CN117978081A - Photovoltaic array self-cleaning system - Google Patents
Photovoltaic array self-cleaning system Download PDFInfo
- Publication number
- CN117978081A CN117978081A CN202410381590.0A CN202410381590A CN117978081A CN 117978081 A CN117978081 A CN 117978081A CN 202410381590 A CN202410381590 A CN 202410381590A CN 117978081 A CN117978081 A CN 117978081A
- Authority
- CN
- China
- Prior art keywords
- photovoltaic array
- cleaning
- photoelectric conversion
- data set
- photovoltaic
- 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.)
- Granted
Links
- 238000004140 cleaning Methods 0.000 title claims abstract description 251
- 238000006243 chemical reaction Methods 0.000 claims abstract description 101
- 230000005856 abnormality Effects 0.000 claims abstract description 51
- 230000003287 optical effect Effects 0.000 claims abstract description 32
- 230000017525 heat dissipation Effects 0.000 claims abstract description 27
- 238000004458 analytical method Methods 0.000 claims abstract description 19
- 230000002159 abnormal effect Effects 0.000 claims description 41
- 238000000034 method Methods 0.000 claims description 34
- 238000003491 array Methods 0.000 claims description 28
- 230000003749 cleanliness Effects 0.000 claims description 12
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 11
- 230000001174 ascending effect Effects 0.000 claims description 9
- 238000007621 cluster analysis Methods 0.000 claims description 9
- 238000010219 correlation analysis Methods 0.000 claims description 8
- 238000005286 illumination Methods 0.000 claims description 8
- 230000008569 process Effects 0.000 claims description 6
- 230000005855 radiation Effects 0.000 claims description 5
- 238000013507 mapping Methods 0.000 claims description 4
- 230000008602 contraction Effects 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 description 5
- 230000007613 environmental effect Effects 0.000 description 5
- 238000012545 processing Methods 0.000 description 3
- 239000000428 dust Substances 0.000 description 2
- 210000003608 fece Anatomy 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 238000002759 z-score normalization Methods 0.000 description 2
- 230000002411 adverse Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000012459 cleaning agent Substances 0.000 description 1
- 238000011109 contamination Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000031700 light absorption Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
- 230000003313 weakening effect Effects 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/50—Photovoltaic [PV] energy
Landscapes
- Photovoltaic Devices (AREA)
Abstract
The invention relates to the technical field of photovoltaic array cleaning, and provides an automatic photovoltaic array cleaning system, which comprises the following components: acquiring a time sequence of related parameters of the photovoltaic array; acquiring a photoelectric conversion quantity sequence according to a time sequence of related parameters of the photovoltaic array, acquiring a cleaning degree index according to the photoelectric conversion quantity sequence, acquiring a heat dissipation data set according to a time sequence of the temperature of the optical component, acquiring a cleaning abnormality index according to the cleaning degree index and the heat dissipation data set, acquiring a shrinkage factor according to the cleaning abnormality index, acquiring a clustering result of the cleaning abnormality data set based on the shrinkage factor by using a CURE hierarchical clustering algorithm, and acquiring a clustering analysis result of the photovoltaic array according to the clustering result of the cleaning abnormality data set; and classifying and cleaning the photovoltaic array according to the clustering analysis result of the photovoltaic array. According to the invention, the accuracy of the clustering analysis result of the photovoltaic array is improved, and the automatic cleaning efficiency of the photovoltaic array is improved.
Description
Technical Field
The invention relates to the technical field of photovoltaic array cleaning, in particular to an automatic cleaning system for a photovoltaic array.
Background
A photovoltaic array is a system of a plurality of photovoltaic modules (photovoltaic cells) that are typically mounted on the roof, floor or other suitable space of a building to maximize the reception of solar radiation. Photovoltaic arrays have wide application in the renewable energy field and have become an important component of global energy conversion.
Covers such as dirt, dust, bird droppings and the like on the surface of the photovoltaic array can have certain influence on the performance of the photovoltaic system, including weakening of light absorption, increase of temperature and increase of vulnerability, and the adverse effects can seriously reduce the efficiency of photovoltaic power generation and reduce the utilization rate of solar energy and the photovoltaic system. Therefore, periodic cleaning and maintenance of the surface of the photovoltaic array is critical. At present, how to provide an automatic cleaning system for a photovoltaic array to solve the problem of low cleaning efficiency of the photovoltaic array is an important issue.
Disclosure of Invention
The invention provides an automatic cleaning system for a photovoltaic array, which aims to solve the problem of lower cleaning efficiency of the photovoltaic array, and adopts the following technical scheme:
One embodiment of the invention provides a photovoltaic array automatic cleaning system, comprising the following modules:
The data acquisition module is used for acquiring a time sequence of related parameters of each photovoltaic array, wherein the related parameters comprise current, voltage, optical component temperature, ambient illumination intensity and ambient wind power;
The cluster analysis module is used for acquiring a photoelectric conversion quantity sequence of the photoelectric conversion quantity of each photovoltaic array according to the time sequence of the related parameters of each photovoltaic array and acquiring each gray correlation degree of the photoelectric conversion quantity of each photovoltaic array according to the photoelectric conversion quantity sequence of the photoelectric conversion quantity of each photovoltaic array; acquiring a cleaning degree index of each photovoltaic array according to each gray correlation degree of the photoelectric conversion quantity of each photovoltaic array; acquiring a heat dissipation data set of each photovoltaic array according to the time sequence of the temperature of the optical component of each photovoltaic array; acquiring a cleaning abnormality index of each photovoltaic array according to the heat radiation data set and the cleaning degree index of each photovoltaic array; acquiring a cleaning abnormal data set according to the cleaning abnormal index of each photovoltaic array, and acquiring the shrinkage coefficient of the cleaning abnormal data set according to the cleaning abnormal data set; acquiring a clustering result of the cleaning abnormal data set based on the shrinkage coefficient by using a CURE hierarchical clustering algorithm; obtaining a clustering analysis result of the photovoltaic array according to the clustering result of the cleaning abnormal data set;
And the cleaning decision module is used for classifying and cleaning the photovoltaic array according to the clustering analysis result of the photovoltaic array.
Preferably, the method for obtaining the gray correlation degree of the photoelectric conversion amount of each photovoltaic array according to the time sequence of the related parameters of each photovoltaic array and the photoelectric conversion amount sequence of the photoelectric conversion amount of each photovoltaic array comprises the following steps:
For each photovoltaic array, taking the product of the current value and the voltage value at each acquisition time of the photovoltaic array as the photoelectric conversion quantity at each acquisition time of the photovoltaic array, and taking a sequence formed by the photoelectric conversion quantities at all the acquisition times of the photovoltaic array according to the ascending order of time as a photoelectric conversion quantity sequence of the photoelectric conversion quantity of the photovoltaic array;
The method comprises the steps of taking the ambient temperature, the ambient illumination intensity and the ambient wind power of a photovoltaic array as each correlation factor of the photoelectric conversion amount of the photovoltaic array, taking a time sequence of all correlation factors of the photoelectric conversion amount of the photovoltaic array and the photoelectric conversion amount of the photovoltaic array as input of a gray correlation analysis algorithm, taking output of the gray correlation analysis algorithm as the correlation degree between the photoelectric conversion amount of the photovoltaic array and each correlation factor of the photoelectric conversion amount of the photovoltaic array, and taking the correlation degree between the photoelectric conversion amount of the photovoltaic array and each correlation factor of the photoelectric conversion amount of the photovoltaic array as each gray correlation degree of the photoelectric conversion amount of the photovoltaic array.
Preferably, the method for obtaining the cleanliness index of each photovoltaic array according to each gray correlation degree of the photoelectric conversion amount of each photovoltaic array comprises the following steps:
Acquiring the correlation intensity of the photoelectric conversion quantity of each photovoltaic array according to the time sequence of the photoelectric conversion quantity of each photovoltaic array and the correlation factor of the photoelectric conversion quantity;
For each photovoltaic array, taking a natural constant as a base number, taking a negative mapping result taking the correlation intensity of the photoelectric conversion quantity of the photovoltaic array as an index as a molecule, calculating the accumulated sum of each gray correlation degree of the photoelectric conversion quantity of the photovoltaic array on all correlation factors, taking the sum of the accumulated sum and a first preset parameter as a denominator, and taking the ratio of the molecule to the denominator as a first characteristic ratio;
And taking the sum of the opposite number of the first characteristic ratio and the first preset parameter as a cleanliness index of the photovoltaic array.
Preferably, the method for obtaining the correlation strength of the photoelectric conversion amount of each photovoltaic array according to the time sequence of the correlation factor of the photoelectric conversion amount and the photoelectric conversion amount of each photovoltaic array comprises the following steps:
For each photovoltaic array, calculating a pearson correlation coefficient between a photoelectric conversion amount sequence of the photoelectric conversion amount of the photovoltaic array and a time sequence of each correlation factor of the photoelectric conversion amount, and taking the average value of the pearson correlation coefficient on all the correlation factors as the correlation intensity of the photovoltaic array.
Preferably, the method for acquiring the heat dissipation data set of each photovoltaic array according to the time sequence of the temperature of the optical component of each photovoltaic array comprises the following steps:
And for the time sequence of the optical component temperature of each photovoltaic array, calculating the difference value between the observed values of two adjacent moments in the time sequence of the optical component temperature, taking the sequence formed by the difference values according to the sequence of time ascending as the differential sequence of the time sequence of the optical component temperature, and taking the set formed by all data with negative values in the differential sequence of the time sequence of the optical component temperature as the heat dissipation data set of the photovoltaic array.
Preferably, the method for obtaining the cleaning abnormality index of each photovoltaic array according to the heat dissipation data set and the cleaning degree index of each photovoltaic array comprises the following steps:
in the method, in the process of the invention, Clean anomaly index representing the xth photovoltaic array,/>Representing the average of all data in the optical component temperature time series of the xth photovoltaic array,/>Representing the mean of all data in the ambient temperature time series of the xth photovoltaic array,/>Index of cleanliness representing the xth photovoltaic array,/>Representing error parameters,/>Representing the number of data in the heat dissipation data set of the xth photovoltaic array,/>Representing an exponential function based on natural constants,/>The value of the g data in the heat dissipation data set of the x photovoltaic array is represented.
Preferably, the method for obtaining the cleaning abnormal data set according to the cleaning abnormal index of each photovoltaic array and obtaining the shrinkage coefficient of the cleaning abnormal data set according to the cleaning abnormal data set comprises the following steps:
Taking a set formed by cleaning abnormality indexes of all the photovoltaic arrays as a cleaning abnormality data set, taking the cleaning abnormality data set as the input of a DPC density peak clustering algorithm, and taking the output of the DPC density peak clustering algorithm as the local density of each data point in the cleaning abnormality data set;
Taking the average value of the local densities of all data points in the cleaning abnormal data set as the tightness index of the cleaning abnormal data set, and taking the natural constant as a base and taking the negative mapping result of the tightness index of the cleaning abnormal data set as an index as the contraction coefficient of the cleaning abnormal data set.
Preferably, the specific method for acquiring the clustering result of the clean abnormal data set based on the shrinkage factor by using the CURE hierarchical clustering algorithm is as follows:
Taking all cleaning abnormality indexes in the cleaning abnormality data set as input of a CURE hierarchical clustering algorithm, taking the shrinkage factor of the cleaning abnormality data set as the shrinkage factor of the CURE hierarchical clustering algorithm, and taking output of the CURE hierarchical clustering algorithm as a clustering result of the cleaning abnormality data set.
Preferably, the method for obtaining the clustering analysis result of the photovoltaic array according to the clustering result of the cleaning abnormal data set comprises the following steps:
Calculating the average value of all cleaning abnormality indexes of each cluster in the clustering result of the cleaning abnormality data set, taking a sequence formed by the average values according to the ascending order of numerical values as a cleaning grade sequence, and taking the cluster corresponding to each data in the cleaning grade sequence as the cluster of each cleaning grade;
and each cleaning abnormality index in the cluster of each cleaning grade corresponds to each photovoltaic array, and the cluster of all the cleaning grades is used as a cluster analysis result of the photovoltaic array.
Preferably, the method for classifying and cleaning the photovoltaic array according to the clustering analysis result of the photovoltaic array comprises the following steps:
And for each cleaning-level cluster in the cluster analysis results of the photovoltaic arrays, cleaning the corresponding photovoltaic arrays in the clusters by using a cleaning execution device to acquire the cleaning results of the photovoltaic arrays by adopting the same-level cleaning parameters, wherein the same-level cleaning parameters comprise the same-level water pressure and the same-level cleaning time.
The beneficial effects of the invention are as follows: according to the method, a photoelectric conversion quantity sequence and gray correlation degree are obtained according to a time sequence of related parameters of the photovoltaic array, a cleaning degree index is obtained according to the photoelectric conversion quantity sequence and the gray correlation degree, a cleaning abnormality index is obtained according to the cleaning degree index in combination with abnormal characteristics which occur when the surface of the photovoltaic array is poor in cleaning degree, a cleaning abnormality data set is constructed according to the cleaning abnormality index, a shrinkage factor in a CURE hierarchical clustering algorithm is improved according to the data compactness in the cleaning abnormality data set, and a clustering analysis result of the photovoltaic array is obtained. The method has the advantages that by improving the contraction factor in the CURE hierarchical clustering algorithm, the accuracy of the clustering analysis result of the photovoltaic array is improved, the automatic cleaning efficiency of the photovoltaic array is improved, and a large amount of waste of clean energy is avoided.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only some embodiments of the invention, and that other drawings can be obtained according to these drawings without inventive faculty for a person skilled in the art.
FIG. 1 is a schematic flow chart of a photovoltaic array automatic cleaning system according to an embodiment of the present invention;
Fig. 2 is a flowchart of an embodiment of a photovoltaic array automatic cleaning system according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, a flowchart of a photovoltaic array automatic cleaning system according to an embodiment of the present invention is shown, where the system includes a data acquisition module, a cluster analysis module, and a cleaning decision module.
And the data acquisition module is used for acquiring a time sequence of relevant parameters of each photovoltaic array.
In order to achieve the aim of sorting and cleaning different photovoltaic arrays, the cleaning degree of each photovoltaic array needs to be analyzed. The method comprises the steps of collecting relevant parameter data of each photovoltaic array by using various sensors, wherein the various sensors comprise a current sensor, a voltage sensor, a temperature sensor, an illumination intensity sensor and a wind sensor, and the relevant parameters of each photovoltaic array comprise current and voltage of each photovoltaic array, and the temperature of an optical component of each photovoltaic array and the environmental temperature, the environmental illumination intensity and the environmental wind force in the environment nearby each photovoltaic array. The data acquisition time interval is 5s, the acquisition times are 500, and the operator can select according to actual conditions. In order to avoid the influence of analysis results among different dimension lines, a sequence formed by each relevant parameter data of each photovoltaic array according to the sequence of time ascending is used as a time sequence of each relevant parameter of each photovoltaic array, and Z-score normalization is carried out on all data in the time sequence of each relevant parameter to obtain the time sequence of each relevant parameter of each photovoltaic array after normalization processing, wherein the Z-score normalization is a known technology and redundant description is omitted.
So far, the time sequence of each relevant parameter of each photovoltaic array after normalization processing is obtained.
The cluster analysis module is used for acquiring a photoelectric conversion quantity sequence and gray correlation degree according to a time sequence of related parameters of the photovoltaic array, acquiring a cleaning degree index according to the photoelectric conversion quantity sequence and the gray correlation degree, acquiring a heat dissipation data set according to a time sequence of the temperature of the optical component, acquiring a cleaning abnormality index according to the cleaning degree index and the heat dissipation data set, acquiring a cleaning abnormality data set and a shrinkage factor according to the cleaning abnormality index, acquiring a clustering result of the cleaning abnormality data set based on the shrinkage factor by using a CURE hierarchical clustering algorithm, and acquiring a clustering analysis result of the photovoltaic array according to the clustering result of the cleaning abnormality data set.
As the environmental impact of different photovoltaic arrays is different, i.e. the degree of cleanliness of different photovoltaic arrays is different. The greater the possibility of abnormal phenomenon of the photovoltaic array with poor cleaning degree, in order to analyze the photovoltaic array with poor cleaning degree, and further to process the photovoltaic arrays with different cleaning degrees differently, the abnormal phenomenon of the photovoltaic array needs to be analyzed first. A flow chart of an embodiment of the present invention is shown in fig. 2.
For each photovoltaic array, if the surface cleaning degree of the photovoltaic array is good, a strong correlation exists between the photoelectric conversion amount of the photovoltaic array and the environment acquisition parameters of the photovoltaic array. However, when the surface of the photovoltaic array is less clean, the relationship between the amount of photoelectric conversion of the photovoltaic array and the environmental collection parameters of the photovoltaic array is broken.
Specifically, for each photovoltaic array, taking the product of the voltage value and the current value at each collection time of the photovoltaic array as the photoelectric conversion quantity at each collection time of the photovoltaic array, taking a sequence formed by the photoelectric conversion quantities at all collection times of the photovoltaic array according to the ascending order of time as a photoelectric conversion quantity sequence of the photoelectric conversion quantity of the photovoltaic array, and taking the ambient temperature, the ambient illumination intensity and the ambient wind power of the photovoltaic array as correlation factors of the photoelectric conversion quantity respectively.
Further, for each photovoltaic array, using a gray correlation analysis algorithm, taking a photoelectric conversion sequence of a photoelectric conversion of the photovoltaic array and a time sequence of all correlation factors of the photoelectric conversion as inputs of the gray correlation analysis algorithm, wherein the time sequence of all correlation factors of the photoelectric conversion comprises an environment temperature time sequence, an environment illumination intensity time sequence and an environment wind power time sequence, taking output of the gray correlation analysis algorithm as a correlation degree between the photoelectric conversion and each correlation factor of the photoelectric conversion, and taking a correlation degree between the photoelectric conversion and each correlation factor of the photoelectric conversion as each gray correlation degree of the photoelectric conversion of the photovoltaic array, wherein the gray correlation analysis algorithm is a known technology and is not redundant.
The cleanliness index for each photovoltaic array was calculated:
in the method, in the process of the invention, Representing the correlation intensity of the x-th photovoltaic array,/>Number of correlation factors representing photoelectric conversion amount of the x-th photovoltaic array,/>, and method for producing the sameRepresenting pearson correlation coefficient function,/>Photoelectric conversion amount sequence representing photoelectric conversion amount of the x-th photovoltaic array,/>Time series of j-th correlation factor representing photoelectric conversion amount of x-th photovoltaic array,/>Index of cleanliness representing the xth photovoltaic array,/>Representing an exponential function based on natural constants,/>The ith gray correlation degree of the photoelectric conversion quantity of the xth photovoltaic array is shown.
Absolute value of pearson correlation coefficient between sequence of photoelectric conversion amounts of the x-th photovoltaic array and time sequence of j-th correlation factor of photoelectric conversion amounts of the x-th photovoltaic arrayThe larger the correlation factor between the photoelectric conversion amount and the photoelectric conversion amount is, the stronger the correlation is, and the better the cleaning degree of the surface of the photovoltaic array is reflected to a certain extent, namely the stronger the correlation is not broken, the larger the cleaning degree index of the photovoltaic array is. Meanwhile, the ith gray correlation degree/>, of photoelectric conversion quantity of the xth photovoltaic arrayThe larger the change with time, the larger the correlation factor influences the photoelectric conversion amount, and the better the cleaning degree of the surface of the photovoltaic array is reflected to a certain extent, namely the stronger the correlation is not broken, the larger the cleaning degree index of the photovoltaic array is.
Further, in general, when the surface of the photovoltaic array is less clean, the heat dissipation capability of the photovoltaic array is poor, and the temperature of the optical components in the optical array may be significantly higher than the ambient temperature. To make the cleaning of the photovoltaic array surfaces clearer, further analysis is required based on the time series of the optical component temperature and the time series of the ambient temperature for each photovoltaic array. In order to improve the comparison result of the optical component temperature and the ambient temperature, so that the surface cleaning information of the photovoltaic array is clearer, further processing needs to be carried out on the time sequence of the optical component temperature and the time sequence of the ambient temperature.
Specifically, for each photovoltaic array, a differential operation is performed on the time sequence of the temperatures of the optical components of the photovoltaic array, that is, the observed value of each current moment is subtracted from the observed value of the previous moment to obtain the differential sequence of the time sequence of the temperatures of the optical components, and the differential operation is a known technology and is not redundant. And for each photovoltaic array, taking a set consisting of all data with negative values in a time sequence differential sequence of the temperature of the optical component of the photovoltaic array as a heat dissipation data set.
The data in the heat dissipation data set reflects the heat dissipation capacity of the photovoltaic array to a certain extent, when the surface cleaning degree of the photovoltaic array is poor, the heat dissipation capacity of the photovoltaic array is poor, and the numerical value in the heat dissipation data set is larger; when the surface cleaning degree of the photovoltaic array is good, the heat radiation capacity of the photovoltaic array is good, and the numerical value in the heat radiation data set is small.
Further, for each photovoltaic array, the average value of all data in the time series of the optical component temperature and the time series of the ambient temperature of the photovoltaic array is obtained respectively.
The cleaning abnormality index for each photovoltaic array was calculated:
in the method, in the process of the invention, Clean anomaly index representing the xth photovoltaic array,/>Mean value of all data in time series representing optical component temperature of xth photovoltaic array,/>Mean value of all data in time series representing ambient temperature of x-th photovoltaic array,/>Index of cleanliness representing the xth photovoltaic array,/>Representing error parameters, avoiding denominator values of 0, and empirical values of 0.1,/>, for error parametersRepresenting the number of data in the heat dissipation data set of the xth photovoltaic array,/>Representing an exponential function based on natural constants,/>The value of the g data in the heat dissipation data set of the x photovoltaic array is represented.
Differences between the mean of all data in the optical assembly temperature time series of the x-th photovoltaic array and the mean of all data in the time series of the ambient temperature of the x-th photovoltaic arrayThe greater reflects the greater the extent to which the optical component temperature is above ambient temperature, and the x-th photovoltaic array has a cleanliness index/>The smaller the surface cleaning degree of the photovoltaic array, the worse the surface cleaning degree, the larger the cleaning abnormality index. Meanwhile, the numerical value of the g data in the radiating data set of the x photovoltaic array/>The greater the degree of heat dissipation, i.e., the degree of value approaching 0, the smaller the degree of heat dissipation, i.e., the poorer the heat dissipation capacity, of the optical component, which reflects to some extent the poorer the surface cleaning degree of the photovoltaic array, the greater the cleaning abnormality index.
The cleaning abnormality index reflects the cleaning degree of the surface of the photovoltaic array, and the greater the cleaning abnormality index, the greater the cleaning abnormality degree of the surface of the photovoltaic array, namely the worse the cleaning degree; the smaller the cleaning anomaly index, the smaller the anomaly degree of cleaning of the photovoltaic array surface, i.e., the better the cleaning degree.
Further, in order to realize the identification of the cleaning degree of the photovoltaic arrays, all the photovoltaic arrays are divided into different categories, and different cleaning strategies are carried out on the photovoltaic arrays in the different categories, so that the cleaning efficiency of the photovoltaic arrays is improved.
Specifically, a set of cleaning abnormality indexes of all photovoltaic arrays is used as a cleaning abnormality data set, a density peak clustering algorithm (DENSITY PEAK clustering, DPC) is utilized, the cleaning abnormality data set is used as an input of the DPC density peak clustering algorithm, an output of the DPC density peak clustering algorithm is used as a local density of each data point in the cleaning abnormality data set, and the DPC density peak clustering algorithm is a known technology and is not redundant.
In general, the cleaning abnormal data set is mainly divided into two major categories, one is a larger category of the included data values, and reflects the poor cleaning degree of the photovoltaic array surface represented by the category; the other is a smaller class of data values contained, reflecting the better degree of surface cleaning of the photovoltaic array represented by the class. However, due to various abnormal conditions, the cleaning abnormal data set often contains a plurality of different categories, such as different abnormal conditions of dirt, dust and bird droppings on the surface of the photovoltaic array. In order to more accurately identify the cleanliness of the photovoltaic array, it is necessary to analyze how closely the data points are in the cleaning anomaly data set.
Further, the mean of the local densities of all data points in the cleaning anomaly data set is taken as the tightness index of the cleaning anomaly data set.
Calculating shrinkage coefficients of the cleaning anomaly data set:
in the method, in the process of the invention, Representing the shrinkage factor of a clean anomaly data set,/>Representing an exponential function based on natural constants,/>A tightness index representing the cleaning anomaly data set.
Tightness index for cleaning abnormal data setsThe larger the data in the clean abnormal data set, the more compact the data, the more the shrinkage coefficient approaches 0; conversely, the closeness index/>, of the cleaning anomaly data setThe smaller the data in the clean anomaly data set, the less tight the data, the more the shrinkage factor is towards 1.
In order to realize the identification of the cleaning degree of the photovoltaic array, a CURE (Clustering Using Representative) hierarchical clustering algorithm is utilized, all data in the cleaning abnormal data set are used as the input of a CURE hierarchical clustering algorithm, the shrinkage factor of the cleaning abnormal data set is used as the shrinkage factor of the CURE hierarchical clustering algorithm, the preset cluster number is 10, the preset representative point number is 15, the output of the CURE hierarchical clustering algorithm is used as the clustering result of the cleaning abnormal data set, and the CURE hierarchical clustering algorithm is a known technology and is not redundant.
Further, calculating the data average value of all data points in each cluster in the clustering result of the cleaning abnormal data set, taking a sequence formed by the data average values corresponding to all the clusters according to the ascending order of the numerical values as a cleaning grade sequence, taking the cluster corresponding to the first data in the cleaning grade sequence as the cluster of the cleaning grade of 1, taking the cluster corresponding to the second data in the cleaning grade sequence as the cluster of the cleaning grade of 2, taking the cluster corresponding to the third data in the cleaning grade sequence as the cluster of the cleaning grade of 3, and so on, and taking the clusters of different cleaning grades as the clustering analysis result of the photovoltaic array.
It should be noted that, each data point in the clusters of different cleaning grades corresponds to a photovoltaic array, for example, the photovoltaic array corresponding to the data point in the cluster of the 1 st cleaning grade, because the cleaning abnormality index of the data point in the cluster of the 1 st cleaning grade is the smallest, the surface cleaning degree of the photovoltaic array corresponding to the data point is the best; the photovoltaic array corresponding to the data points in the cluster of the 10 th cleaning grade indicates that the surface cleaning degree of the photovoltaic array corresponding to the data points is worst because the cleaning abnormality index of the data points in the cluster of the 10 th cleaning grade is the largest.
Thus, a clustering analysis result of the photovoltaic array is obtained.
And the cleaning decision module acquires the adjusted cleaning parameters according to the clustering analysis result of the photovoltaic array, and completes the automatic cleaning of the photovoltaic array.
And (3) formulating a cleaning strategy according to the clustering analysis result of the photovoltaic array, wherein the cleaning strategy comprises cleaning parameter settings under different cleaning degrees. The cleaning parameters include:
(1) Hydraulic pressure: photovoltaic arrays of different degrees of cleanliness require different levels of water pressure for cleaning.
(2) Cleaning time: photovoltaic cells with higher levels of contamination may require longer cleaning times.
When the photovoltaic array belongs to the photovoltaic array corresponding to the data points in the cluster of the 1 st cleaning level, the cleaning energy is saved by adopting lower 1 st level water pressure and shorter 1 st level cleaning time as the surface of the photovoltaic array has the best cleaning degree; when the photovoltaic array belongs to the photovoltaic array corresponding to the data point in the cluster of the 2 nd cleaning level, the photovoltaic array corresponding to the data point in the cluster of the 2 nd cleaning level is compared with the photovoltaic array corresponding to the data point in the cluster of the 1 st cleaning level, and the cleaning degree of the photovoltaic array corresponding to the data point in the cluster of the 2 nd cleaning level is relatively poor, and then the 2 nd water pressure slightly higher than the 1 st water pressure and the 2 nd cleaning time slightly higher than the 1 st cleaning time are adopted; when the photovoltaic array belongs to the photovoltaic array corresponding to the data point in the cluster of the 3 rd cleaning level, the photovoltaic array corresponding to the data point in the cluster of the 3 rd cleaning level is compared with the photovoltaic array corresponding to the data point in the cluster of the 2 nd cleaning level, and the cleaning degree of the photovoltaic array corresponding to the data point in the cluster of the 3 rd cleaning level is relatively poor, the 3 rd level water pressure slightly higher than the 2 nd level water pressure and the 3 rd level cleaning time slightly higher than the 2 nd level cleaning time are adopted; and so on; when the photovoltaic array belongs to the photovoltaic array corresponding to the data points in the clustering cluster of the 10 th cleaning level, because the surface cleaning degree of the photovoltaic array is worst, the higher 10 th level water pressure and the longer 10 th level cleaning time are adopted, so that the utilization efficiency of clean energy is improved.
And (3) automatically cleaning all the photovoltaic arrays based on the cleaning parameters by using a cleaning execution device, when each photovoltaic array belongs to the photovoltaic array corresponding to the data points in the cluster of the U-th cleaning level, the cleaning parameters are automatically adjusted to be the U-th level water pressure and the U-th level cleaning time (U=1, 2,3, … and 10), and the automatic cleaning of the photovoltaic arrays is realized by using a cleaning agent, a nozzle connected with a water pump and a cleaning brush.
Thus, the photovoltaic array automatic cleaning system is completed.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. The above description is only of the preferred embodiments of the present invention and is not intended to limit the invention, but any modifications, equivalent substitutions, improvements, etc. within the principles of the present invention should be included in the scope of the present invention.
Claims (10)
1. A photovoltaic array self-cleaning system, comprising the following modules:
The data acquisition module is used for acquiring a time sequence of related parameters of each photovoltaic array, wherein the related parameters comprise current, voltage, optical component temperature, ambient illumination intensity and ambient wind power;
The cluster analysis module is used for acquiring a photoelectric conversion quantity sequence of the photoelectric conversion quantity of each photovoltaic array according to the time sequence of the related parameters of each photovoltaic array and acquiring each gray correlation degree of the photoelectric conversion quantity of each photovoltaic array according to the photoelectric conversion quantity sequence of the photoelectric conversion quantity of each photovoltaic array; acquiring a cleaning degree index of each photovoltaic array according to each gray correlation degree of the photoelectric conversion quantity of each photovoltaic array; acquiring a heat dissipation data set of each photovoltaic array according to the time sequence of the temperature of the optical component of each photovoltaic array; acquiring a cleaning abnormality index of each photovoltaic array according to the heat radiation data set and the cleaning degree index of each photovoltaic array; acquiring a cleaning abnormal data set according to the cleaning abnormal index of each photovoltaic array, and acquiring the shrinkage coefficient of the cleaning abnormal data set according to the cleaning abnormal data set; acquiring a clustering result of the cleaning abnormal data set based on the shrinkage coefficient by using a CURE hierarchical clustering algorithm; obtaining a clustering analysis result of the photovoltaic array according to the clustering result of the cleaning abnormal data set;
And the cleaning decision module is used for classifying and cleaning the photovoltaic array according to the clustering analysis result of the photovoltaic array.
2. The automatic cleaning system for photovoltaic arrays according to claim 1, wherein the method for obtaining the photoelectric conversion value sequence of the photoelectric conversion value of each photovoltaic array according to the time sequence of the related parameters of each photovoltaic array and obtaining each gray correlation degree of the photoelectric conversion value of each photovoltaic array according to the photoelectric conversion value sequence of each photovoltaic array comprises the following steps:
For each photovoltaic array, taking the product of the current value and the voltage value at each acquisition time of the photovoltaic array as the photoelectric conversion quantity at each acquisition time of the photovoltaic array, and taking a sequence formed by the photoelectric conversion quantities at all the acquisition times of the photovoltaic array according to the ascending order of time as a photoelectric conversion quantity sequence of the photoelectric conversion quantity of the photovoltaic array;
The method comprises the steps of taking the ambient temperature, the ambient illumination intensity and the ambient wind power of a photovoltaic array as each correlation factor of the photoelectric conversion amount of the photovoltaic array, taking a time sequence of all correlation factors of the photoelectric conversion amount of the photovoltaic array and the photoelectric conversion amount of the photovoltaic array as input of a gray correlation analysis algorithm, taking output of the gray correlation analysis algorithm as the correlation degree between the photoelectric conversion amount of the photovoltaic array and each correlation factor of the photoelectric conversion amount of the photovoltaic array, and taking the correlation degree between the photoelectric conversion amount of the photovoltaic array and each correlation factor of the photoelectric conversion amount of the photovoltaic array as each gray correlation degree of the photoelectric conversion amount of the photovoltaic array.
3. The automatic cleaning system for photovoltaic arrays according to claim 1, wherein the method for obtaining the cleaning degree index of each photovoltaic array according to each gray correlation degree of the photoelectric conversion amount of each photovoltaic array comprises the following steps:
Acquiring the correlation intensity of the photoelectric conversion quantity of each photovoltaic array according to the time sequence of the photoelectric conversion quantity of each photovoltaic array and the correlation factor of the photoelectric conversion quantity;
For each photovoltaic array, taking a natural constant as a base number, taking a negative mapping result taking the correlation intensity of the photoelectric conversion quantity of the photovoltaic array as an index as a molecule, calculating the accumulated sum of each gray correlation degree of the photoelectric conversion quantity of the photovoltaic array on all correlation factors, taking the sum of the accumulated sum and a first preset parameter as a denominator, and taking the ratio of the molecule to the denominator as a first characteristic ratio;
And taking the sum of the opposite number of the first characteristic ratio and the first preset parameter as a cleanliness index of the photovoltaic array.
4. A photovoltaic array automatic cleaning system according to claim 3, wherein the method for obtaining the correlation intensity of the photovoltaic conversion amount of each photovoltaic array according to the time series of the correlation factor of the photovoltaic conversion amount and the photovoltaic conversion amount of each photovoltaic array comprises the following steps:
For each photovoltaic array, calculating a pearson correlation coefficient between a photoelectric conversion amount sequence of the photoelectric conversion amount of the photovoltaic array and a time sequence of each correlation factor of the photoelectric conversion amount, and taking the average value of the pearson correlation coefficient on all the correlation factors as the correlation intensity of the photovoltaic array.
5. The system of claim 1, wherein the method for obtaining the heat dissipation data set of each photovoltaic array according to the time sequence of the temperature of the optical component of each photovoltaic array comprises:
And for the time sequence of the optical component temperature of each photovoltaic array, calculating the difference value between the observed values of two adjacent moments in the time sequence of the optical component temperature, taking the sequence formed by the difference values according to the sequence of time ascending as the differential sequence of the time sequence of the optical component temperature, and taking the set formed by all data with negative values in the differential sequence of the time sequence of the optical component temperature as the heat dissipation data set of the photovoltaic array.
6. The automatic cleaning system for photovoltaic arrays according to claim 1, wherein the method for obtaining the cleaning abnormality index of each photovoltaic array according to the heat dissipation data set and the cleaning degree index of each photovoltaic array comprises the following steps:
in the method, in the process of the invention, Clean anomaly index representing the xth photovoltaic array,/>Representing the average of all data in the optical component temperature time series of the xth photovoltaic array,/>Representing the mean of all data in the ambient temperature time series of the xth photovoltaic array,/>Index of cleanliness representing the xth photovoltaic array,/>Representing error parameters,/>Representing the number of data in the heat dissipation data set of the xth photovoltaic array,/>Representing an exponential function based on natural constants,/>The value of the g data in the heat dissipation data set of the x photovoltaic array is represented.
7. The automatic cleaning system for photovoltaic arrays according to claim 1, wherein the method for acquiring the cleaning anomaly data set according to the cleaning anomaly index of each photovoltaic array and the shrinkage factor of the cleaning anomaly data set according to the cleaning anomaly data set comprises the following steps:
Taking a set formed by cleaning abnormality indexes of all the photovoltaic arrays as a cleaning abnormality data set, taking the cleaning abnormality data set as the input of a DPC density peak clustering algorithm, and taking the output of the DPC density peak clustering algorithm as the local density of each data point in the cleaning abnormality data set;
Taking the average value of the local densities of all data points in the cleaning abnormal data set as the tightness index of the cleaning abnormal data set, and taking the natural constant as a base and taking the negative mapping result of the tightness index of the cleaning abnormal data set as an index as the contraction coefficient of the cleaning abnormal data set.
8. The automatic cleaning system for a photovoltaic array according to claim 1, wherein the specific method for acquiring the clustering result of the cleaning abnormal data set based on the shrinkage factor by using the CURE hierarchical clustering algorithm is as follows:
Taking all cleaning abnormality indexes in the cleaning abnormality data set as input of a CURE hierarchical clustering algorithm, taking the shrinkage factor of the cleaning abnormality data set as the shrinkage factor of the CURE hierarchical clustering algorithm, and taking output of the CURE hierarchical clustering algorithm as a clustering result of the cleaning abnormality data set.
9. The automatic cleaning system for a photovoltaic array according to claim 1, wherein the method for obtaining the clustering analysis result of the photovoltaic array according to the clustering result of the cleaning abnormal data set comprises the following steps:
Calculating the average value of all cleaning abnormality indexes of each cluster in the clustering result of the cleaning abnormality data set, taking a sequence formed by the average values according to the ascending order of numerical values as a cleaning grade sequence, and taking the cluster corresponding to each data in the cleaning grade sequence as the cluster of each cleaning grade;
and each cleaning abnormality index in the cluster of each cleaning grade corresponds to each photovoltaic array, and the cluster of all the cleaning grades is used as a cluster analysis result of the photovoltaic array.
10. The automatic cleaning system for photovoltaic arrays according to claim 1, wherein the method for classifying and cleaning the photovoltaic arrays according to the result of the cluster analysis of the photovoltaic arrays comprises the following steps:
And for each cleaning-level cluster in the cluster analysis results of the photovoltaic arrays, cleaning the corresponding photovoltaic arrays in the clusters by using a cleaning execution device to acquire the cleaning results of the photovoltaic arrays by adopting the same-level cleaning parameters, wherein the same-level cleaning parameters comprise the same-level water pressure and the same-level cleaning time.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410381590.0A CN117978081B (en) | 2024-04-01 | 2024-04-01 | Photovoltaic array self-cleaning system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410381590.0A CN117978081B (en) | 2024-04-01 | 2024-04-01 | Photovoltaic array self-cleaning system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117978081A true CN117978081A (en) | 2024-05-03 |
CN117978081B CN117978081B (en) | 2024-06-14 |
Family
ID=90864951
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202410381590.0A Active CN117978081B (en) | 2024-04-01 | 2024-04-01 | Photovoltaic array self-cleaning system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117978081B (en) |
Citations (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105701727A (en) * | 2016-01-14 | 2016-06-22 | 国家电网公司 | User load classification method based on optimal clustering number and apparatus thereof |
CN114004139A (en) * | 2021-09-24 | 2022-02-01 | 沈阳工程学院 | Photovoltaic power generation power prediction method |
CN114169229A (en) * | 2021-11-24 | 2022-03-11 | 华能新能源股份有限公司 | Data-driven identification method and device for optimizing cleaning time of photovoltaic array |
CN116402289A (en) * | 2023-03-27 | 2023-07-07 | 苏州瑞得恩光能科技有限公司 | Method, device and storage medium for determining cleaning frequency of photovoltaic panel |
CN116582086A (en) * | 2023-04-07 | 2023-08-11 | 国家电投集团内蒙古新能源有限公司 | Photovoltaic power station cleanliness diagnosis method |
CN117118339A (en) * | 2023-08-02 | 2023-11-24 | 华能南京金陵发电有限公司 | Self-cleaning control method and device for photovoltaic power generation assembly |
CN117176075A (en) * | 2023-08-31 | 2023-12-05 | 尚特杰电力科技有限公司 | Method and device for cleaning and detecting photovoltaic cell panel, photovoltaic power station and storage medium |
CN117277958A (en) * | 2023-11-21 | 2023-12-22 | 江苏达海智能***股份有限公司 | Intelligent operation and maintenance management method and system for photovoltaic power station based on big data |
CN117349630A (en) * | 2023-12-04 | 2024-01-05 | 邢台医学高等专科学校 | Method and system for biochemical data analysis |
CN117421616A (en) * | 2023-12-18 | 2024-01-19 | 北京年管家信息科技有限公司 | Mine shaft detection system and method |
CN117584033A (en) * | 2022-08-17 | 2024-02-23 | 株式会社荏原制作所 | Information processing apparatus and method, machine learning apparatus and method |
CN117614110A (en) * | 2023-10-27 | 2024-02-27 | 国网河北省电力有限公司营销服务中心 | Photovoltaic equipment operation and maintenance system and method |
CN117783658A (en) * | 2024-01-08 | 2024-03-29 | 北京广源佳鑫科技有限公司 | Current digital measurement method for power equipment |
-
2024
- 2024-04-01 CN CN202410381590.0A patent/CN117978081B/en active Active
Patent Citations (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105701727A (en) * | 2016-01-14 | 2016-06-22 | 国家电网公司 | User load classification method based on optimal clustering number and apparatus thereof |
CN114004139A (en) * | 2021-09-24 | 2022-02-01 | 沈阳工程学院 | Photovoltaic power generation power prediction method |
CN114169229A (en) * | 2021-11-24 | 2022-03-11 | 华能新能源股份有限公司 | Data-driven identification method and device for optimizing cleaning time of photovoltaic array |
CN117584033A (en) * | 2022-08-17 | 2024-02-23 | 株式会社荏原制作所 | Information processing apparatus and method, machine learning apparatus and method |
CN116402289A (en) * | 2023-03-27 | 2023-07-07 | 苏州瑞得恩光能科技有限公司 | Method, device and storage medium for determining cleaning frequency of photovoltaic panel |
CN116582086A (en) * | 2023-04-07 | 2023-08-11 | 国家电投集团内蒙古新能源有限公司 | Photovoltaic power station cleanliness diagnosis method |
CN117118339A (en) * | 2023-08-02 | 2023-11-24 | 华能南京金陵发电有限公司 | Self-cleaning control method and device for photovoltaic power generation assembly |
CN117176075A (en) * | 2023-08-31 | 2023-12-05 | 尚特杰电力科技有限公司 | Method and device for cleaning and detecting photovoltaic cell panel, photovoltaic power station and storage medium |
CN117614110A (en) * | 2023-10-27 | 2024-02-27 | 国网河北省电力有限公司营销服务中心 | Photovoltaic equipment operation and maintenance system and method |
CN117277958A (en) * | 2023-11-21 | 2023-12-22 | 江苏达海智能***股份有限公司 | Intelligent operation and maintenance management method and system for photovoltaic power station based on big data |
CN117349630A (en) * | 2023-12-04 | 2024-01-05 | 邢台医学高等专科学校 | Method and system for biochemical data analysis |
CN117421616A (en) * | 2023-12-18 | 2024-01-19 | 北京年管家信息科技有限公司 | Mine shaft detection system and method |
CN117783658A (en) * | 2024-01-08 | 2024-03-29 | 北京广源佳鑫科技有限公司 | Current digital measurement method for power equipment |
Also Published As
Publication number | Publication date |
---|---|
CN117978081B (en) | 2024-06-14 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112871938B (en) | Photovoltaic power plant cleans control integration management system | |
CN111245365B (en) | Photovoltaic module fault diagnosis method based on dynamic weighted depth forest | |
CN107064165B (en) | Photovoltaic module surface area gray scale online detection device and cleaning method | |
CN115495505A (en) | Airport group data correlation analysis method based on new energy | |
CN117540859B (en) | Method and system for estimating generated energy of photovoltaic power station based on neural network | |
CN117978081B (en) | Photovoltaic array self-cleaning system | |
CN117713218B (en) | Knowledge and data fusion driving-based photovoltaic output prediction method | |
CN110796292A (en) | Photovoltaic power short-term prediction method considering haze influence | |
CN110729767A (en) | Water-electricity-containing regional power grid wind-solar capacity optimal configuration method | |
CN112505518B (en) | Method, device and system for detecting dust accumulation of photovoltaic string | |
CN116800172B (en) | Photoelectric conversion method and energy storage system | |
CN116433217A (en) | Distributed photovoltaic power generation management method and system | |
CN112085350A (en) | Method for evaluating photovoltaic array state in large photovoltaic power station | |
Makkar et al. | IoT based framework: mathematical modelling and analysis of dust impact on solar panels | |
CN115456440A (en) | Method, device and system for evaluating power generation characteristics of distributed photovoltaic power station | |
CN115310839A (en) | Photovoltaic power station dust deposition early warning assessment method and system | |
CN114021838A (en) | Photovoltaic power station dust coverage prediction method | |
Mahmoud et al. | Classification of solar variability using K-means method for the evaluation of solar photovoltaic systems performance | |
CN111881969A (en) | Photovoltaic system tracking optimization method based on small-scale irradiation prediction | |
CN112288285B (en) | Photovoltaic power station string attenuation rate evaluation method based on generalized extreme value distribution | |
CN117937478B (en) | Photovoltaic power prediction method, device, computer equipment and storage medium | |
CN116843105A (en) | Photovoltaic module power management system based on neural network and model training method | |
Wang et al. | A multi-sensor information fusion monitoring system for photovoltaic power generation | |
CN118279084A (en) | Photovoltaic intra-day output prediction method and device | |
CN115764861A (en) | Wind-solar integrated power prediction method based on machine learning |
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 |