a certain correlation with each other, the score to be revised can be revised according to all quality indicator scores at each sampling moment in the time window of the i-th sampling moment.

[0102] Among them, the method for correcting the score to be corrected according to all quality indicator scores at each sampling moment in the time window of the i-th sampling moment is:

[0103] (1) For any score to be corrected at the sampling moment, a first related indicator and a second related indicator related to the power quality indicator corresponding to the score to be corrected are obtained from the N types of power quality indicators.

[0104] In one implementation, since the power quality indicators have a certain correlation with each other, the first related indicator and the second related indicator of each power quality indicator are set according to the correlation strength between the power quality indicators, as shown below:

[0105] For the overload rate, a decrease in voltage deviation will lead to a larger overload rate, and an increase in the harmonic impact index will also lead to a larger overload rate. Therefore, the first related index related to the overload rate is set to be the voltage deviation, and the second related index related to the overload rate is set to be the harmonic impact index;

[0106] For voltage deviation, an increase in overload rate will lead to a decrease in voltage deviation, and an increase in harmonic impact index will also lead to a decrease in voltage deviation. Therefore, the first related index related to voltage deviation is set to be the overload rate, and the second related index related to voltage deviation is set to be the harmonic impact index;

[0107] For the harmonic impact index, an increase in the three-phase voltage unbalance rate will lead to an increase in the harmonic impact index, and a decrease in the voltage deviation will also lead to an increase in the harmonic impact index. Therefore, the first related index related to the harmonic impact index is set to be the three-phase voltage unbalance rate, and the second related index related to the harmonic impact index is set to be the voltage deviation;

[0108] For the three-phase voltage imbalance rate, the aggravation of the harmonic impact index will lead to an increase in the three-phase voltage imbalance rate, and the reduction of voltage deviation will also lead to an increase in the three-phase voltage imbalance rate. Therefore, the first related indicator related to the three-phase voltage imbalance rate is set as the harmonic impact index, and the second related indicator related to the three-phase voltage imbalance rate is set as the voltage deviation.

[0109] For any score to be corrected at the i-th sampling moment, according to the correlation between the above power quality indicators, a first related indicator and a second related indicator related to the power quality indicator corresponding to the score to be corrected are obtained from the N power quality indicators.

[0110] (2) Obtaining a first correlation score of the score to be corrected according to the quality indicator score corresponding to the first correlation indicator at the sampling time, and obtaining a correlation coefficient between the score to be corrected and the first correlation score, recorded as a first correlation coefficient.

[0111] Wherein, obtaining the first relevant score of the score to be corrected according to the quality indicator score corresponding to the first relevant indicator at the sampling time includes:

[0112] When the quality indicator score corresponding to the first relevant indicator at the sampling time is a normal score, taking the quality indicator score corresponding to the first relevant indicator at the sampling time as the first relevant score of the score to be corrected;

[0113] When the quality indicator score corresponding to the first related indicator at the sampling moment is a score to be revised, detecting whether there is a normal score at the sampling moment, and if it is detected that there is at least one normal score at the sampling moment, taking any normal score at the sampling moment as an available score at the sampling moment; if it is detected that there is no normal score at the sampling moment, obtaining a preset target power quality indicator, and obtaining an average of the quality indicator scores corresponding to the target power quality indicator in the first preset sub-period and the second preset sub-period as the available score at the sampling moment;

[0114] The expected score of the first relevant indicator at the sampling moment is obtained according to the available score at the sampling moment, and the expected score of the first relevant indicator at the sampling moment is used as the first relevant score of the score to be revised.

[0115] In one implementation, according to a method for detecting a quality indicator score corresponding to an nth type of power quality indicator at an i-th sampling moment, it is detected whether the quality indicator score corresponding to a first related indicator at an i-th sampling moment is a normal score or a score to be corrected; when the quality indicator score corresponding to the first related indicator at an i-th sampling moment is a normal score, the quality indicator score corresponding to the first related indicator at an i-th sampling moment is used as a first related score of the score to be corrected;

[0116] When the quality indicator score corresponding to the first related indicator at the i-th sampling moment is a score to be corrected, it is detected whether there is a normal score at the i-th sampling moment. If it is detected that there is a normal score in the quality indicator scores corresponding to the i-th sampling moment, any normal score in the quality indicator scores corresponding to the i-th sampling moment is used as the available score at the i-th sampling moment; if it is detected that there is no normal score in the quality indicator scores corresponding to the i-th sampling moment, it is necessary to select a power quality indicator that is least affected by the external environment from the power quality indicators corresponding to the i-th sampling moment as the target power quality indicator. Since the generation of harmonics mainly depends on the characteristics of nonlinear loads, the impact of load changes on harmonics is relatively small, and the correlation between harmonics and power supply fluctuations and environmental factors is relatively low, the harmonic impact indicator is more stable than other power quality indicators when the external environment changes, so the harmonic impact indicator is selected as the target power quality indicator.

[0117] In the time window of the i-th sampling moment, the first 10 sampling moments of the i-th sampling moment are combined into a first preset sub-period (if there are no 10 sampling moments before the i-th sampling moment, all sampling moments before the i-th sampling moment are combined into the first preset sub-period), and the last 10 sampling moments of the i-th sampling moment are combined into a second preset sub-period (if there are no 10 sampling moments after the i-th sampling moment, all sampling moments after the i-th sampling moment are combined into the second preset sub-period), all quality indicator scores of the harmonic impact index are obtained in the first preset sub-period, and all quality indicator scores of the harmonic impact index are obtained in the second preset sub-period, and the average of all quality indicator scores of the harmonic impact index in the first preset sub-period and the second preset sub-period is calculated as the available score in the i-th sampling moment; the expected score of the first related indicator at the i-th sampling moment is obtained according to the available score in the i-th sampling moment, and the expected score of the first related indicator at the i-th sampling moment is used as the first related score to be corrected.

[0118] The correlation coefficient between the score to be corrected and the first correlation score is obtained, recorded as the first correlation coefficient, including:

[0119] In a first preset sub-period, all quality indicator scores that belong to the same power quality indicator as the score to be corrected are obtained to form a first scoring sequence. In a first preset sub-period, all quality indicator scores that belong to the same power quality indicator as the first related score are obtained to form a second scoring sequence. The absolute value of the correlation coefficient between the first scoring sequence and the second scoring sequence is obtained and recorded as a first coefficient.

[0120] In the second preset sub-period, all quality indicator scores that belong to the same power quality indicator as the score to be corrected are obtained to form a third scoring sequence. In the second preset sub-period, all quality indicator scores that belong to the same power quality indicator as the first related score are obtained to form a fourth scoring sequence. The absolute value of the correlation coefficient between the third scoring sequence and the fourth scoring sequence is obtained and recorded as the second coefficient.

[0121] An average value of the first coefficient and the second coefficient is obtained as a correlation coefficient between the score to be corrected and the first relevant score.

[0122] In one implementation, taking the i-th sampling moment as an example, within the time window of the i-th sampling moment, the first 10 sampling moments of the i-th sampling moment are composed of a first preset sub-period, and the last 10 sampling moments of the i-th sampling moment are composed of a second preset sub-period, in the first preset sub-period, all quality indicator scores belonging to the same power quality indicator as the score to be corrected are obtained to form a first scoring sequence, in the first preset sub-period, all quality indicator scores belonging to the same power quality indicator as the first related score are obtained to form a second scoring sequence, and the Pearson correlation coefficient between the first scoring sequence and the second scoring sequence is obtained; in the second preset sub-period, all quality indicator scores belonging to the same power quality indicator as the score to be corrected are obtained to form a third scoring sequence, in the second preset sub-period, all quality indicator scores belonging to the same power quality indicator as the first related score are obtained to form a fourth scoring sequence, and the Pearson correlation coefficient between the third scoring sequence and the fourth scoring sequence is obtained, wherein the Pearson correlation coefficient belongs to the prior art and will not be repeated here; the correlation coefficient between the score to be corrected and the first related score is calculated:

[0123]

[0124] in, is the correlation coefficient between the score to be corrected and the first related score; r1 is the Pearson correlation coefficient between the first score sequence and the second score sequence; r2 is the Pearson correlation coefficient between the third score sequence and the fourth score sequence; | | is the absolute value symbol.

[0125] It should be noted that the greater the absolute value of the Pearson correlation coefficient between the first scoring sequence and the second scoring sequence, and the greater the absolute value of the Pearson correlation coefficient between the third scoring sequence and the fourth scoring sequence, the greater the correlation coefficient between the score to be revised and the first relevant score, which means that the greater the correlation between the score to be revised and the first relevant score, the greater the proportion of the first relevant score in the process of revising the score to be revised.

[0126] If there are not 10 sampling moments before the i-th sampling moment, the first preset sub-period is not needed, and the 10 sampling moments after the i-th sampling moment form a second preset sub-period, and all quality indicator scores belonging to the same power quality indicator as the score to be corrected are obtained in the second preset sub-period to form a third scoring sequence, and all quality indicator scores belonging to the same power quality indicator as the first related score are obtained in the second preset sub-period to form a fourth scoring sequence, and the Pearson correlation coefficient of the third scoring sequence and the fourth scoring sequence is obtained, and the correlation coefficient between the score to be corrected and the first related score is calculated:

[0127]

[0128] in, is the correlation coefficient between the score to be corrected and the first related score; r2 is the Pearson correlation coefficient between the third score sequence and the fourth score sequence; | | is the absolute value symbol.

[0129] It should be noted that the larger the absolute value of the Pearson correlation coefficient between the third scoring sequence and the fourth scoring sequence, the larger the correlation coefficient between the score to be revised and the first related score, which means that the greater the correlation between the score to be revised and the first related score, the greater the proportion of the first related score in the process of revising the score to be revised.

[0130] If there are no 10 sampling moments after the i-th sampling moment, the second preset sub-period is not needed, and the first 10 sampling moments of the i-th sampling moment form a first preset sub-period, and all quality indicator scores belonging to the same power quality indicator as the score to be corrected are obtained in the first preset sub-period to form a first scoring sequence, and all quality indicator scores belonging to the same power quality indicator as the first related score are obtained in the first preset sub-period to form a second scoring sequence, and the Pearson correlation coefficient of the first scoring sequence and the second scoring sequence is obtained, and the correlation coefficient between the score to be corrected and the first related score is calculated:

[0131]

[0132] in, is the correlation coefficient between the score to be corrected and the first related score; r1 is the Pearson correlation coefficient between the first score sequence and the second score sequence; | | is the absolute value symbol.

[0133] It should be noted that the larger the absolute value of the Pearson correlation coefficient between the first scoring sequence and the second scoring sequence, the larger the correlation coefficient between the score to be revised and the first related score, which means that the greater the correlation between the score to be revised and the first related score, the greater the proportion of the first related score in the process of revising the score to be revised.

[0134] Wherein, obtaining the expected score of the first related indicator at the i-th sampling moment according to the available score at the i-th sampling moment includes:

[0135] The quality indicator score corresponding to the first related indicator at the sampling moment is recorded as the abnormal score. If the first related indicator or the second related indicator related to the power quality indicator corresponding to the available score is the power quality indicator corresponding to the abnormal score, the correlation coefficient between the available score and the abnormal score is obtained, and the product of the correlation coefficient between the available score and the abnormal score and the available score is calculated as the first product. The mean of the quality indicator scores of the power quality indicator corresponding to the available score in the first preset sub-period and the second preset sub-period is obtained as the first mean, the product of the correlation coefficient between the available score and the abnormal score and the first mean is calculated as the second product, and the mean of the quality indicator scores of the power quality indicator corresponding to the abnormal score in the first preset sub-period and the second preset sub-period is obtained as the second mean. The addition result of the first product and the second mean is calculated, and the difference between the addition result and the second product is obtained as the expected score of the first related indicator at the sampling moment.

[0136] In one embodiment, the quality index score corresponding to the first related index at the i-th sampling time is recorded as the abnormality score When the first related indicator or the second related indicator related to the power quality indicator corresponding to the available score is the power quality indicator corresponding to the abnormal score, the correlation coefficient between the available score and the abnormal score is obtained according to the method for obtaining the correlation coefficient between the score to be corrected and the first related score. , the available score obtained is recorded as , obtaining the average of the quality index scores of the power quality index corresponding to the available score in the first preset sub-period and the second preset sub-period as the first average , obtaining the average of the quality index scores of the power quality index corresponding to the abnormal score in the first preset sub-period and the second preset sub-period as the second average , calculate the expected score of the first relevant indicator at the i-th sampling time:

[0137]

[0138] in, is the expected score of the first relevant indicator at the i-th sampling time; is the correlation coefficient between the available score and the abnormal score; Rating for availability; is the second mean; is the first mean; i is the sequence number of the sampling moment in the preset time period; n is the sequence number of the power quality indicator among the N types of power quality indicators.

[0139] It should be noted that the larger the correlation coefficient between the available score and the abnormal score, the greater the correlation between the expected available score and the abnormal score, and the greater the proportion of the available score in the calculation of the expected score.

[0140] If the first correlation indicator and the second correlation indicator related to the power quality indicator corresponding to the available score are not the power quality indicators corresponding to the abnormal score, the first correlation indicator related to the power quality indicator corresponding to the available score is marked as the first indicator, and the second correlation indicator related to the power quality indicator corresponding to the available score is marked as the second indicator. According to the correlation coefficients between the quality indicator score corresponding to the first indicator and the quality indicator score corresponding to the second indicator and the available score, respectively, the expected scores of the first indicator and the second indicator at the sampling time are obtained, which are recorded as candidate expected scores. According to the correlations between the power quality indicator corresponding to the abnormal score and the first indicator and the second indicator, respectively, a target expected score is obtained from the candidate expected scores. The target expected score is used as a new available score. According to the method for obtaining the expected score of the first correlation indicator at the sampling time according to the available score at the sampling time, the expected score of the first correlation indicator at the sampling time is obtained using the new available score.

[0141] Preferably, the method for obtaining the target expected score from the candidate expected scores is as follows:

[0142] If the first related indicator or the second related indicator related to the first indicator is the power quality indicator corresponding to the abnormal score, the expected score of the first indicator at the sampling time is used as the target expected score;

[0143] If neither the first correlation indicator and the second correlation indicator related to the first indicator are the power quality indicators corresponding to the abnormal score, detect whether any of the first correlation indicator and the second correlation indicator related to the second indicator is the power quality indicator corresponding to the abnormal score; if it is detected that the first correlation indicator or the second correlation indicator related to the second indicator is the power quality indicator corresponding to the abnormal score, take the expected score of the second indicator at the sampling time as the target expected score.

[0144] For example, when the quality indicator score corresponding to the overload rate at the i-th sampling moment is an abnormal score, and the only available score at the i-th sampling moment is the quality indicator score corresponding to the three-phase voltage imbalance rate, the first related indicator and the second related indicator of the three-phase voltage imbalance rate are the harmonic impact index and the voltage deviation, respectively. The harmonic impact index is marked as the first indicator, and the voltage deviation is recorded as the second indicator. According to the method for obtaining the expected score of the first related indicator at the i-th sampling moment, the expected scores of the first indicator (harmonic impact index) and the second indicator (voltage deviation) at the i-th sampling moment are obtained respectively and recorded as candidate expected scores. Since the first related indicator related to the first indicator (harmonic impact index) is the three-phase voltage imbalance rate, the second related indicator related to the first indicator (harmonic impact index) is the voltage deviation, and the first related indicator related to the second indicator (voltage deviation) is the overload rate, the expected score of the second indicator (i.e., voltage deviation) at the i-th sampling moment is taken as the current expected score, and the target expected score (i.e., the expected score of the voltage deviation at the i-th sampling moment) is taken as the new available score. According to the method for obtaining the expected score of the first related indicator at the sampling moment according to the available score at the sampling moment, the expected score of the first related indicator at the i-th sampling moment is obtained using the new available score.

[0145] (3) Obtaining a second correlation score of the score to be corrected according to the quality indicator score corresponding to the second correlation indicator at the sampling time, and obtaining a correlation coefficient between the score to be corrected and the second correlation score, recorded as a second correlation coefficient.

[0146] According to the method for obtaining the first relevant score of the score to be revised and the method for obtaining the correlation coefficient between the score to be revised and the first relevant score, the correlation coefficient between the score to be revised and the second relevant score is obtained and recorded as the second correlation coefficient.

[0147] (4) Obtaining the instantaneous change rates of the first correlation coefficient and the second correlation coefficient within the time window respectively, obtaining the absolute value of the difference between the instantaneous change rate of the first correlation coefficient and the instantaneous change rate of the second correlation coefficient, and normalizing the absolute value of the difference to obtain an error coefficient.

[0148] In one embodiment, according to the method for obtaining the correlation coefficient between the score to be corrected and the first correlation score, within the time window of the i-th sampling moment, the quality index score of each sampling moment that belongs to the same power quality index as the score to be corrected, and the correlation coefficient between the quality index score that belongs to the same power quality index as the first correlation score are obtained to form a first correlation coefficient sequence, and a first correlation coefficient curve is drawn according to the first correlation coefficient sequence, and the instantaneous change rate of the i-th sampling moment in the first correlation coefficient curve is calculated as the instantaneous change rate of the first correlation coefficient in the time window of the i-th sampling moment; according to the method for obtaining the instantaneous change rate of the first correlation coefficient in the time window of the i-th sampling moment, the instantaneous change rate of the second correlation coefficient in the time window of the i-th sampling moment is obtained, and the instantaneous change rate belongs to the prior art and will not be repeated here, and the error coefficient is calculated:

[0149]

[0150] in, is the error coefficient; K1 is the instantaneous change rate of the first correlation coefficient; K2 is the instantaneous change rate of the second correlation coefficient; norm() is the normalization function; | | is the absolute value symbol.

[0151] It should be noted that the instantaneous change rate of the first correlation coefficient represents the change trend of the first correlation coefficient in the time window at the i-th sampling moment, the instantaneous change rate of the second correlation coefficient represents the change trend of the second correlation coefficient in the time window at the i-th sampling moment, and the error coefficient represents the similarity of the change trends of the first correlation coefficient and the second correlation coefficient. Under normal circumstances, the change trends of the first correlation coefficient and the second correlation coefficient are the same. Therefore, the greater the difference between the instantaneous change rate of the first correlation coefficient and the instantaneous change rate of the second correlation coefficient, the greater the error coefficient, indicating that the change trends of the first correlation coefficient and the second correlation coefficient do not conform to the normal situation, that is, the first correlation coefficient and the second correlation coefficient are less credible. Therefore, the smaller the proportion of the first correlation coefficient and the first score in the process of correcting the score to be corrected, the greater the proportion of the score to be corrected itself in the process of correction.

[0152] (5) Using the first correlation coefficient as the first weight of the first correlation score, using the error coefficient as the second weight of the score to be revised, and performing weighted summation on the first correlation score and the score to be revised to obtain a revised score of the score to be revised.

[0153] In one embodiment, the quality indicator score corresponding to the nth type of power quality indicator at the i-th sampling moment is used as the score to be corrected. , the first relevant score of the quality index score corresponding to the nth power quality index at the i-th sampling moment is , calculate the revised score of the score to be revised:

[0154]

[0155] in, is the revised score; k1 is the first correlation coefficient (i.e., the correlation coefficient between the score to be revised and the first correlation score); is the error coefficient; Score the first correlation; Rating for revision.

[0156] It should be noted that, the larger the first correlation coefficient is, the greater the proportion of the first correlation score in the revised score is; the larger the error coefficient is, the greater the proportion of the score to be revised itself in the revised score is.

[0157] According to the method for obtaining the corrected score, the corrected score of each score to be corrected at the i-th sampling moment is obtained, and according to the corrected score at the i-th sampling moment, the final power quality score at the i-th sampling moment is obtained.

[0158] Specifically, all non-corrected scores and all corrected scores at the i-th sampling moment are accumulated to obtain a final power quality score at the i-th sampling moment.

[0159] In one embodiment, the mth uncorrected score at the i-th sampling moment is recorded as , the sth modified score at the i-th sampling moment is recorded as , calculate the final power quality score at the i-th sampling moment:

[0160]

[0161] in, is the final power quality score at the i-th sampling moment; is the mth uncorrected score at the ith sampling moment; is the sth corrected score in the ith sampling moment; i is the serial number of the sampling moment in the preset time period; m is the serial number of the non-corrected score in the ith sampling moment; s is the serial number of the corrected score in the ith sampling moment; M is the total number of non-corrected scores in the ith sampling moment; S is the total number of corrected scores in the ith sampling moment.

[0162] It should be noted that the larger the non-corrected score at the i-th sampling moment, the larger the final power quality score at the i-th sampling moment; the larger the corrected score at the i-th sampling moment, the larger the final power quality score at the i-th sampling moment.

[0163] At this point, the final power quality score at the i-th sampling moment is obtained.

[0164] Step S104, respectively obtaining the final power quality score at each sampling moment in the preset time period, and evaluating the power quality of the residential area in the preset time period according to the final power quality score at each sampling moment in the preset time period.

[0165] After obtaining the final power quality score at the i-th sampling moment, the final power quality score at each sampling moment in the preset time period is obtained according to the method for obtaining the final power quality score at the i-th sampling moment, and the power quality of the residential area in the preset time period is evaluated according to the final power quality score at each sampling moment in the preset time period.

[0166] Among them, according to the final power quality score at each sampling moment in the preset period, the method for evaluating the power quality of the residential area in the preset period is as follows:

[0167] (1) Setting a power quality threshold. If the final power quality score at any sampling moment within the preset time period is greater than or equal to the power quality threshold, the power quality at the sampling moment is determined to be qualified.

[0168] In one implementation, the power quality threshold is set to 1.5, which is not limited here and can be set according to a specific implementation scenario. Taking the i-th sampling moment in a preset time period as an example, if the final power quality score at the i-th sampling moment is greater than or equal to 1.5, the final power quality score at the i-th sampling moment is qualified.

[0169] (2) Obtaining a pass rate of the power quality in the preset time period; when the pass rate of the power quality in the preset time period is greater than a preset pass rate threshold, confirming that the power quality of the residential area in the preset time period is qualified.

[0170] In one embodiment, the pass rate threshold of the power quality in the preset time period is set to 90%. There is no limitation here and it can be set according to the specific implementation scenario. According to the above method of judging whether the final power quality score at the i-th sampling moment is qualified, the final power quality score at each sampling moment in the preset time period is judged to obtain the pass rate of the power quality in the preset time period. When the pass rate of the power quality in the preset time period is greater than 90%, it is determined that the power quality of the residential area in the preset time period is qualified.

[0171] In summary, the present embodiment obtains multi-dimensional electric energy data of the residential area at any sampling moment, obtains the parameter values ​​corresponding to the preset N kinds of electric energy quality indicators at the sampling moments, obtains the quality indicator scores corresponding to the N kinds of electric energy quality indicators at the sampling moments according to all the parameter values, and obtains the quality indicator scores corresponding to each sampling moment of the N kinds of electric energy quality indicators within the preset time period; for any sampling moment within the preset time period, a time window of preset length is established with the sampling moment as the center, and according to the quality indicator score corresponding to each sampling moment in the time window, it is detected whether there is a score to be corrected in the sampling moment; if it is detected that there is at least one score to be corrected in the sampling moment, then according to all the quality indicator scores of each sampling moment in the time window, each score to be corrected at the sampling moment is corrected to obtain the corresponding corrected score, and all the non-corrected scores and all the corrected scores at the sampling moment are accumulated to obtain the final electric energy quality score at the sampling moment; the final electric energy quality score of each sampling moment in the preset time period is obtained respectively, and the electric energy quality of the residential area within the preset time period is evaluated according to the final electric energy quality score of each sampling moment in the preset time period. This embodiment detects the scores of various quality indicators of residential substations, obtains the scores to be corrected due to instantaneous fluctuations and corrects them, and evaluates the power quality of the residential substations through the final power quality scores obtained by correcting the scores, thereby reducing the problem of inaccurate final power quality scores caused by abnormal data generated by instantaneous fluctuations, thereby improving the accuracy and reliability of power quality evaluation of residential substations.

[0172] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A residential area power quality intelligent detection system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the following method is implemented: Obtain multi-dimensional electric energy data of the residential area at any sampling time, obtain parameter values ​​corresponding to N preset power quality indicators at the sampling time, obtain quality indicator scores corresponding to the N preset power quality indicators at the sampling time according to all parameter values, and obtain quality indicator scores corresponding to each sampling time of the N preset power quality indicators within a preset time period; For any sampling moment within the preset time period, a time window of preset length is established with the sampling moment as the center, and according to the quality indicator score corresponding to each sampling moment within the time window, it is detected whether there is a score to be corrected in the sampling moment; If it is detected that there is at least one score to be corrected at the sampling moment, then according to all quality indicator scores at each sampling moment in the time window, each score to be corrected at the sampling moment is corrected to obtain a corresponding corrected score, and all non-corrected scores and all corrected scores at the sampling moment are accumulated to obtain a final power quality score at the sampling moment; Respectively obtain the final power quality score at each sampling moment in the preset time period, and evaluate the power quality of the residential area in the preset time period according to the final power quality score at each sampling moment in the preset time period; The step of detecting whether there is a score to be corrected at each sampling moment according to the quality indicator score corresponding to each sampling moment in the time window comprises: For any quality indicator score at the sampling moment, all quality indicator scores belonging to the same power quality indicator as the quality indicator score are obtained in the time window, and form a quality indicator score sequence with the quality indicator score; Fitting the quality indicator score sequence to obtain a fitting function, using the fitting function to predict the quality indicator score to obtain a predicted value, obtaining an absolute value of a difference between the quality indicator score and the predicted value, normalizing the absolute value of the difference to obtain an abnormal coefficient of the quality indicator score; Obtaining an abnormal coefficient of each quality indicator score in the quality indicator score sequence, and when the abnormal coefficient of the quality indicator score is greater than or equal to a preset abnormal coefficient threshold, and the abnormal coefficient of each quality indicator score in the quality indicator score sequence except the quality indicator score is less than the preset abnormal coefficient threshold, confirming that the quality indicator score is a score to be revised; When the abnormal coefficient of the quality indicator score is greater than or equal to the preset abnormal coefficient threshold, and in the quality indicator score sequence, there is at least one quality indicator score other than the quality indicator score whose abnormal coefficient is greater than or equal to the preset abnormal coefficient threshold, confirming that the quality indicator score is a normal score; When the abnormal coefficient of the quality indicator score is less than a preset abnormal coefficient threshold, the quality indicator score is confirmed to be a normal score.

2. According to claim 1, a residential area power quality intelligent detection system is characterized in that: After detecting whether there is a score to be corrected at the sampling moment, the method further includes: If no score to be corrected is detected at the sampling moment, the sum of all quality indicator scores at the sampling moment is accumulated as the final power quality score at the sampling moment.

3. According to claim 1, a residential area power quality intelligent detection system is characterized in that: The obtaining, according to all parameter values, quality index scores corresponding to the N types of power quality indicators at the sampling time, respectively, includes: For any power quality indicator among the N power quality indicators, setting a first parameter threshold and a second parameter threshold of the power quality indicator, wherein the first parameter threshold is less than the second parameter threshold; If the parameter value corresponding to the power quality indicator at the sampling time is less than or equal to the first parameter threshold, obtain the ratio of the parameter value of a first preset multiple to the first parameter threshold as a first ratio, and calculate the difference between a constant 1 and the first ratio as the quality indicator score corresponding to the power quality indicator at the sampling time; If the parameter value corresponding to the power quality indicator at the sampling time is greater than the first parameter threshold value and less than or equal to the second parameter threshold value, obtain the difference between the first parameter threshold value and the second parameter threshold value as the first difference value, calculate the ratio of the parameter value of the second preset multiple to the first difference value as the second ratio, calculate the ratio of the second parameter threshold value of the second preset multiple to the first difference value as the third ratio, and obtain the difference between the second ratio and the third ratio as the quality indicator score corresponding to the power quality indicator at the sampling time; If the parameter value corresponding to the power quality indicator at the sampling time is greater than the second parameter threshold, the quality indicator score corresponding to the power quality indicator at the sampling time is set to a preset value.

4. The intelligent detection system for power quality in residential areas according to claim 1 is characterized in that: The step of correcting each score to be corrected at the sampling moment according to all quality indicator scores at each sampling moment within the time window to obtain a corresponding corrected score includes: For any score to be corrected at the sampling moment, obtaining a first related indicator and a second related indicator related to the power quality indicator corresponding to the score to be corrected from the N types of power quality indicators; According to the quality indicator score corresponding to the first relevant indicator at the sampling time, obtaining a first relevant score of the score to be corrected, and obtaining a correlation coefficient between the score to be corrected and the first relevant score, recorded as a first correlation coefficient; According to the quality indicator score corresponding to the second related indicator at the sampling time, obtaining a second related score of the score to be corrected, and obtaining a correlation coefficient between the score to be corrected and the second related score, recorded as a second correlation coefficient; Respectively obtaining the instantaneous change rates of the first correlation coefficient and the second correlation coefficient within the time window, obtaining the absolute value of the difference between the instantaneous change rate of the first correlation coefficient and the instantaneous change rate of the second correlation coefficient, and performing normalization processing on the absolute value of the difference to obtain an error coefficient; The first correlation coefficient is used as the first weight of the first correlation score, the error coefficient is used as the second weight of the score to be revised, and the first correlation score and the score to be revised are weightedly summed to obtain a revised score of the score to be revised.

5. The intelligent detection system for power quality in residential areas according to claim 4 is characterized in that: The obtaining, according to the quality indicator score corresponding to the first related indicator at the sampling time, the first related score of the score to be corrected comprises: When the quality indicator score corresponding to the first relevant indicator at the sampling time is a normal score, taking the quality indicator score corresponding to the first relevant indicator at the sampling time as the first relevant score of the score to be corrected; When the quality indicator score corresponding to the first related indicator at the sampling moment is a score to be revised, detecting whether there is a normal score at the sampling moment, and if it is detected that there is at least one normal score at the sampling moment, taking any normal score at the sampling moment as an available score at the sampling moment; if it is detected that there is no normal score at the sampling moment, obtaining a preset target power quality indicator, and obtaining an average of the quality indicator scores corresponding to the target power quality indicator in the first preset sub-period and the second preset sub-period as the available score at the sampling moment; The expected score of the first relevant indicator at the sampling moment is obtained according to the available score at the sampling moment, and the expected score of the first relevant indicator at the sampling moment is used as the first relevant score of the score to be revised.

6. The intelligent detection system for power quality in residential areas according to claim 4 is characterized in that: The obtaining of the correlation coefficient between the score to be corrected and the first related score includes: In a first preset sub-period, all quality indicator scores that belong to the same power quality indicator as the score to be corrected are obtained to form a first scoring sequence. In a first preset sub-period, all quality indicator scores that belong to the same power quality indicator as the first related score are obtained to form a second scoring sequence. The absolute value of the correlation coefficient between the first scoring sequence and the second scoring sequence is obtained and recorded as a first coefficient. In the second preset sub-period, all quality indicator scores that belong to the same power quality indicator as the score to be corrected are obtained to form a third scoring sequence. In the second preset sub-period, all quality indicator scores that belong to the same power quality indicator as the first related score are obtained to form a fourth scoring sequence. The absolute value of the correlation coefficient between the third scoring sequence and the fourth scoring sequence is obtained and recorded as the second coefficient. An average value of the first coefficient and the second coefficient is obtained as a correlation coefficient between the score to be corrected and the first relevant score.

7. The intelligent detection system for power quality in residential areas according to claim 5 is characterized in that: The obtaining the expected score of the first related indicator at the sampling moment according to the available score at the sampling moment includes: Record the quality index score corresponding to the first related indicator at the sampling moment as the abnormal score; if the first related indicator or the second related indicator related to the power quality indicator corresponding to the available score is the power quality indicator corresponding to the abnormal score, obtain the correlation coefficient between the available score and the abnormal score, calculate the product of the correlation coefficient between the available score and the abnormal score and the available score as the first product, obtain the mean of the quality index scores of the power quality indicator corresponding to the available score in the first preset sub-period and the second preset sub-period as the first mean, calculate the product of the correlation coefficient between the available score and the abnormal score and the first mean as the second product, obtain the mean of the quality index scores of the power quality indicator corresponding to the abnormal score in the first preset sub-period and the second preset sub-period as the second mean, calculate the addition result of the first product and the second mean, and obtain the difference between the addition result and the second product as the expected score of the first related indicator at the sampling moment; If the first correlation indicator and the second correlation indicator related to the power quality indicator corresponding to the available score are not the power quality indicators corresponding to the abnormal score, the first correlation indicator related to the power quality indicator corresponding to the available score is marked as the first indicator, and the second correlation indicator related to the power quality indicator corresponding to the available score is marked as the second indicator. According to the correlation coefficients between the quality indicator score corresponding to the first indicator and the quality indicator score corresponding to the second indicator and the available score, respectively, the expected scores of the first indicator and the second indicator at the sampling time are obtained, which are recorded as candidate expected scores. According to the correlations between the power quality indicator corresponding to the abnormal score and the first indicator and the second indicator, respectively, a target expected score is obtained from the candidate expected scores. The target expected score is used as a new available score. According to the method for obtaining the expected score of the first correlation indicator at the sampling time according to the available score at the sampling time, the expected score of the first correlation indicator at the sampling time is obtained using the new available score.

8. The intelligent detection system for power quality in residential areas according to claim 7 is characterized in that: The obtaining a target expected score from the candidate expected scores according to the correlation between the power quality indicator corresponding to the abnormal score and the first indicator and the second indicator respectively includes: If the first related indicator or the second related indicator related to the first indicator is the power quality indicator corresponding to the abnormal score, the expected score of the first indicator at the sampling time is used as the target expected score; If neither the first correlation indicator and the second correlation indicator related to the first indicator are the power quality indicators corresponding to the abnormal score, detect whether any of the first correlation indicator and the second correlation indicator related to the second indicator is the power quality indicator corresponding to the abnormal score; if it is detected that the first correlation indicator or the second correlation indicator related to the second indicator is the power quality indicator corresponding to the abnormal score, take the expected score of the second indicator at the sampling time as the target expected score.

9. The intelligent detection system for residential power quality according to claim 1 is characterized in that: The method of evaluating the power quality of the residential area within the preset period of time according to the final power quality score at each sampling moment within the preset period of time includes: Setting a power quality threshold, if the final power quality score at any sampling moment within the preset time period is greater than or equal to the power quality threshold, then determining that the power quality at the sampling moment is qualified; The qualified rate of the power quality in the preset time period is obtained, and when the qualified rate of the power quality in the preset time period is greater than a preset qualified rate threshold, it is confirmed that the power quality of the residential area in the preset time period is qualified.

Images