CN117648554B - Intelligent data acquisition method for photovoltaic multifunctional circuit breaker - Google Patents
Intelligent data acquisition method for photovoltaic multifunctional circuit breaker Download PDFInfo
- Publication number
- CN117648554B CN117648554B CN202410114972.7A CN202410114972A CN117648554B CN 117648554 B CN117648554 B CN 117648554B CN 202410114972 A CN202410114972 A CN 202410114972A CN 117648554 B CN117648554 B CN 117648554B
- Authority
- CN
- China
- Prior art keywords
- current
- data
- value
- moment
- current moment
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 24
- 238000005070 sampling Methods 0.000 claims abstract description 33
- 238000007499 fusion processing Methods 0.000 claims abstract description 13
- 238000000556 factor analysis Methods 0.000 claims abstract description 9
- 238000012545 processing Methods 0.000 claims abstract description 9
- 238000010606 normalization Methods 0.000 claims description 15
- 230000007787 long-term memory Effects 0.000 claims description 8
- 238000004458 analytical method Methods 0.000 claims description 7
- 230000006403 short-term memory Effects 0.000 claims description 6
- 230000010485 coping Effects 0.000 abstract description 3
- 230000009471 action Effects 0.000 description 11
- 238000010248 power generation Methods 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 230000008859 change Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 238000005286 illumination Methods 0.000 description 4
- 238000012423 maintenance Methods 0.000 description 3
- 230000001960 triggered effect Effects 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000005452 bending Methods 0.000 description 2
- 238000013480 data collection Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012854 evaluation process Methods 0.000 description 1
- 238000007667 floating Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000002459 sustained effect Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02H—EMERGENCY PROTECTIVE CIRCUIT ARRANGEMENTS
- H02H1/00—Details of emergency protective circuit arrangements
- H02H1/0092—Details of emergency protective circuit arrangements concerning the data processing means, e.g. expert systems, neural networks
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R19/00—Arrangements for measuring currents or voltages or for indicating presence or sign thereof
- G01R19/0092—Arrangements for measuring currents or voltages or for indicating presence or sign thereof measuring current only
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
- G06F18/15—Statistical pre-processing, e.g. techniques for normalisation or restoring missing data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24133—Distances to prototypes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2433—Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02H—EMERGENCY PROTECTIVE CIRCUIT ARRANGEMENTS
- H02H1/00—Details of emergency protective circuit arrangements
- H02H1/0007—Details of emergency protective circuit arrangements concerning the detecting means
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02H—EMERGENCY PROTECTIVE CIRCUIT ARRANGEMENTS
- H02H3/00—Emergency protective circuit arrangements for automatic disconnection directly responsive to an undesired change from normal electric working condition with or without subsequent reconnection ; integrated protection
- H02H3/08—Emergency protective circuit arrangements for automatic disconnection directly responsive to an undesired change from normal electric working condition with or without subsequent reconnection ; integrated protection responsive to excess current
- H02H3/087—Emergency protective circuit arrangements for automatic disconnection directly responsive to an undesired change from normal electric working condition with or without subsequent reconnection ; integrated protection responsive to excess current for dc applications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2123/00—Data types
- G06F2123/02—Data types in the time domain, e.g. time-series data
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Probability & Statistics with Applications (AREA)
- Photovoltaic Devices (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of data processing, in particular to an intelligent data acquisition method for a photovoltaic multifunctional breaker, which comprises the steps of acquiring historical current time sequence data and current prediction time sequence data to form a current time sequence; acquiring a data local distance selection factor at the current moment, and acquiring a K distance neighborhood when performing COF outlier factor analysis on a current predicted value at the current moment according to the data local distance selection factor; according to the current data in the K distance neighborhood of the current moment, a weight factor of a current predicted value of the current moment is obtained, fusion processing is carried out on a current sampling value and the current predicted value of the current moment according to the weight factor, a target current value for threshold judgment of the photovoltaic multifunctional breaker at the current moment is obtained, and the capability of the photovoltaic multifunctional breaker for coping with instantaneous peak current is improved.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to an intelligent data acquisition method for a photovoltaic multifunctional breaker.
Background
In photovoltaic power generation, current fluctuation of a direct current power generation end can directly influence power supply quality of a power grid, and when events such as overhigh current, electric leakage, circuit breaking, power failure of the power grid and the like occur, a circuit breaker needs to act in time to cut off power supply so as to protect the whole power supply module and operators. The multifunctional photovoltaic circuit breaker can be used for setting the action threshold value of the trigger circuit breaker according to the theoretical maximum value of the actual working load in the existing photovoltaic power generation, and can use a modularization technology to add the circuit breaker function according to the actual required scene, so that the problem that the traditional circuit breaker cannot flexibly judge the action threshold value by using single data is solved, and the multifunctional photovoltaic circuit breaker can act on faults including current, voltage and comprehensive load at the same time to protect equipment safety.
In practical application, factors influencing the direct current end of photovoltaic power generation are many, and the working efficiency of the photovoltaic panel of the photovoltaic power generation is influenced by illumination intensity, temperature, shadow shielding and the like. The problem that the voltage of light Fu Bangong is increased due to temperature dip, the current is increased due to suddenly-increased illumination intensity, the current is too high due to equipment short circuit, the current is unstable due to large fine-crushing cloud layer floating and the like can be generated at an unavoidable direct current generating end in the photovoltaic grid-connected power generation, the special sensibility of the photovoltaic multifunctional circuit breaker can well trigger action to break a circuit to protect equipment, but the operation of the photovoltaic multifunctional circuit breaker can be triggered due to some accidental instant fluctuation, so that difficulties are caused to the stable operation and maintenance of the equipment, for example: the phenomenon that the current suddenly rises to be above the threshold value and then falls back is generated under the conditions including but not limited to shielding of fine cloud layers, current noise and the like, and the phenomenon also triggers the photovoltaic multifunctional breaker to work.
Therefore, how to improve the judging capability of the photovoltaic multi-functional circuit breaker to the instantaneous peak current so as to reduce the misjudgment of the photovoltaic multi-functional circuit breaker is a problem to be solved.
Disclosure of Invention
In view of the above, the embodiment of the invention provides an intelligent data acquisition method for a photovoltaic multifunctional breaker, so as to solve the problem of how to improve the judging capability of the photovoltaic multifunctional breaker on the instantaneous peak current and reduce the misjudgment of the photovoltaic multifunctional breaker.
The embodiment of the invention provides an intelligent data acquisition method for a photovoltaic multifunctional breaker, which comprises the following steps of:
acquiring historical current time sequence data of a photovoltaic panel direct current output end before the current moment and current prediction time sequence data in a preset future time period including the current moment, and forming a current time sequence by the historical current time sequence data and the current prediction time sequence data;
Acquiring a data local distance selection factor of the current moment according to the current data difference at two sides of the current predicted value corresponding to the current moment in the current time sequence, and acquiring a K distance neighborhood when performing COF outlier factor analysis on the current predicted value of the current moment according to the data local distance selection factor;
And acquiring a weight factor of the current predicted value at the current moment according to the current data in the K-distance neighborhood at the current moment, acquiring a current sampling value at the current moment, and carrying out fusion processing on the current sampling value at the current moment and the current predicted value according to the weight factor to obtain a target current value for threshold judgment of the photovoltaic multifunctional breaker at the current moment.
Further, the obtaining the local distance selection factor of the current time according to the current data difference at two sides of the current predicted value corresponding to the current time in the current time sequence includes:
Taking a current predicted value corresponding to the current moment in the current time sequence as target data, and forming a first preset number of historical current values before the target data into a local data set of the target data in the current time sequence;
and acquiring the data local distance selection factor at the current moment according to the current data difference at two sides of each historical current value in the local data set of the target data and the distance between the current data and the target data.
Further, the obtaining the local distance selection factor of the current time according to the current data difference at two sides of each historical current value in the local data set of the target data and the distance between the current data and the target data includes:
Selecting a second preset number of current data to form a first subsequence on the left side of the historical current value, and selecting a second preset number of current data to form a second subsequence on the right side of the historical current value, wherein the historical current value is taken as a center in the current time sequence, so as to obtain the similarity between the first subsequence and the second subsequence;
acquiring all neighborhood current data in an initial K-distance neighborhood of the historical current value in the current time sequence, respectively calculating the absolute value of a difference value between a current sampling value and a current prediction value of each neighborhood current data to obtain the sum of the absolute value of the difference value, obtaining the product between the sum of the absolute value of the difference value and the similarity degree, carrying out normalization processing on the product to obtain a corresponding normalization value, obtaining the distance between the historical current value and the target data, and obtaining the difference characteristic value of the historical current value according to the distance and the normalization value;
Taking the product of the difference characteristic values of all the historical current values in the local set data of the target data as a data local distance selection factor at the current moment.
Further, the obtaining, according to the data local distance selection factor, a K-distance neighborhood when performing COF outlier analysis on the current predicted value at the current moment includes:
And obtaining the product between the initial K distance neighborhood and the data local distance selection factor as the K distance neighborhood when the COF outlier analysis is carried out on the current predicted value at the current moment.
Further, the obtaining the weight factor of the current predicted value at the current moment according to the current data in the K-distance neighborhood at the current moment includes:
And acquiring a COF value of the current predicted value at the current moment according to the current data in the K-distance neighborhood at the current moment, acquiring an absolute value of a difference between the COF value and a constant 1, and carrying out normalization processing, wherein the obtained normalized value is used as a weight factor of the current predicted value at the current moment.
Further, the fusing processing is performed on the current sampling value and the current prediction value at the current moment according to the weight factor to obtain a target current value for threshold judgment of the photovoltaic multifunctional breaker at the current moment, including:
And respectively acquiring weight coefficients of the current sampling value and the current predicted value at the current moment according to the weight factors, and carrying out weighted summation on the current sampling value and the current predicted value at the current moment according to the weight coefficients, wherein an obtained weighted summation result is used as a target current value for threshold judgment of the photovoltaic multifunctional breaker at the current moment.
Further, the method for acquiring current prediction time sequence data in the preset future time period including the current moment includes:
And acquiring a trained LSTM long-term and short-term memory network model, and respectively acquiring current predicted values of the current moment and each moment after the current moment by utilizing the trained LSTM long-term and short-term memory network model to acquire current predicted time sequence data in a preset future period.
Compared with the prior art, the embodiment of the invention has the beneficial effects that:
The method comprises the steps of obtaining historical current time sequence data of a photovoltaic panel direct-current output end before the current moment and current prediction time sequence data in a preset future time period including the current moment, and forming a current time sequence by the historical current time sequence data and the current prediction time sequence data; acquiring a data local distance selection factor of the current moment according to the current data difference at two sides of the current predicted value corresponding to the current moment in the current time sequence, and acquiring a K distance neighborhood when performing COF outlier factor analysis on the current predicted value of the current moment according to the data local distance selection factor; and acquiring a weight factor of the current predicted value at the current moment according to the current data in the K-distance neighborhood at the current moment, acquiring a current sampling value at the current moment, and carrying out fusion processing on the current sampling value at the current moment and the current predicted value according to the weight factor to obtain a target current value for threshold judgment of the photovoltaic multifunctional breaker at the current moment. The current value conforming to the actual situation at the current moment is obtained by collecting historical current time sequence data and current prediction time sequence data at the current moment and carrying out fusion processing on the current sampling value and the current prediction value at the current moment, namely, the current value is used for judging whether the photovoltaic multifunctional circuit breaker needs to make a judgment basis for cutting off power supply action, the capability of the photovoltaic multifunctional circuit breaker for coping with instantaneous peak current is improved, the occurrence of misoperation of the photovoltaic multifunctional circuit breaker is reduced, the anti-interference capability is increased while the original working performance is maintained, and the stability of the whole system is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the embodiments or the description of the prior art will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a method for intelligent data acquisition for a photovoltaic multifunctional breaker according to an embodiment of the present invention.
Detailed Description
Embodiments of the present disclosure are described in detail below, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to like or similar elements or elements having like or similar functions throughout. The embodiments described below by referring to the drawings are exemplary and intended for the purpose of explaining the present disclosure and are not to be construed as limiting the present disclosure.
It should be noted that the terms "first," "second," and the like in the description of the present disclosure and the above-described figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the disclosure described herein may be capable of operation in sequences other than those illustrated or described herein. The implementations described in the following exemplary examples are not representative of all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with aspects as disclosed herein.
In order to illustrate the technical scheme of the invention, the following description is made by specific examples.
The invention aims at the specific scene: in grid-connected photovoltaic power generation using a photovoltaic multifunctional breaker to participate in work, the photovoltaic multifunctional breaker needs to be added at a direct-current output end of a photovoltaic panel to ensure equipment safety, wherein the photovoltaic multifunctional breaker meets the stable and efficient working requirements in practical application in a threshold triggering mode. However, since the current output of the photovoltaic panel is greatly influenced by external environment in the operation process of the photovoltaic panel, the current of the photovoltaic panel often fluctuates and possibly exceeds the theoretical maximum value of the photovoltaic panel, the photovoltaic multifunctional circuit breaker usually takes the theoretical maximum value of the photovoltaic panel as the threshold value, and when the output current of the photovoltaic panel exceeds the threshold value due to the environmental fluctuation, the photovoltaic multifunctional circuit breaker triggers the circuit breaking action, so that the current fluctuation caused by the environmental influence can cause an instant current to exceed the threshold value set by the photovoltaic multifunctional circuit breaker, thereby causing the problem of misoperation of the photovoltaic multifunctional circuit breaker to break the circuit.
Referring to fig. 1, a method flowchart of an intelligent data collection method for a photovoltaic multifunctional circuit breaker according to an embodiment of the present invention is shown in fig. 1, where the intelligent data collection method may include:
Step S101, historical current time sequence data of a photovoltaic panel direct current output end before the current moment and current prediction time sequence data in a preset future time period including the current moment are obtained, and the historical current time sequence data and the current prediction time sequence data form a current time sequence.
In the direct current power generation process of the photovoltaic panel, current data of a direct current output end of the photovoltaic panel can be acquired in real time through a data acquisition module of the photovoltaic multifunctional circuit breaker. The sampling frequency of the current data is not limited, and an implementer can set according to a real-time scene, and preferably, the sampling frequency in the embodiment of the invention is 1 second.
Based on the above sampling manner, the historical current time sequence data collected by the photovoltaic panel direct current output end in the historical period before the current time t can be obtained, so that the current prediction time sequence data in the preset future period including the current time is obtained through the time sequence data prediction model, and preferably, in the embodiment of the invention, the length of the current prediction time sequence data is 50. The method for acquiring the current prediction time sequence data in the preset future time period including the current moment comprises the following steps: and acquiring a trained LSTM long-term and short-term memory network model, and respectively acquiring current predicted values of the current moment and each moment after the current moment by utilizing the trained LSTM long-term and short-term memory network model to acquire current predicted time sequence data in a preset future period.
It should be noted that, training the LSTM long-term memory network model according to the historical current time sequence data belongs to a known technology, and will not be described herein.
Further, the historical current time sequence data and the current prediction time sequence data are formed into a current time sequence, wherein in the current time sequence, current data before the current time belongs to a historical current value, current data after the current time and including the current time belongs to a current prediction value, and the current prediction value corresponding to the current time t in the current time sequence is recorded as。
Step S102, according to the current data difference at two sides of the current predicted value corresponding to the current moment in the current time sequence, acquiring a data local distance selection factor at the current moment, and according to the data local distance selection factor, acquiring a K distance neighborhood when performing COF outlier analysis on the current predicted value at the current moment.
In the process of data acquisition, current fluctuation conditions caused by the environment exist, and the environment state cannot be accurately adapted under different fluctuation conditions, so that in order to better avoid misjudgment of the photovoltaic multifunctional circuit breaker, a current predicted value corresponding to the current moment is requiredSum current sample value/>And carrying out numerical fusion processing to serve as a judgment basis of the photovoltaic multifunctional circuit breaker at the current moment.
Therefore, when the current predicted value corresponding to the current time t is obtainedThen, the current prediction value/>, corresponding to the current time tWhen the current prediction value/>When a stable current value change is presented in the current sequence, the reference predicted current value is more prone to a higher degree, and when the current predicted value/>, in the current sequenceWhen the local neighborhood is present as unstable current value change, the reference predicted current value tends to be lower, so that the current predicted value/>, corresponding to the current moment, corresponds toSum current sample value/>In the numerical fusion process, the current predicted value/>, is neededThe local stability of the presentation is measured and fusion weights are obtained based on the measurement.
In the embodiment of the invention, the COF outlier factor of the current predicted value is obtained to obtain the current predicted valueThe local stability degree is measured, and the current predicted value/>In the COF outlier of (2), since the COF outlier is obtained by the selection of a fixed K-distance neighborhood (K-distance neighborhood is the current predictor selected in the current timing sequence)Recent K pieces of current data), therefore, in the current time series, the current predicted value/>, at the current time instantWhen COF outlier factor is obtained, current prediction value/>The selection of the K-distance neighborhood needs to be made in its left historical current value and right current prediction value. When the historical current value on the left side and the current predicted value on the right side are in similar fluctuation states, the current predicted value/>, can be improved through selection of a smaller K distance neighborhoodOn the contrary, when the historical current value on the left side and the current predicted value on the right side are in larger fluctuation difference, the selection of a larger K distance neighborhood can ensure that more current data with higher fluctuation can be covered when the COF outlier factor is acquired, so that the acquired current predicted value/>Accuracy of COF outlier factor(s). Therefore, firstly, according to the current data difference at two sides of the current predicted value corresponding to the current moment in the current time sequence, a data local distance selection factor at the current moment is obtained, and then, according to the data local distance selection factor, a K distance neighborhood when COF outlier factor analysis is carried out on the current predicted value at the current moment is obtained.
Preferably, the obtaining the local distance selection factor of the current time according to the current data difference at two sides of the current predicted value corresponding to the current time in the current time sequence includes:
(1) And taking the current predicted value corresponding to the current moment in the current time sequence as target data, and forming a local data set of the target data by a first preset number of historical current values before the target data in the current time sequence.
Specifically, an initial K distance neighborhood when acquiring the COF outlier factor is set to 20, and then a current prediction value corresponding to the current moment is formed by a historical current value after the 20 th historical current value in the current time sequenceIs described herein). It should be noted that the present invention does not limit the initial K-distance neighborhood, and the implementer may set according to the implementation scenario.
(2) And acquiring the data local distance selection factor at the current moment according to the current data difference at two sides of each historical current value in the local data set of the target data and the distance between the current data and the target data.
Specifically, for any historical current value in the local data set, selecting a second preset number of current data to form a first subsequence on the left side of the historical current value in the current time sequence by taking the historical current value as the center, selecting the second preset number of current data to form a second subsequence on the right side of the historical current value, and obtaining the similarity between the first subsequence and the second subsequence;
acquiring all neighborhood current data in an initial K-distance neighborhood of the historical current value in the current time sequence, respectively calculating the absolute value of a difference value between a current sampling value and a current prediction value of each neighborhood current data to obtain the sum of the absolute value of the difference value, obtaining the product between the sum of the absolute value of the difference value and the similarity degree, carrying out normalization processing on the product to obtain a corresponding normalization value, obtaining the distance between the historical current value and the target data, and obtaining the difference characteristic value of the historical current value according to the distance and the normalization value;
Taking the product of the difference characteristic values of all the historical current values in the local set data of the target data as a data local distance selection factor at the current moment.
In one embodiment, the second predetermined number is set toK refers to an initial K distance neighborhood, and the calculation expression of the data local distance selection factor at the current time t is as follows:
wherein, A data local distance selection factor representing the current time t, M representing the data quantity in the local data set of the current predicted value corresponding to the current time t,/>Representing the distance between the xth historical current value in the local data set and the current predicted value corresponding to the current time t,/>Representing normalization process,/>An initial K-distance neighborhood representing the xth historical current value in the local data set,/>Current prediction value of the q-th neighborhood current data in the initial K-distance neighborhood representing the x-th historical current value,/>Current sample value of the q-th neighborhood current data in the initial K-distance neighborhood representing the x-th historical current value, |represents absolute value sign,/>Representing a dynamic time warping algorithm,/>Left/>, representing the xth historical current value in the local data setThe current data form a first subsequence,/>Right side/>, representing the xth historical current value in the local data setThe current data constitutes a second sub-sequence.
When the weather changes, the fluctuation condition of the current has continuous fluctuation increasing or decreasing conditions, namely the current is stable to fluctuation and fluctuation to stable conditions, so that in the process of evaluating the data local distance selection factor at the current time t, the continuous change of the initial K distance neighborhood evaluation is required to be evaluated, and therefore, the data local distance selection factor is acquired by a continuous multiplication mode in the formula; after obtaining a trained LSTM long-term memory network model, evaluating the prediction reliability of a local range through the difference between the current predicted value and the current sampling value in the initial K-distance neighborhood of each historical current value in the local data set of the current predicted value corresponding to the current time t, wherein the greater the prediction and actual difference is, the lower the reliability of prediction information is in the evaluation process of the data local distance selection factor of the current predicted value corresponding to the current time t, and then the two lengths of the current time t are/>, Subsequences of (first and second subsequences)And (3) carrying out distance optimization on the similarity degree of the real data and the predicted data obtained by calculating the dynamic bending distance, so as to ensure that the evaluation difference between the two subsequences is increased when the prediction reliability is low, and further, carrying out accurate optimization on the selection of the subsequent K-distance neighborhood.
Wherein between the first subsequence and the second subsequenceThe dynamic bending distance illustrates how the ripple condition of the two sub-sequences requires a reduction in the amount of data points in the subsequent calculation of the average connection distance in the ripple plateau current data, so that its characteristics can also be highlighted in the plateau ripple curve.
Preferably, the obtaining the K-distance neighborhood when performing COF outlier analysis on the current predicted value at the current moment according to the data local distance selection factor includes:
And obtaining the product between the initial K distance neighborhood and the data local distance selection factor as the K distance neighborhood when the COF outlier analysis is carried out on the current predicted value at the current moment.
In one embodiment, the local distance selection factor of the data at the current time t is obtainedThereafter, by calculating the expression/>Obtaining K distance neighborhood/>, when performing COF outlier factor analysis on current predicted value at current time t。
So far, according to the current data difference in the current time sequence, the K distance neighborhood when the current predicted value at the current time t is subjected to COF outlier factor analysis is obtained。
Step S103, obtaining a weight factor of a current predicted value at the current moment according to current data in the K-distance neighborhood at the current moment, obtaining a current sampling value at the current moment, and carrying out fusion processing on the current sampling value at the current moment and the current predicted value according to the weight factor to obtain a target current value for threshold judgment of the photovoltaic multifunctional breaker at the current moment.
K distance neighborhood when COF outlier factor analysis is carried out on current predicted value of current time tThen, a weight factor of a current predicted value at the current moment can be obtained according to current data in a K-distance neighborhood at the current moment, and the weight factor obtaining method comprises the following steps: and acquiring a COF value of the current predicted value at the current moment according to the current data in the K-distance neighborhood at the current moment, acquiring an absolute value of a difference between the COF value and a constant 1, and carrying out normalization processing, wherein the obtained normalized value is used as a weight factor of the current predicted value at the current moment.
In one embodiment, the calculation expression of the weighting factor of the current prediction value at the present time t is:
wherein, Weight factor representing current prediction value at present time t,/>The normalization function is represented as a function of the normalization,Current prediction value/>, which represents current time tAnd 1 represents a constant, || represents an absolute value symbol.
It should be noted that, because the problem to be solved by the embodiment of the invention is that the instantaneous abnormal current output in the working of the photovoltaic panel causes the photovoltaic multifunctional breaker to perform the power-off action, the invention has the following advantages thatHigher, the factor decreases/>, based on the local distance of the dataThe value, get the weight factor which keeps the detail characteristic through the calculation of the average connection distance, if appear the instantaneous current peak value at present moment t, the weight factor will bias to the predicted value, when there is fluctuation in the true sampling value or predicted value and true value deviate greatly, lead to the degree of similarity to be lower, then according to the local distance selection factor of data increase/>The value is in this case more biased towards the true value so that the current is more believing the predicted value at a steady output and the true value at a fluctuating time, without impairing the response to a sustained over-threshold current while avoiding the effects of the instantaneous fluctuating current produced by the circuit.
Furthermore, the embodiment of the invention aims to bias to the predicted value to eliminate the influence of the peak when the instantaneous peak appears in the output current of the photovoltaic panel under the steady state under the action of the weight factor, and bias to the true value when the actual frequent fluctuation appears in the output current of the photovoltaic panel, so that the photovoltaic multifunctional circuit breaker can cope with the possible overcurrent phenomenon, such as the phenomenon that equipment short circuit happens suddenly when the current is relatively steady, and the photovoltaic multifunctional circuit breaker can still accurately trigger the action because the short circuit current is much higher than the current peak value generated by noise or illumination influence, therefore, the current sampling value and the current predicted value at the current moment are fused according to the weight factor to obtain the target current value for threshold judgment of the photovoltaic multifunctional circuit breaker at the current moment.
Preferably, according to the weight factor, fusion processing is performed on the current sampling value and the current prediction value at the current moment to obtain a target current value for threshold judgment of the photovoltaic multifunctional breaker at the current moment, including:
And respectively acquiring weight coefficients of the current sampling value and the current predicted value at the current moment according to the weight factors, and carrying out weighted summation on the current sampling value and the current predicted value at the current moment according to the weight coefficients, wherein an obtained weighted summation result is used as a target current value for threshold judgment of the photovoltaic multifunctional breaker at the current moment.
In one embodiment, firstly, a current sampling value at the current time t, that is, a current value of a photovoltaic panel direct current output end is collected, and a calculation expression of a target current value for threshold judgment of the photovoltaic multifunctional breaker at the current time is as follows:
wherein, Target current value for threshold judgment of photovoltaic multifunctional breaker at current moment,/>, andWeight factor representing the current time t,/>Representing the current sample value (i.e. the actual acquired current data) at the present moment,/>The current predicted value at the present time t is indicated.
When a weight factor of the current time t is obtained, a current sampling value and a current prediction value are weighted and fused through the weight factor, so that a target current value for threshold judgment of the photovoltaic multifunctional breaker is obtained。
After the target current value at the current time t is obtained, the current value can be used as a judgment basis of the photovoltaic multifunctional circuit breaker at the current time, if the target current value is judged to be higher than a threshold value, the action of opening the circuit is triggered, and the resistance of the photovoltaic multifunctional circuit breaker to triggering by abnormal signals generated by the change of the illumination environment is enhanced while the basic working requirement of the photovoltaic multifunctional circuit breaker is not influenced. When the photovoltaic multifunctional breaker triggers the action of breaking the circuit because the current exceeds the threshold value, the photovoltaic multifunctional breaker can provide maintenance personnel with current data outlier information including current data fluctuation conditions and when the photovoltaic multifunctional breaker is triggered to act, and the current data outlier information is used for judging whether frequent current fluctuation is caused by unstable equipment or environment or over-current problem is caused by equipment short circuit, so that the maintenance personnel can conveniently adjust equipment and parameters in time.
In summary, the embodiment of the invention obtains the historical current time sequence data of the photovoltaic panel direct current output end before the current moment and the current prediction time sequence data in the preset future time period including the current moment, and forms the current time sequence by the historical current time sequence data and the current prediction time sequence data; obtaining a data local distance selection factor at the current moment according to current data differences at two sides of a current predicted value corresponding to the current moment in a current time sequence, and obtaining a K distance neighborhood when performing COF outlier factor analysis on the current predicted value at the current moment according to the data local distance selection factor; and acquiring a weight factor of a current predicted value at the current moment according to the current data in the K-distance neighborhood at the current moment, acquiring a current sampling value at the current moment, and carrying out fusion processing on the current sampling value at the current moment and the current predicted value according to the weight factor to obtain a target current value for threshold judgment of the photovoltaic multifunctional breaker at the current moment. The current value conforming to the actual situation at the current moment is obtained by collecting historical current time sequence data and current prediction time sequence data at the current moment and carrying out fusion processing on the current sampling value and the current prediction value at the current moment, namely, the current value is used for judging whether the photovoltaic multifunctional circuit breaker needs to make a judgment basis for cutting off power supply action, the capability of the photovoltaic multifunctional circuit breaker for coping with instantaneous peak current is improved, the occurrence of misoperation of the photovoltaic multifunctional circuit breaker is reduced, the anti-interference capability is increased while the original working performance is maintained, and the stability of the whole system is improved.
The above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention, and are intended to be included in the scope of the present invention.
Claims (3)
1. An intelligent data acquisition method for a photovoltaic multifunctional circuit breaker is characterized by comprising the following steps:
acquiring historical current time sequence data of a photovoltaic panel direct current output end before the current moment and current prediction time sequence data in a preset future time period including the current moment, and forming a current time sequence by the historical current time sequence data and the current prediction time sequence data;
Acquiring a data local distance selection factor of the current moment according to the current data difference at two sides of the current predicted value corresponding to the current moment in the current time sequence, and acquiring a K distance neighborhood when performing COF outlier factor analysis on the current predicted value of the current moment according to the data local distance selection factor;
Acquiring a weight factor of a current predicted value at the current moment according to current data in the K-distance neighborhood at the current moment, acquiring a current sampling value at the current moment, and carrying out fusion processing on the current sampling value at the current moment and the current predicted value according to the weight factor to obtain a target current value for threshold judgment of the photovoltaic multifunctional breaker at the current moment;
The obtaining the data local distance selection factor of the current moment according to the current data difference at two sides of the current predicted value corresponding to the current moment in the current time sequence comprises the following steps:
Taking a current predicted value corresponding to the current moment in the current time sequence as target data, and forming a first preset number of historical current values before the target data into a local data set of the target data in the current time sequence;
Acquiring a data local distance selection factor at the current moment according to the current data difference at two sides of each historical current value in the local data set of the target data and the distance between the current data and the target data;
The obtaining the data local distance selection factor at the current moment according to the current data difference at two sides of each historical current value in the local data set of the target data and the distance between the current data and the target data comprises the following steps:
Selecting a second preset number of current data to form a first subsequence on the left side of the historical current value, and selecting a second preset number of current data to form a second subsequence on the right side of the historical current value, wherein the historical current value is taken as a center in the current time sequence, so as to obtain the similarity between the first subsequence and the second subsequence;
acquiring all neighborhood current data in an initial K-distance neighborhood of the historical current value in the current time sequence, respectively calculating the absolute value of a difference value between a current sampling value and a current prediction value of each neighborhood current data to obtain the sum of the absolute value of the difference value, obtaining the product between the sum of the absolute value of the difference value and the similarity degree, carrying out normalization processing on the product to obtain a corresponding normalization value, obtaining the distance between the historical current value and the target data, and obtaining the difference characteristic value of the historical current value according to the distance and the normalization value;
taking the product of the difference characteristic values of all the historical current values in the local set data of the target data as a data local distance selection factor at the current moment;
The obtaining the K-distance neighborhood when performing COF outlier analysis on the current predicted value at the current moment according to the data local distance selection factor includes:
obtaining a product between the initial K distance neighborhood and the data local distance selection factor as a K distance neighborhood when performing COF outlier analysis on the current predicted value at the current moment;
the fusion processing is carried out on the current sampling value and the current prediction value at the current moment according to the weight factors to obtain a target current value for threshold judgment of the photovoltaic multifunctional breaker at the current moment, and the fusion processing comprises the following steps:
And respectively acquiring weight coefficients of the current sampling value and the current predicted value at the current moment according to the weight factors, and carrying out weighted summation on the current sampling value and the current predicted value at the current moment according to the weight coefficients, wherein an obtained weighted summation result is used as a target current value for threshold judgment of the photovoltaic multifunctional breaker at the current moment.
2. The intelligent data acquisition method for a photovoltaic multifunctional circuit breaker according to claim 1, wherein the acquiring the weight factor of the current predicted value at the current moment according to the current data in the K-distance neighborhood at the current moment comprises:
And acquiring a COF value of the current predicted value at the current moment according to the current data in the K-distance neighborhood at the current moment, acquiring an absolute value of a difference between the COF value and a constant 1, and carrying out normalization processing, wherein the obtained normalized value is used as a weight factor of the current predicted value at the current moment.
3. The intelligent data acquisition method for a photovoltaic multi-function circuit breaker according to claim 1, wherein the acquisition method of current prediction time series data in a preset future period including the current moment comprises the following steps:
And acquiring a trained LSTM long-term and short-term memory network model, and respectively acquiring current predicted values of the current moment and each moment after the current moment by utilizing the trained LSTM long-term and short-term memory network model to acquire current predicted time sequence data in a preset future period.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410114972.7A CN117648554B (en) | 2024-01-29 | 2024-01-29 | Intelligent data acquisition method for photovoltaic multifunctional circuit breaker |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410114972.7A CN117648554B (en) | 2024-01-29 | 2024-01-29 | Intelligent data acquisition method for photovoltaic multifunctional circuit breaker |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117648554A CN117648554A (en) | 2024-03-05 |
| CN117648554B true CN117648554B (en) | 2024-04-26 |
Family
ID=90049849
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410114972.7A Active CN117648554B (en) | 2024-01-29 | 2024-01-29 | Intelligent data acquisition method for photovoltaic multifunctional circuit breaker |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117648554B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117970182A (en) * | 2024-03-28 | 2024-05-03 | 国网山东省电力公司曲阜市供电公司 | Electric leakage early warning method and system based on DTW algorithm |
| CN118245957B (en) * | 2024-05-28 | 2024-07-26 | 山东德源电力科技股份有限公司 | Electric energy meter system based on cloud edge cooperation |
| CN118539378B (en) * | 2024-07-26 | 2024-10-08 | 浙江亿腾电气科技有限公司 | Intelligent circuit breaker, method and control equipment for solar photovoltaic power station |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102799616A (en) * | 2012-06-14 | 2012-11-28 | 北京大学 | Outlier point detection method in large-scale social network |
| CN115933787A (en) * | 2023-03-14 | 2023-04-07 | 西安英图克环境科技有限公司 | Indoor multi-terminal intelligent control system based on indoor environment monitoring |
| CN116013087A (en) * | 2023-03-27 | 2023-04-25 | 东莞市城建规划设计院 | Traffic flow statistical method based on urban moving vehicle detection |
| CN116609651A (en) * | 2023-07-20 | 2023-08-18 | 山东德源电力科技股份有限公司 | Detection method of multifunctional circuit breaker for photovoltaic |
| CN116794385A (en) * | 2023-08-21 | 2023-09-22 | 山东德源电力科技股份有限公司 | High-voltage current monitoring method based on multidimensional data analysis |
| CN116845878A (en) * | 2023-07-10 | 2023-10-03 | 杭州齐智能源科技股份有限公司 | Electric load prediction method for micro-grid |
-
2024
- 2024-01-29 CN CN202410114972.7A patent/CN117648554B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102799616A (en) * | 2012-06-14 | 2012-11-28 | 北京大学 | Outlier point detection method in large-scale social network |
| CN115933787A (en) * | 2023-03-14 | 2023-04-07 | 西安英图克环境科技有限公司 | Indoor multi-terminal intelligent control system based on indoor environment monitoring |
| CN116013087A (en) * | 2023-03-27 | 2023-04-25 | 东莞市城建规划设计院 | Traffic flow statistical method based on urban moving vehicle detection |
| CN116845878A (en) * | 2023-07-10 | 2023-10-03 | 杭州齐智能源科技股份有限公司 | Electric load prediction method for micro-grid |
| CN116609651A (en) * | 2023-07-20 | 2023-08-18 | 山东德源电力科技股份有限公司 | Detection method of multifunctional circuit breaker for photovoltaic |
| CN116794385A (en) * | 2023-08-21 | 2023-09-22 | 山东德源电力科技股份有限公司 | High-voltage current monitoring method based on multidimensional data analysis |
Non-Patent Citations (8)
| Title |
|---|
| Outlier detection algorithm based on k-nearest neighbors-local outlier factor;He Xu et al.;《Journal of Algorithms & Computational Technology》;20220315;第16卷;第1-12页 * |
| Outlier Detection for Multidimensional Time Series Using Deep Neural Networks;Tung Kieu et al.;《2018 19th IEEE International Conference on Mobile Data Management (MDM)》;20180716;第125-134页 * |
| 基于LSTM神经网络的光伏系统功率预测;朱润泽等;《电力科技与环保》;20230615;第39卷(第03期);第201-206页 * |
| 基于SVR-UKF的光伏电站功率预测;俞娜燕等;《自动化与仪器仪表》;20200425(第04期);第73-77页 * |
| 基于改进VMD-XGBoost-BILSTM组合模型的光伏发电异常检测方法;赵博超等;《计算机工程》;20230626;第1-13页 * |
| 智能物联断路器在智慧台区建设中的应用;梁波等;《大众用电》;20201105;第35卷(第11期);第23-26页 * |
| 离群点检测算法综述;孔翎超等;《计算机科学》;20240109;第1-17页 * |
| 融合孤立森林和局部离群因子的离群点检测方法;凌莉等;《计算机应用与软件》;20221231;第第39卷卷(第第12期期);第278-283页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117648554A (en) | 2024-03-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117648554B (en) | Intelligent data acquisition method for photovoltaic multifunctional circuit breaker | |
| Larsson | Flicker emission of wind turbines during continuous operation | |
| JP5600570B2 (en) | Photovoltaic power generation prediction device, solar power generation amount prediction method, system voltage control device, system voltage control method | |
| JP7289995B2 (en) | Method and apparatus for recognizing operating state of photovoltaic string and storage medium | |
| CN112653393B (en) | Control method and device for photovoltaic system IV diagnosis | |
| CN117332215B (en) | High-low voltage power distribution cabinet abnormal fault information remote monitoring system | |
| CN117056813B (en) | Detection method for intelligent circuit breaker data acquisition | |
| CN109286201A (en) | A kind of duty ratio control method based on power feedforward mode | |
| CN111894814A (en) | Fault processing system and method for power generation system | |
| CN118432000B (en) | High-voltage isolation protection method and system | |
| Iqbal et al. | Real-time event detection based on Weibull distribution using synchrophasor measurements for enhanced situational awareness | |
| KR20190037657A (en) | Method and apparatus for measuring degradation property of the solar cell module | |
| CN111864709A (en) | Overload protection method, control method and device and wind power converter | |
| CN117117780A (en) | Circuit breaker anti-blocking method and system based on secondary information fusion of transformer substation | |
| CN115144745B (en) | Detection system is used in column switch production | |
| CN116048182A (en) | Photovoltaic module flexible power point tracking method, device, equipment, medium and system | |
| CN115614112A (en) | Electric power control method and system for steam turbine generator of nuclear power plant | |
| CN108227818B (en) | Self-adaptive step size photovoltaic maximum power tracking method and system based on conductance increment | |
| CN118157332B (en) | Power supply equipment operation data acquisition and analysis method based on wireless communication technology | |
| CN117833824B (en) | Performance analysis method, device and equipment of photovoltaic inverter and storage medium | |
| CN118539378B (en) | Intelligent circuit breaker, method and control equipment for solar photovoltaic power station | |
| CN117519418B (en) | Maximum power point tracking method executed by direct current electronic load | |
| CN115825664A (en) | Linear load electrical anomaly detection method based on convolution model | |
| Fang et al. | Research on Overcurrent Setting Protection of Distribution Network Based on Adaptive Principle | |
| KR20240111576A (en) | Apparatus and method for evaluating the risk of failure due to arc |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |