CN116050592A - Multi-dimensional photovoltaic power prediction method and system - Google Patents
Multi-dimensional photovoltaic power prediction method and system Download PDFInfo
- Publication number
- CN116050592A CN116050592A CN202211681975.6A CN202211681975A CN116050592A CN 116050592 A CN116050592 A CN 116050592A CN 202211681975 A CN202211681975 A CN 202211681975A CN 116050592 A CN116050592 A CN 116050592A
- Authority
- CN
- China
- Prior art keywords
- power
- module
- data
- photovoltaic power
- prediction
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 40
- 238000010248 power generation Methods 0.000 claims abstract description 35
- 238000004458 analytical method Methods 0.000 claims abstract description 25
- 238000004364 calculation method Methods 0.000 claims abstract description 19
- 238000012545 processing Methods 0.000 claims description 25
- 238000012549 training Methods 0.000 claims description 23
- 238000011156 evaluation Methods 0.000 claims description 21
- 238000010276 construction Methods 0.000 claims description 18
- 238000003672 processing method Methods 0.000 claims description 17
- 238000007689 inspection Methods 0.000 claims description 14
- 238000007405 data analysis Methods 0.000 claims description 10
- 238000001514 detection method Methods 0.000 claims description 2
- 230000000694 effects Effects 0.000 abstract description 10
- 230000007613 environmental effect Effects 0.000 abstract description 5
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 238000005286 illumination Methods 0.000 description 12
- 238000012217 deletion Methods 0.000 description 5
- 230000037430 deletion Effects 0.000 description 5
- 238000005457 optimization Methods 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000005855 radiation Effects 0.000 description 3
- 238000012795 verification Methods 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 230000001502 supplementing effect Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 238000001816 cooling Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000004907 flux Effects 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 230000009469 supplementation Effects 0.000 description 1
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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- 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—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- 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
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Entrepreneurship & Innovation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Artificial Intelligence (AREA)
- Quality & Reliability (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Development Economics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a multi-dimensional photovoltaic power prediction method and system, and belongs to the technical field of photovoltaic power prediction methods and equipment. The technical scheme of the invention is as follows: acquiring basic information of a photovoltaic power station, performing historical data calling, acquiring historical data information, acquiring a first power limiting factor, a second power limiting factor and a third power limiting factor, performing influence correlation calculation, acquiring factor-power influence correlation, constructing a photovoltaic power prediction model, performing real-time data acquisition on the power generation of the photovoltaic power station, and inputting a real-time data acquisition result into the photovoltaic power prediction model to obtain predicted power data. The beneficial effects of the invention are as follows: by means of deep analysis of environmental factors, a photovoltaic power prediction model is built, real-time data acquisition results are input into the photovoltaic power prediction model, predicted power data are obtained, and the effect of improving the power prediction accuracy of the photovoltaic power station is achieved.
Description
Technical Field
The invention relates to a multi-dimensional photovoltaic power prediction method and system, and belongs to the technical field of photovoltaic power prediction methods and equipment.
Background
The photovoltaic power generation is used as a novel renewable energy source, has the advantages of being renewable, pollution-free, wide in source and the like compared with the traditional energy source, is a main choice for replacing fossil energy sources, and is continuously developed due to the advantages. At present, along with the increasingly wide application of large-scale photovoltaic power generation systems, more and more problems also appear. Because solar radiation quantity is related to weather conditions such as quarterly, night, sunny and sunny, the inherent defects of the output power of the photovoltaic power generation system such as the worry and the randomness are caused, when a large number of photovoltaic power generation systems are connected into a power grid, serious challenges are brought to safe and stable operation, electric energy quality and the like of the power system, and the development speed and the scale of photovoltaic power generation are limited, so that the prediction of the output power of the photovoltaic power generation system has very important significance for the operation of the power system. The conventional photovoltaic power generation power prediction analysis method has certain defects, and certain lifting space exists for the photovoltaic power generation power prediction analysis.
In the prior art, the influence of environmental factors on the photovoltaic power station is not mastered enough, so that the photovoltaic power prediction is not accurate enough.
Disclosure of Invention
The invention aims to provide a multi-dimensional photovoltaic power prediction method and system, which are used for constructing a photovoltaic power prediction model through deep analysis of environmental factors, inputting real-time data acquisition results into the photovoltaic power prediction model to obtain predicted power data, so that the effect of improving the power prediction accuracy of a photovoltaic power station is achieved, and the problems in the background technology are effectively solved.
The technical scheme of the invention is as follows: a multi-dimensional photovoltaic power prediction method comprising the steps of:
(1) Acquiring basic information of a photovoltaic power station;
(2) Based on a preset time period, historical data of the photovoltaic power station is called, and historical data information is obtained;
(3) Acquiring a first power limiting factor, a second power limiting factor and a third power limiting factor according to the basic information and the historical data information of the photovoltaic power station;
(4) Performing influence correlation calculation on the power of the photovoltaic power station on the first power limiting factor, the second power limiting factor and the third power limiting factor respectively to obtain factor-power influence correlation;
(5) Constructing a photovoltaic power prediction model according to the factor-power influence correlation degree;
(6) The method comprises the steps of carrying out real-time data acquisition on a power limiting factor of a photovoltaic power station through a data acquisition device, and acquiring a real-time data acquisition result;
(7) And inputting the real-time data acquisition result into a photovoltaic power prediction model to obtain predicted power data.
Further, the step (1) includes the following steps:
retrieving a plurality of sets of historical data based on a preset time period, wherein each set of historical data comprises a multi-dimensional data type;
carrying out data analysis on multiple groups of historical data to obtain information missing data points;
performing data missing inspection on the information missing data points to obtain a data missing inspection result;
and processing the plurality of groups of historical data according to the data missing checking result to acquire historical data information.
Further, the processing of the plurality of sets of historical data according to the data missing inspection result, to obtain historical data information, includes the following steps:
performing feature analysis on the data missing inspection result to obtain a feature analysis result;
matching corresponding missing value processing methods according to the feature analysis result;
and processing a plurality of groups of historical data according to the missing value processing method.
Further, the step (5) includes the following steps:
taking the influence correlation degree of the first power limiting factor and the photovoltaic power station power as first training information, and constructing a first power prediction module;
taking the influence correlation degree of the second power limiting factor and the photovoltaic power station power as second training information, and constructing a second power prediction module;
taking the influence correlation degree of the third power limiting factor and the photovoltaic power station power as third training information, and constructing a third power prediction module;
and constructing the photovoltaic power prediction model based on the first power prediction module, the second power prediction module and the third power prediction module.
Further, the method also comprises the following steps:
acquiring a prediction period, and generating a power recording node according to the prediction period;
when the photovoltaic power station runs to the power recording node, recording the real-time power generation power of the photovoltaic power station, and acquiring real-time power generation power data;
and comparing the predicted power data with the real-time power generation power data to obtain a power prediction evaluation result.
Further, the method also comprises the following steps:
acquiring a preset power prediction threshold value;
and when the power prediction evaluation result does not meet the power prediction threshold value, optimizing the photovoltaic power prediction model according to the power prediction evaluation result.
The multi-dimensional photovoltaic power prediction system comprises a basic information acquisition module, a historical data calling module, a limiting factor acquisition module, a correlation calculation module, a prediction model construction module, a real-time data acquisition module and a prediction power data acquisition module, wherein the basic information acquisition module and the historical data calling module are respectively connected with the limiting factor acquisition module, the correlation calculation module and the prediction model construction module are sequentially connected, and the output ends of the prediction model construction module and the real-time data acquisition module are respectively connected with the prediction power data acquisition module.
Further, the basic information acquisition module comprises a historical data calling module, a data analysis module, a data missing checking module and a historical data processing module, which are sequentially connected.
Further, the historical data processing module comprises a characteristic analysis module, a processing method matching module and a data processing module which are sequentially connected.
Further, the prediction model building module comprises a first power prediction module building module, a second power prediction module building module, a third power prediction module building module and a photovoltaic power prediction model building module, and the first power prediction module building module, the second power prediction module building module and the third power prediction module building module are respectively connected with the photovoltaic power prediction model building module.
Further, the system also comprises a power record node generation module, a real-time power generation power record module and a power prediction evaluation result acquisition module which are sequentially connected.
Further, the system also comprises a preset power prediction threshold acquisition module and a model optimization module which are connected with each other.
The beneficial effects of the invention are as follows: by means of deep analysis of environmental factors, a photovoltaic power prediction model is built, real-time data acquisition results are input into the photovoltaic power prediction model, predicted power data are obtained, and the effect of improving the power prediction accuracy of the photovoltaic power station is achieved.
Drawings
FIG. 1 is a schematic flow chart of the method of the present invention;
FIG. 2 is a schematic diagram of a process for obtaining historical data information in the method of the present invention;
FIG. 3 is a schematic diagram of a system architecture of the present invention;
in the figure: the system comprises a basic information acquisition module 10, a historical data calling module 20, a limiting factor acquisition module 30, a correlation calculation module 40, a prediction model construction module 50, a real-time data acquisition module 60 and a prediction power data acquisition module 70.
Detailed Description
The following describes the technical scheme of the present invention in further detail by referring to the accompanying drawings and examples, which are preferred examples of the present invention. It should be understood that the described embodiments are merely some, but not all, embodiments of the present invention; it should be noted that, without conflict, the embodiments of the present invention and features of the embodiments may be combined with each other. 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.
A multi-dimensional photovoltaic power prediction method comprising the steps of:
(1) Acquiring basic information of a photovoltaic power station;
(2) Based on a preset time period, historical data of the photovoltaic power station is called, and historical data information is obtained;
(3) Acquiring a first power limiting factor, a second power limiting factor and a third power limiting factor according to the basic information and the historical data information of the photovoltaic power station;
(4) Performing influence correlation calculation on the power of the photovoltaic power station on the first power limiting factor, the second power limiting factor and the third power limiting factor respectively to obtain factor-power influence correlation;
(5) Constructing a photovoltaic power prediction model according to the factor-power influence correlation degree;
(6) The method comprises the steps of carrying out real-time data acquisition on a power limiting factor of a photovoltaic power station through a data acquisition device, and acquiring a real-time data acquisition result;
(7) And inputting the real-time data acquisition result into a photovoltaic power prediction model to obtain predicted power data.
The step (1) comprises the following steps:
retrieving a plurality of sets of historical data based on a preset time period, wherein each set of historical data comprises a multi-dimensional data type;
carrying out data analysis on multiple groups of historical data to obtain information missing data points;
performing data missing inspection on the information missing data points to obtain a data missing inspection result;
and processing the plurality of groups of historical data according to the data missing checking result to acquire historical data information.
The processing of the plurality of groups of historical data according to the data missing checking result, obtaining historical data information, comprises the following steps:
performing feature analysis on the data missing inspection result to obtain a feature analysis result;
matching corresponding missing value processing methods according to the feature analysis result;
and processing a plurality of groups of historical data according to the missing value processing method.
The step (5) comprises the following steps:
taking the influence correlation degree of the first power limiting factor and the photovoltaic power station power as first training information, and constructing a first power prediction module;
taking the influence correlation degree of the second power limiting factor and the photovoltaic power station power as second training information, and constructing a second power prediction module;
taking the influence correlation degree of the third power limiting factor and the photovoltaic power station power as third training information, and constructing a third power prediction module;
and constructing the photovoltaic power prediction model based on the first power prediction module, the second power prediction module and the third power prediction module.
Further, the method also comprises the following steps:
acquiring a prediction period, and generating a power recording node according to the prediction period;
when the photovoltaic power station runs to the power recording node, recording the real-time power generation power of the photovoltaic power station, and acquiring real-time power generation power data;
and comparing the predicted power data with the real-time power generation power data to obtain a power prediction evaluation result.
Further, the method also comprises the following steps:
acquiring a preset power prediction threshold value;
and when the power prediction evaluation result does not meet the power prediction threshold value, optimizing the photovoltaic power prediction model according to the power prediction evaluation result.
The multi-dimensional photovoltaic power prediction system comprises a basic information acquisition module 10, a historical data calling module 20, a restriction factor acquisition module 30, a correlation calculation module 40, a prediction model construction module 50, a real-time data acquisition module 60 and a prediction power data acquisition module 70, wherein the basic information acquisition module 10 and the historical data calling module 20 are respectively connected with the restriction factor acquisition module 30, the correlation calculation module 40 and the prediction model construction module 50 are sequentially connected, and the output ends of the prediction model construction module 50 and the real-time data acquisition module 60 are respectively connected with the prediction power data acquisition module 70.
The basic information acquisition module 10 comprises a historical data calling module, a data analysis module, a data missing checking module and a historical data processing module which are sequentially connected.
The historical data processing module comprises a characteristic analysis module, a processing method matching module and a data processing module which are sequentially connected.
The prediction model building module 50 includes a first power prediction module building module, a second power prediction module building module, a third power prediction module building module, and a photovoltaic power prediction model building module, where the first power prediction module building module, the second power prediction module building module, and the third power prediction module building module are respectively connected with the photovoltaic power prediction model building module.
Further, the system also comprises a power record node generation module, a real-time power generation power record module and a power prediction evaluation result acquisition module which are sequentially connected.
Further, the system also comprises a preset power prediction threshold acquisition module and a model optimization module which are connected with each other.
In practical application, the multi-dimensional photovoltaic power prediction method comprises the following steps: acquiring basic information of a photovoltaic power station; based on a preset time period, historical data of the photovoltaic power station is called, and historical data information is obtained; acquiring a first power limiting factor, a second power limiting factor and a third power limiting factor according to the basic information and the historical data information of the photovoltaic power station; performing influence correlation calculation on the power of the photovoltaic power station on the first power limiting factor, the second power limiting factor and the third power limiting factor respectively to obtain factor-power influence correlation; constructing a photovoltaic power prediction model according to the factor-power influence correlation degree; the method comprises the steps of carrying out real-time data acquisition on a power limiting factor of a photovoltaic power station through a data acquisition device, and acquiring a real-time data acquisition result; and inputting the real-time data acquisition result into a photovoltaic power prediction model to obtain predicted power data.
The multi-dimensional photovoltaic power prediction system of the present invention comprises: the basic information acquisition module is used for acquiring basic information of the photovoltaic power station; the historical data calling module is used for calling historical data of the photovoltaic power station based on a preset time period to acquire historical data information; the limiting factor acquisition module is used for acquiring a first power limiting factor, a second power limiting factor and a third power limiting factor according to the basic information and the historical data information of the photovoltaic power station; the correlation calculation module is used for calculating influence correlation of the first power limiting factor, the second power limiting factor and the third power limiting factor on the power of the photovoltaic power station respectively, and obtaining factor-power influence correlation; the prediction model construction module is used for constructing a photovoltaic power prediction model according to the factor-power influence correlation degree; the real-time data acquisition module is used for acquiring real-time data of the power limiting factor of the photovoltaic power station through the data acquisition device and acquiring a real-time data acquisition result; and the predicted power data acquisition module is used for inputting the real-time data acquisition result into the photovoltaic power prediction model to obtain predicted power data.
Example 1
As shown in fig. 1, the multi-dimensional photovoltaic power prediction method includes:
step S100: acquiring basic information of a photovoltaic power station;
the photovoltaic power prediction system is in communication connection with the data acquisition device, and the data acquisition device is used for carrying out real-time data acquisition on the generated power of the photovoltaic power station.
Firstly, the basic information of the photovoltaic power station comprises various factor information influencing the output power of the photovoltaic power station, wherein the main factor is external environment factors, the types of the related external environment factors of the photovoltaic power generation power are numerous, the types and the number of weather factors which can be collected and recorded in a designated area and a time range are limited by actual conditions, and according to the photovoltaic power generation principle, the weather factors influencing the photovoltaic power generation power mainly comprise: irradiance, component temperature, ambient temperature, wind speed, relative humidity, cloud cover, and barometric pressure. The corresponding data acquisition equipment is used for acquiring external environment data in real time, such as a sun tracking array, an inverter, a weather station, an anemometer, a hygrometer, an optical monitoring instrument and the like of an external environment. The method realizes the acquisition of the basic information of the photovoltaic power station and lays a foundation for the follow-up construction of the photovoltaic power prediction model.
Step S200: based on a preset time period, historical data of the photovoltaic power station is called, and historical data information is obtained;
the preset time period is set according to the condition of the photovoltaic power station and is used for limiting the time period for data retrieval of the photovoltaic power station. Because the influence of the environment on the photovoltaic power station is greatly different in different time periods, the influence of the temperature in summer and winter on the photovoltaic power station can be effectively avoided by setting the time period according to seasons, and meanwhile, the influence of the temperature on the photovoltaic power station can be clearly determined by comparing the two data sets in summer and winter. And according to the set time period, the historical data of the photovoltaic power station are retrieved through the work log, the corresponding data in the same period are used as the same group of data, and one group of historical data comprises the illumination intensity, the air temperature, the sunshine time and the like of the historical photovoltaic power station together with the corresponding historical power data.
Step S300: acquiring a first power limiting factor, a second power limiting factor and a third power limiting factor according to the basic information and the historical data information of the photovoltaic power station;
comparing the same period data with different period data in the historical data information, sequencing the influences of all environmental factors on the power of the photovoltaic power station, and obtaining the first three factors with the largest influence, wherein the first three factors are respectively used as a first power limiting factor, a second power limiting factor and a third power limiting factor. The second power limiting factor is the air temperature, i.e. the degree of cooling and heating of the air, which can characterize the thermal condition of a place. The third power limiting factor is sunlight time, i.e. the accumulated time length of receiving illumination in one period of the photovoltaic panel, which is generally influenced by environment and climate.
Step S400: performing influence correlation calculation on the power of the photovoltaic power station on the first power limiting factor, the second power limiting factor and the third power limiting factor respectively to obtain factor-power influence correlation;
and because the influence of each item of data in the first power limiting factor, the second power limiting factor and the third power limiting factor on the power is different, acquiring the correlation coefficient between each item of data and the power by adopting a control variable method.
Illustratively, a solar cell utilizes the photovoltaic effect to convert solar radiation energy into electrical energy, irradiance being taken as indicative of solar radiationThe physical quantity of the radiation intensity, defined as the radiant energy flux impinging on a surface element at a point divided by the area of the element, is measured in units of
. Clearly, irradiance is the most direct meteorological factor that determines the output of a photovoltaic module, and the random and periodic changes of irradiance cause the photovoltaic power to exhibit obvious intermittence and fluctuation. In general, the larger the irradiance, the larger the photovoltaic power generation power, and thus, the output power of the photovoltaic module is substantially proportional to the irradiance. Wherein the influence correlation coefficient r is calculated as follows:
in the middle of
、/>
The ith data point for variables x, y; />
、/>
N is the number of data points, which is the mean of the variables x, y. The scatter diagram of the photovoltaic power generation power and irradiance has a regular shape, a certain proportional corresponding relation, r is larger, and the irradiance and the photovoltaic power generation power have stronger correlation.
The influence correlation coefficient between other factors and power, i.e. factor-power influence correlation, is calculated in the same way.
Step S500: constructing a photovoltaic power prediction model according to the factor-power influence correlation degree;
the photovoltaic power prediction model is a model for predicting the photovoltaic power station power according to a plurality of limiting factors, 80% of historical data information is randomly extracted as a training set to be used for training the model according to a factor-power mapping relation, 10% of historical data information is randomly extracted as a verification set to verify the accuracy of the model, an optimal training model is obtained through continuous verification, and the remaining 10% of historical data information is used as a test set to test the final effect of the model. By setting up the model, the output power is simply and efficiently calculated, and the effect of improving the accuracy and efficiency of power prediction is achieved.
Step S600: the method comprises the steps of carrying out real-time data acquisition on a power limiting factor of a photovoltaic power station through a data acquisition device, and acquiring a real-time data acquisition result;
the data acquisition device is equipment for acquiring power limiting factors of the photovoltaic power station in real time, data detection is carried out on the real-time illumination intensity, air temperature and sunlight time through the data acquisition equipment, a rectangular coordinate system is established, the time is taken as a horizontal axis, the illumination intensity, the air temperature and the sunlight time are taken as vertical axes, curves of the changes of the illumination intensity, the air temperature and the sunlight time along with time are drawn, the curves are taken as real-time data acquisition results, and when a certain node A needs to be predicted, the illumination intensity, the air temperature and the sunlight time data of a point t=A can be immediately acquired, and the curves are taken as input information of a photovoltaic power prediction model.
Step S700: and inputting the real-time data acquisition result into a photovoltaic power prediction model to obtain predicted power data.
The real-time data acquisition result is illumination intensity, air temperature and sunlight time information on the photovoltaic power station under the real-time node, the information is used as input data to be input into a photovoltaic power prediction model, based on factor-power influence correlation, each factor data in the obtained real-time data acquisition result is substituted, output power can be obtained, and the power in a future period of time can be quickly matched to the power in the current state through calculation of the model. The high-precision power prediction is realized, and the effect of improving the accuracy and efficiency of the power prediction of the photovoltaic power station is achieved.
Further, as shown in fig. 2, step S200 of the present invention further includes:
step S210: retrieving a plurality of sets of historical data based on the preset time period, wherein each set of historical data comprises a multi-dimensional data type;
step S220: carrying out data analysis on multiple groups of historical data to obtain information missing data points;
step S230: performing data missing inspection on the information missing data points to obtain a data missing inspection result;
step S240: and processing the plurality of groups of historical data according to the data missing checking result to acquire the historical data information.
The preset time period is the frequency for calling the historical data, which is set in advance according to the actual situation, such as four seasons and alternate sunny days and cloudy days. The multiple sets of historical data are data obtained after the historical data are called according to different preset time periods, wherein each set of historical data comprises multi-dimensional data types, and each multi-dimensional data type comprises illumination intensity, air temperature and illumination time, and the multi-dimensional data types and the corresponding historical power data.
The multi-dimensional data type is taken as a data identification index, and illustratively, the illumination intensity A, the air temperature B, the illumination time C and the corresponding power S randomly extract a group of data, the index is marked as uppercase, the index is not marked as lowercase, so that data analysis is carried out on multiple groups of historical data, the historical data marked as lowercase a is used for representing missing illumination intensity data, the missing information of each group is analyzed, whether the data identification index can be deduced and supplemented according to other information is judged, the temperature of the previous and later days is 40 ℃ high temperature, and the high probability of the missing data is 40 ℃. When the inference and the supplementation can not be carried out, the fact that the group of data lacks key information is indicated, and the key information cannot be continuously utilized is not utilized, so that the group of data is removed, and the accuracy of the data is ensured.
Further, step S240 of the present invention further includes:
step S241: performing feature analysis on the data missing inspection result to obtain a feature analysis result;
step S242: matching corresponding missing value processing methods according to the feature analysis result;
step S243: and processing a plurality of groups of historical data according to the missing value processing method.
Specifically, according to the data missing check result, if there is no missing data, other data preprocessing steps are continued, and if there is missing data, the steps can be classified as follows according to a missing mechanism: a completely random miss, i.e. a miss of data is completely random, does not depend on any incomplete or complete variable, does not affect the unbiasedness of the sample, simply speaking, the probability of data loss is completely independent of its hypothesized value and other variable values; random loss means that the probability of data loss is independent of the lost data itself, but only of the partially observed data, i.e. the absence of data is not completely random, the absence of such data being dependent on other complete variables; non-random misses, i.e. the absence of data is related to the value of the incomplete variable itself, in two cases, i.e. the missing value depends on its hypothesized value or the missing value depends on the other variable value. The deletion processing method comprises deleting data, interpolating data and not processing, and matching corresponding deletion value processing methods according to the characteristic analysis result, for example, for the time series problem, the deletion processing method can be used for supplementing the deletion data according to the front-back relation, and the interpolation method can be used for supplementing the deletion data.
Further, the step S500 of the present invention further includes:
step S510: taking the influence correlation degree of the first power limiting factor and the photovoltaic power station power as first training information to construct a first power prediction module;
step S520: taking the influence correlation degree of the second power limiting factor and the photovoltaic power station power as second training information to construct a second power prediction module;
step S530: taking the influence correlation degree of the third power limiting factor and the photovoltaic power station power as third training information, and constructing a third power prediction module;
step S540: and constructing a photovoltaic power prediction model based on the first power prediction module, the second power prediction module and the third power prediction module.
According to the mapping relation between the first power limiting factor and the first power, randomly extracting 80% of a data set corresponding to the first power limiting factor in the historical data information as a training set for training a model, randomly extracting 10% of the data set as a verification set for verifying the accuracy of the model, continuously verifying to obtain an optimal training model, and testing the final effect of the model by using the remaining 10% as a testing set to construct the first power prediction module. And acquiring a second power prediction module and a third power prediction module by using the same method. The first power prediction module, the second power prediction module and the third power prediction module are distributed in parallel, and the data can be classified quickly when input and then calculated according to the corresponding modules. The effect of improving the calculation efficiency of the model is achieved.
Further, the invention also comprises:
step S810: acquiring a prediction period, and generating a power recording node according to the prediction period;
step S820: when the photovoltaic power station runs to a power recording node, recording the real-time power generation power of the photovoltaic power station, and acquiring real-time power data;
step S830: and comparing the predicted power data with the real-time power generation power data to obtain a power prediction evaluation result.
Dividing the period to be predicted into k stages, dividing each stage into m nodes, calculating the predicted power corresponding to each node, namely predicted power data, through a photovoltaic power prediction model, recording the actual power corresponding to each node, namely real-time power data when a photovoltaic power station runs to each node, comparing the calculated predicted power data with the real-time power data recorded in real time, wherein the percentage of the ratio is the power prediction accuracy, taking the power prediction accuracy as a power prediction evaluation result, obtaining the target power prediction accuracy, and determining whether the photovoltaic power prediction model meets the requirement or not through the comparison of the power prediction accuracy and the target power prediction accuracy.
Further, the invention also comprises:
step S840: acquiring a preset power prediction threshold value;
step S850: and when the power prediction evaluation result does not meet the power prediction threshold value, optimizing the photovoltaic power prediction model according to the power prediction evaluation result.
And setting a power prediction threshold value for limiting a power prediction evaluation result, if the power prediction accuracy is smaller than the power prediction threshold value, describing that the model is accurate to calculate, and not needing to be optimized, if the power prediction accuracy is larger than the power prediction threshold value, describing that the calculation error is large, acquiring an optimization instruction, taking the difference value between the power prediction threshold value and the power prediction evaluation result and the power prediction threshold value as a feedback target, adjusting training data according to a factor generating deviation, and training again, so that the photovoltaic power prediction model is optimized. The effect of continuously optimizing and improving the accuracy of the prediction model is achieved.
Example two
Based on the same inventive concept as the multi-dimensional photovoltaic power prediction method in the first embodiment, as shown in fig. 3, the present invention provides a multi-dimensional photovoltaic power prediction system, which includes:
the basic information acquisition module 10 is used for acquiring basic information of the photovoltaic power station;
the historical data calling module 20 is used for calling historical data of the photovoltaic power station based on a preset time period to acquire historical data information;
the limiting factor obtaining module 30 is configured to obtain a first power limiting factor, a second power limiting factor, and a third power limiting factor according to the photovoltaic power station basic information and the historical data information;
the correlation calculation module 40 is configured to perform influence correlation calculation on the photovoltaic power station power for the first power limiting factor, the second power limiting factor and the third power limiting factor, so as to obtain a factor-power influence correlation;
a prediction model construction module 50, wherein the prediction model construction module 50 is used for constructing a photovoltaic power prediction model according to the factor-power influence correlation degree;
the real-time data acquisition module 60 is used for acquiring real-time data of the power limiting factor of the photovoltaic power station through the data acquisition device, and acquiring a real-time data acquisition result;
the predicted power data acquisition module 70 is configured to input a real-time data acquisition result to the photovoltaic power prediction model to obtain predicted power data.
Further, the invention also comprises:
the historical data calling module is used for calling a plurality of groups of historical data based on a preset time period, wherein each group of historical data comprises a multi-dimensional data type;
the data analysis module is used for carrying out data analysis on a plurality of groups of historical data to obtain information missing data points;
the data missing checking module is used for carrying out data missing checking on the information missing data points and obtaining a data missing checking result;
and the historical data processing module is used for processing a plurality of groups of historical data according to the data missing checking result to acquire historical data information.
Further, the system further comprises:
the feature analysis module is used for carrying out feature analysis on the data missing inspection result to obtain a feature analysis result;
the processing method matching module is used for matching corresponding missing value processing methods according to the characteristic analysis result;
and the data processing module is used for processing a plurality of groups of historical data according to the missing value processing method.
Further, the system further comprises:
the first power prediction module construction module is used for constructing a first power prediction module by taking the influence correlation degree of the first power limiting factor and the power of the photovoltaic power station as first training information;
the second power prediction module construction module is used for constructing a second power prediction module by taking the influence correlation degree of the second power limiting factor and the power of the photovoltaic power station as second training information;
the third power prediction module construction module is used for constructing a third power prediction module by taking the influence correlation degree of the third power limiting factor and the power of the photovoltaic power station as third training information;
and the photovoltaic power prediction model construction module is used for constructing a photovoltaic power prediction model based on the first power prediction module, the second power prediction module and the third power prediction module.
Further, the system further comprises:
the power record node generation module is used for acquiring a prediction period and generating a power record node according to the prediction period;
the real-time power generation power recording module is used for recording the real-time power generation power of the photovoltaic power station when the photovoltaic power station runs to the power recording node, and acquiring real-time power data;
the power prediction evaluation result acquisition module is used for comparing the prediction power data with the real-time power generation power data to obtain a power prediction evaluation result.
Further, the system further comprises:
the preset power prediction threshold value acquisition module is used for acquiring a preset power prediction threshold value;
and the model optimization module is used for optimizing the photovoltaic power prediction model according to the power prediction evaluation result when the power prediction evaluation result does not meet the power prediction threshold value.
Through the foregoing detailed description of a multi-dimensional photovoltaic power prediction method, those skilled in the art can clearly know a multi-dimensional photovoltaic power prediction method and a system in this embodiment, and for the device disclosed in the embodiment, the description is relatively simple because it corresponds to the method disclosed in the embodiment, and the relevant points refer to the description of the method section.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. The multi-dimensional photovoltaic power prediction method is characterized by comprising the following steps of:
(1) Acquiring basic information of a photovoltaic power station;
(2) Based on a preset time period, historical data of the photovoltaic power station is called, and historical data information is obtained;
(3) Acquiring a first power limiting factor, a second power limiting factor and a third power limiting factor according to the basic information and the historical data information of the photovoltaic power station;
(4) Performing influence correlation calculation on the power of the photovoltaic power station on the first power limiting factor, the second power limiting factor and the third power limiting factor respectively to obtain factor-power influence correlation;
(5) Constructing a photovoltaic power prediction model according to the factor-power influence correlation degree;
(6) The method comprises the steps of carrying out real-time data acquisition on a power limiting factor of a photovoltaic power station through a data acquisition device, and acquiring a real-time data acquisition result;
(7) And inputting the real-time data acquisition result into a photovoltaic power prediction model to obtain predicted power data.
2. The multi-dimensional photovoltaic power prediction method according to claim 1, wherein: the step (1) comprises the following steps:
retrieving a plurality of sets of historical data based on a preset time period, wherein each set of historical data comprises a multi-dimensional data type;
carrying out data analysis on multiple groups of historical data to obtain information missing data points;
performing data missing inspection on the information missing data points to obtain a data missing inspection result;
and processing the plurality of groups of historical data according to the data missing checking result to acquire historical data information.
3. The multi-dimensional photovoltaic power prediction method according to claim 2, wherein: the processing of the plurality of groups of historical data according to the data missing checking result, obtaining historical data information, comprises the following steps:
performing feature analysis on the data missing inspection result to obtain a feature analysis result;
matching corresponding missing value processing methods according to the feature analysis result;
and processing a plurality of groups of historical data according to the missing value processing method.
4. The multi-dimensional photovoltaic power prediction method according to claim 1, wherein: the step (5) comprises the following steps:
taking the influence correlation degree of the first power limiting factor and the photovoltaic power station power as first training information, and constructing a first power prediction module;
taking the influence correlation degree of the second power limiting factor and the photovoltaic power station power as second training information, and constructing a second power prediction module;
taking the influence correlation degree of the third power limiting factor and the photovoltaic power station power as third training information, and constructing a third power prediction module;
and constructing the photovoltaic power prediction model based on the first power prediction module, the second power prediction module and the third power prediction module.
5. The multi-dimensional photovoltaic power prediction method according to claim 1, wherein: the method also comprises the following steps:
acquiring a prediction period, and generating a power recording node according to the prediction period;
when the photovoltaic power station runs to the power recording node, recording the real-time power generation power of the photovoltaic power station, and acquiring real-time power generation power data;
and comparing the predicted power data with the real-time power generation power data to obtain a power prediction evaluation result.
6. The multi-dimensional photovoltaic power generation method according to claim 5, wherein: the method also comprises the following steps:
acquiring a preset power prediction threshold value;
and when the power prediction evaluation result does not meet the power prediction threshold value, optimizing the photovoltaic power prediction model according to the power prediction evaluation result.
7. A multi-dimensional photovoltaic power prediction system, characterized by: the system comprises a basic information acquisition module (10), a historical data acquisition module (20), a restriction factor acquisition module (30), a correlation calculation module (40), a prediction model construction module (50), a real-time data acquisition module (60) and a prediction power data acquisition module (70), wherein the basic information acquisition module (10) and the historical data acquisition module (20) are respectively connected with the restriction factor acquisition module (30), the correlation calculation module (40) and the prediction model construction module (50) are sequentially connected, and the output ends of the prediction model construction module (50) and the real-time data acquisition module (60) are respectively connected with the prediction power data acquisition module (70).
8. The multi-dimensional photovoltaic power generation system of claim 7, wherein: the basic information acquisition module (10) comprises a historical data acquisition module, a data analysis module, a data missing detection module and a historical data processing module which are sequentially connected.
9. The multi-dimensional photovoltaic power generation system of claim 8, wherein: the historical data processing module comprises a characteristic analysis module, a processing method matching module and a data processing module which are sequentially connected.
10. The multi-dimensional photovoltaic power generation system of claim 7, wherein: the prediction model building module (50) comprises a first power prediction module building module, a second power prediction module building module, a third power prediction module building module and a photovoltaic power prediction model building module, and the first power prediction module building module, the second power prediction module building module and the third power prediction module building module are respectively connected with the photovoltaic power prediction model building module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211681975.6A CN116050592A (en) | 2022-12-27 | 2022-12-27 | Multi-dimensional photovoltaic power prediction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211681975.6A CN116050592A (en) | 2022-12-27 | 2022-12-27 | Multi-dimensional photovoltaic power prediction method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116050592A true CN116050592A (en) | 2023-05-02 |
Family
ID=86122909
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211681975.6A Pending CN116050592A (en) | 2022-12-27 | 2022-12-27 | Multi-dimensional photovoltaic power prediction method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116050592A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116756528A (en) * | 2023-08-18 | 2023-09-15 | 杭州鸿晟电力设计咨询有限公司 | Method, device, equipment and medium for complementing electricity load data |
-
2022
- 2022-12-27 CN CN202211681975.6A patent/CN116050592A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116756528A (en) * | 2023-08-18 | 2023-09-15 | 杭州鸿晟电力设计咨询有限公司 | Method, device, equipment and medium for complementing electricity load data |
| CN116756528B (en) * | 2023-08-18 | 2023-11-28 | 杭州鸿晟电力设计咨询有限公司 | Method, device, equipment and medium for complementing electricity load data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111275367B (en) | Regional comprehensive energy system energy efficiency state evaluation method | |
| CN110909919A (en) | Photovoltaic power prediction method of depth neural network model with attention mechanism fused | |
| CN113657662B (en) | Downscaling wind power prediction method based on data fusion | |
| CN109272258B (en) | Regional wind and solar power generation resource evaluation method based on K-means clustering | |
| CN109902340A (en) | It is a kind of meter and complicated meteorology coupled characteristic multi-source-lotus combine scene generating method | |
| CN111509768A (en) | Scene simulation method and system of comprehensive energy system | |
| CN116632823A (en) | Short-term photovoltaic power prediction method based on power conversion model and multi-layer perceptron | |
| CN114399081A (en) | Photovoltaic power generation power prediction method based on weather classification | |
| CN111985678A (en) | Photovoltaic power short-term prediction method | |
| Aliberti et al. | Forecasting Short-term Solar Radiation for Photovoltaic Energy Predictions. | |
| CN116050592A (en) | Multi-dimensional photovoltaic power prediction method and system | |
| CN115204444A (en) | Photovoltaic power prediction method based on improved cluster analysis and fusion integration algorithm | |
| CN111275238B (en) | Large-scale power station photovoltaic output sequence generation method based on hourly clear sky index | |
| CN114240003A (en) | New energy output prediction method, system, storage medium and equipment | |
| CN115660232A (en) | Ultra-short-term prediction method, device and system for wind power | |
| CN113449847B (en) | Offshore wind power rolling prediction method considering second-level time series wind speed change | |
| Zheng et al. | Short-term photovoltaic power prediction based on daily feature matrix and deep neural network | |
| Matushkin et al. | Analysis of factors for forecasting electric power generation by solar power plants | |
| CN113449466B (en) | Solar radiation prediction method and system for optimizing RELM based on PCA and chaos GWO | |
| CN110796292A (en) | Photovoltaic power short-term prediction method considering haze influence | |
| CN115587664A (en) | Solar radiation prediction method and system based on graph convolution long-term and short-term memory network | |
| CN111027816B (en) | Photovoltaic power generation efficiency calculation method based on data envelope analysis | |
| CN113962432A (en) | Wind power prediction method and system integrating three-dimensional convolution and light-weight convolution threshold unit | |
| CN117828488B (en) | Solar radiance prediction method based on random forest and robust regression | |
| CN116502070B (en) | Intelligent monitoring system for state of miniature wind turbine generator |
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 |