CN108335042B - Method for calculating cleaning index of dynamic photovoltaic panel - Google Patents
Method for calculating cleaning index of dynamic photovoltaic panel Download PDFInfo
- Publication number
- CN108335042B CN108335042B CN201810114996.7A CN201810114996A CN108335042B CN 108335042 B CN108335042 B CN 108335042B CN 201810114996 A CN201810114996 A CN 201810114996A CN 108335042 B CN108335042 B CN 108335042B
- Authority
- CN
- China
- Prior art keywords
- value
- index
- power generation
- clusters
- loss rate
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Operations Research (AREA)
- Mathematical Physics (AREA)
- Tourism & Hospitality (AREA)
- Mathematical Analysis (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Administration (AREA)
- Marketing (AREA)
- Pure & Applied Mathematics (AREA)
- Development Economics (AREA)
- Mathematical Optimization (AREA)
- Computational Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Probability & Statistics with Applications (AREA)
- Life Sciences & Earth Sciences (AREA)
- Algebra (AREA)
- Quality & Reliability (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Photovoltaic Devices (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a method for calculating a dynamic photovoltaic panel cleaning index based on PM value data. The method comprises the following steps: collecting the dust accumulation degree, the power generation loss rate, the expected power generation value and the PM value of the location of a local power station; performing linear regression analysis using the PM value and the dust accumulation degree; predicting the power generation loss rate according to linear regression, and expanding an original data set; and rearranging by using a clustering method according to the power generation loss rate, wherein the final sequence is the recommended cleaning index. According to the invention, the dust accumulation degree and the PM value are fitted through regression analysis, after data fitting is completed, the PM value which is easy to obtain can be used as a characteristic parameter to be cooperatively filtered with other characteristics such as generating capacity and electric quantity loss index, an accurate cleaning index is obtained under the condition of minimum investment, a user is automatically reminded whether the dust deposition degree of the photovoltaic panel needs to be cleaned, the photovoltaic panel is cleaned in time, the power generation efficiency is improved, and the cost is saved.
Description
Technical Field
The invention relates to the technical field of photovoltaic panel dust monitoring, in particular to a method for calculating a dynamic photovoltaic panel cleaning index estimated based on PM value data.
Background
Because the generating capacity of the photovoltaic power generation system is the most important index for evaluating the performance of the photovoltaic power station, the influence of dust on the generating performance of the photovoltaic module is universal, atmospheric dust is one of key factors influencing the solar power generation efficiency, and especially, long-time wind and sand cause dust and other pollutants to shield the photovoltaic module, influence the transmittance of light rays and further influence the radiation quantity received by the surface of the module.
Meanwhile, because the distances between the pollutants and the photovoltaic cell pieces are very close, shadows are formed, and a hot spot effect is formed on the photovoltaic module. If the photovoltaic module is not cleaned in time for a long time, the generated energy of the photovoltaic power station can be greatly reduced, the requirement of a power grid cannot be met, and the utilization rate of a photovoltaic power generation system is also reduced. At present, the research on dust mainly stays in monitoring the accumulation degree of dust by using a sensor, and the method has high laying cost and needs to install a large number of sensors.
Disclosure of Invention
Aiming at the problems that the existing photovoltaic monitoring is lack of effective means for monitoring dust, high in complexity, extremely high in dependence on a sensor and not mature and complete enough, the invention provides a calculation method for evaluating a dynamic photovoltaic panel cleaning index according to simple and easily obtained PM value data, regression analysis of the PM value and dust accumulation degree can be realized, and a recommendation algorithm for the cleaning index is obtained by analyzing characteristic values.
The technical scheme adopted by the invention for solving the technical problems is as follows:
a method for calculating a dynamic photovoltaic panel cleaning index is carried out according to the following steps:
data acquisition → regression analysis → standardized data → clustering → index rearrangement; the data acquisition is to collect the dust accumulation degree, the power generation loss rate, the expected power generation value and the PM value of the location of a local power station; regression analysis linear regression analysis was performed using PM values and dust accumulation degrees; the standardized data is used for predicting the power generation loss rate according to linear regression and expanding an original data set; and clustering division is to use a clustering method and rearrange according to the power generation loss rate, and the final sequence is the recommended cleaning index.
Compared with the prior art, the invention adopting the technical scheme has the beneficial effects that:
according to the invention, the dust accumulation degree and the PM value are fitted through regression analysis, after data fitting is completed, the PM value which is easy to obtain can be used as a characteristic parameter to be cooperatively filtered with other characteristics such as generating capacity and electric quantity loss index, an accurate cleaning index is obtained under the condition of minimum investment, a user is automatically reminded whether the dust deposition degree of the photovoltaic panel needs to be cleaned, the photovoltaic panel is cleaned in time, the power generation efficiency is improved, and the cost is saved.
Further, the preferred scheme of the invention is as follows:
in the linear regression analysis, in the initial stage, a linear regression equation of a PM index, a dust formation index and a power loss rate is used for carrying out regression analysis, and then new regression analysis is automatically generated according to sensor data; the specific linear regression equation is:
wherein x is a PM value, y is a dust accumulation degree, R is a linear relation coefficient between y and x, and n is a constant.
The PM value is the PM value of the power station where the power station is located and is obtained according to the public API.
The clustering method uses an improved algorithm of a K-means algorithm, and sets a K value as 100, and specifically comprises the following steps:
(1) randomly taking 100 elements from the dissimilarity degree calculation formula d as respective centers of the k clusters;
(2) respectively calculating the dissimilarity degree of the rest elements to the centers of 100 clusters, and classifying the elements into the clusters with the lowest dissimilarity degree;
(3) according to the clustering result, re-calculating the respective centers of the k clusters, wherein the calculation method is to take the arithmetic mean of the respective dimensions of all elements in the clusters;
(4) re-clustering all elements in the dissimilarity degree calculation formula d according to the new centers;
(5) repeating the step 4 until the clustering result is not changed;
(6) outputting the result;
(7) and arranging 100 clusters from small to large according to the loss rate, and dividing the clusters into 1-100% of cleaning indexes respectively.
Drawings
FIG. 1 is a graphical representation of the correlation of PM index to soot index;
FIG. 2 is a schematic illustration of a recommendation index.
Detailed Description
In order to describe the present invention more specifically, the following detailed description of the technical solutions and the related principles of the present invention is provided with reference to the accompanying drawings and embodiments.
The method for calculating the cleaning index of the dynamic photovoltaic panel comprises the following specific steps:
(1) and acquiring the PM value of the power station according to the public API, and substituting the PM value into a linear regression equation according to the predicted correlation to obtain the correlation between the PM value and the output reduction rate.
(2) The parameters are input as characteristic values and normalized.
(3) Clustering was performed using the K-means method.
(4) And rearranging the clusters according to the loss rate, wherein the clusters from low to high are the recommended cleaning index.
Referring to fig. 1, a regression analysis is performed on the local PM value and the measured data according to the test data, and the regression equations of the PM index, the ash accumulation index, and the power loss rate are as follows:
in the formula: x is a PM value, y is a dust accumulation degree, R is a linear relation coefficient between y and x, and n is a constant. And a linear relation exists between y and x, the relation degree is described by a correlation coefficient R, and the magnitude of the absolute value of the correlation coefficient R represents the degree of the correlation. For example, a current coefficient of 0.82 indicates partial correlation, a large value of the brillouin indicates a high degree of correlation.
Estimating the standard deviation S using the predicted value of the dust accumulation degree with the current PM value knownyAnd a prediction interval, the formula being as follows:
estimating standard deviation:
in the formula: syIs a standard deviation, x is a PM value, y is a dust accumulation degree, n is a constant,
is a subset.
Prediction interval:
in the formula: syIs a standard deviation, x is a PM value, y is a dust accumulation degree, n is a constant,
for the purpose of the subset,
are averages.
The values in the prediction interval formula cannot be directly applied to the actual power station, and content analysis is performed according to other parameters of the current power station. For example, the selected feature values include: PM value, installation angle, power generation loss rate and cleaning period.
To obtain a reliable index of cleaning, a normalization process is required. The Min-max standardization method is to perform linear transformation on original data, set minA and maxA as the minimum value and the maximum value of the attribute A respectively, and map an original value x of A into a value x' in an interval [0,1] through Min-max standardization, wherein the formula is as follows:
new data is (original data-min)/(max-min).
After the data are normalized, calculating the dissimilarity degree, wherein the dissimilarity degree calculation formula is as follows:
wherein the formula is as follows: x is the PM value, y is the dust accumulation degree, m is the quantification, lambda is the constant, and i is the variable;
using a modified algorithm of the K-means algorithm, set the K value to 100:
randomly taking 100 elements from a dissimilarity degree calculation formula d as respective centers of k clusters;
calculating the dissimilarity degree of the remaining elements to the centers of 100 clusters respectively, and classifying the elements into the cluster with the lowest dissimilarity degree respectively;
recalculating centers of the k clusters according to the clustering result, wherein the calculation method is to take an arithmetic mean of dimensions of all elements in the clusters;
fourthly, clustering all elements in the d again according to the new center;
fifthly, repeating the step 4 until the clustering result is not changed;
sixthly, outputting the result;
seventhly, arranging 100 clusters from small to large according to the loss rate, respectively dividing the clusters into recommended cleaning indexes of 1-100 percent,
the final result is shown in fig. 2.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the scope of the present invention, which is defined in the appended claims.
Claims (2)
1. A method for calculating a dynamic photovoltaic panel cleaning index is characterized by comprising the following steps:
data acquisition → regression analysis → standardized data → clustering → index rearrangement; the data acquisition is to collect the dust accumulation degree, the power generation loss rate, the expected power generation value and the PM value of the location of a local power station; regression analysis linear regression analysis was performed using PM values and dust accumulation degrees; the standardized data is used for predicting the power generation loss rate according to linear regression and expanding an original data set; clustering division is to use a clustering method and rearrange according to the power generation loss rate, and the final sequence is the recommended cleaning index; specifically, the method comprises the following steps:
in the linear regression analysis, in the initial stage, a linear regression equation of a PM index, a dust formation index and a power loss rate is used for carrying out regression analysis, and then new regression analysis is automatically generated according to sensor data; the specific linear regression equation is:
wherein x is a PM value, y is a dust accumulation degree, R is a linear relation coefficient between y and x, and n is a constant;
the clustering method uses an improved algorithm of a K-means algorithm, and sets a K value as 100, and specifically comprises the following steps:
(1) randomly taking 100 elements from the dissimilarity degree calculation formula d as respective centers of the k clusters;
(2) respectively calculating the dissimilarity degree of the rest elements to the centers of 100 clusters, and classifying the elements into the clusters with the lowest dissimilarity degree;
(3) according to the clustering result, re-calculating the respective centers of the k clusters, wherein the calculation method is to take the arithmetic mean of the respective dimensions of all elements in the clusters;
(4) re-clustering all elements in the dissimilarity degree calculation formula d according to the new centers;
(5) repeating the step 4 until the clustering result is not changed;
(6) outputting the result;
(7) and arranging 100 clusters from small to large according to the loss rate, and dividing the clusters into 1-100% of cleaning indexes respectively.
2. The method for calculating the index of cleaning of a dynamic photovoltaic panel according to claim 1, characterized in that: the PM value is the PM value of the power station where the power station is located and is obtained according to the public API.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810114996.7A CN108335042B (en) | 2018-02-06 | 2018-02-06 | Method for calculating cleaning index of dynamic photovoltaic panel |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810114996.7A CN108335042B (en) | 2018-02-06 | 2018-02-06 | Method for calculating cleaning index of dynamic photovoltaic panel |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108335042A CN108335042A (en) | 2018-07-27 |
| CN108335042B true CN108335042B (en) | 2022-02-01 |
Family
ID=62928106
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810114996.7A Active CN108335042B (en) | 2018-02-06 | 2018-02-06 | Method for calculating cleaning index of dynamic photovoltaic panel |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108335042B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112926798A (en) * | 2021-03-26 | 2021-06-08 | 苏州朋友保科技有限公司 | Method, device, equipment and medium for predicting photovoltaic power generation loss caused by dust |
| CN113705881B (en) * | 2021-08-25 | 2024-03-29 | 陕西启迪瑞行清洁能源研究院有限公司 | Method for predicting cleaning index of plate heat exchanger based on logarithmic temperature difference |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102201037A (en) * | 2011-06-14 | 2011-09-28 | 中国农业大学 | Agricultural disaster forecast method |
| CN103268366A (en) * | 2013-03-06 | 2013-08-28 | 辽宁省电力有限公司电力科学研究院 | Combined wind power prediction method suitable for distributed wind power plant |
| CN104021427A (en) * | 2014-06-10 | 2014-09-03 | 上海电力学院 | Method for predicting daily generating capacity of grid-connected photovoltaic power station based on factor analysis |
| CN107064165A (en) * | 2017-05-23 | 2017-08-18 | 扬州鸿淏新能源科技有限公司 | A kind of photovoltaic module surface area gray scale on-line measuring device and cleaning method |
| US9740545B2 (en) * | 2015-03-20 | 2017-08-22 | Kabushiki Kaisha Toshiba | Equipment evaluation device, equipment evaluation method and non-transitory computer readable medium |
-
2018
- 2018-02-06 CN CN201810114996.7A patent/CN108335042B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102201037A (en) * | 2011-06-14 | 2011-09-28 | 中国农业大学 | Agricultural disaster forecast method |
| CN103268366A (en) * | 2013-03-06 | 2013-08-28 | 辽宁省电力有限公司电力科学研究院 | Combined wind power prediction method suitable for distributed wind power plant |
| CN104021427A (en) * | 2014-06-10 | 2014-09-03 | 上海电力学院 | Method for predicting daily generating capacity of grid-connected photovoltaic power station based on factor analysis |
| US9740545B2 (en) * | 2015-03-20 | 2017-08-22 | Kabushiki Kaisha Toshiba | Equipment evaluation device, equipment evaluation method and non-transitory computer readable medium |
| CN107064165A (en) * | 2017-05-23 | 2017-08-18 | 扬州鸿淏新能源科技有限公司 | A kind of photovoltaic module surface area gray scale on-line measuring device and cleaning method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108335042A (en) | 2018-07-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112952820B (en) | Intelligent community micro-grid ultra-multi-target energy management method considering retired batteries | |
| CN112633924B (en) | District electric energy substitution demand analysis method based on load decomposition | |
| CN108335042B (en) | Method for calculating cleaning index of dynamic photovoltaic panel | |
| CN115829105A (en) | Photovoltaic power prediction method based on historical data feature search | |
| CN106384015A (en) | Energy consumption data collection-based distributed photovoltaic system energy efficiency pre-estimation method | |
| CN109460419A (en) | A kind of equipment state change events monitoring method | |
| CN116796403A (en) | Building energy saving method based on comprehensive energy consumption prediction of commercial building | |
| CN112257964B (en) | Load-intensive urban intelligent park demand aggregation modeling method | |
| CN110796292A (en) | Photovoltaic power short-term prediction method considering haze influence | |
| CN108876060B (en) | Big data based prediction method for wind power output probability of sample collection | |
| CN109064353B (en) | Large building user behavior analysis method based on improved cluster fusion | |
| CN115932435A (en) | Resident non-invasive load monitoring method based on low-frequency acquisition signals | |
| CN116050592A (en) | Multi-dimensional photovoltaic power prediction method and system | |
| CN115733258A (en) | Control method of all-indoor intelligent substation system based on Internet of things technology | |
| CN112052430B (en) | Photovoltaic station correlation comprehensive evaluation system based on improved TOPSIS | |
| CN113344293B (en) | Photovoltaic power prediction method based on NCA-fusion regression tree model | |
| CN115575737A (en) | Current-carrying capacity detection method and system | |
| CN112925827A (en) | User property abnormity analysis method based on power acquisition Internet of things data | |
| CN112991574B (en) | Method for analyzing attenuation of electric pile | |
| CN114662563A (en) | Industrial electricity non-invasive load decomposition method based on gradient lifting algorithm | |
| CN109359134B (en) | Method for identifying energy consumption implicit abnormal data of lighting socket based on data mining | |
| CN109378816B (en) | Method and system for searching running opportunity of electricity conversion gas in energy system | |
| Su et al. | Research on Two-stage Identification of Distributed Photovoltaic Output Based on WGAN Data Reconstruction Technology | |
| CN111162566B (en) | Method and system for determining joint output characteristic of wind-solar combined power generation system | |
| CN117291299B (en) | Moon electricity quantity prediction method considering various influence factors |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20220507 Address after: 266100 1808, 18 / F, block B, Shenghe building, No. 58, shangdongtou Road, Laoshan District, Qingdao, Shandong Province Patentee after: Qingdao Guanxin Technology Co.,Ltd. Address before: 063100 middle section of Aimin Road, Tangjiazhuang, Guye District, Tangshan City, Hebei Province Patentee before: TANGSHAN YINGTONG TECHNOLOGY CO.,LTD. |
|
| TR01 | Transfer of patent right |