CN114169229A - Data-driven identification method and device for optimizing cleaning time of photovoltaic array - Google Patents
Data-driven identification method and device for optimizing cleaning time of photovoltaic array Download PDFInfo
- Publication number
- CN114169229A CN114169229A CN202111405477.4A CN202111405477A CN114169229A CN 114169229 A CN114169229 A CN 114169229A CN 202111405477 A CN202111405477 A CN 202111405477A CN 114169229 A CN114169229 A CN 114169229A
- Authority
- CN
- China
- Prior art keywords
- model
- photovoltaic
- cleaning
- hunting
- photovoltaic array
- 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 110
- 238000004140 cleaning Methods 0.000 title claims abstract description 109
- 239000000428 dust Substances 0.000 claims abstract description 69
- 241000282994 Cervidae Species 0.000 claims abstract description 57
- 230000008569 process Effects 0.000 claims abstract description 54
- 238000005457 optimization Methods 0.000 claims abstract description 47
- 238000010248 power generation Methods 0.000 claims abstract description 41
- 230000008021 deposition Effects 0.000 claims abstract description 34
- 238000004458 analytical method Methods 0.000 claims abstract description 17
- 238000011156 evaluation Methods 0.000 claims abstract description 13
- 238000009825 accumulation Methods 0.000 claims abstract description 12
- 238000004445 quantitative analysis Methods 0.000 claims abstract description 10
- 239000013598 vector Substances 0.000 claims description 33
- 230000008859 change Effects 0.000 claims description 16
- 238000005070 sampling Methods 0.000 claims description 12
- 239000004071 soot Substances 0.000 claims description 12
- XOFYZVNMUHMLCC-ZPOLXVRWSA-N prednisone Chemical compound O=C1C=C[C@]2(C)[C@H]3C(=O)C[C@](C)([C@@](CC4)(O)C(=O)CO)[C@@H]4[C@@H]3CCC2=C1 XOFYZVNMUHMLCC-ZPOLXVRWSA-N 0.000 claims description 7
- 230000005611 electricity Effects 0.000 claims description 5
- 230000007613 environmental effect Effects 0.000 claims description 4
- 230000005855 radiation Effects 0.000 claims description 4
- 238000012360 testing method Methods 0.000 claims description 4
- 230000000007 visual effect Effects 0.000 claims description 4
- 230000008901 benefit Effects 0.000 abstract description 7
- 238000011002 quantification Methods 0.000 abstract description 3
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 4
- 239000003795 chemical substances by application Substances 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 229910052760 oxygen Inorganic materials 0.000 description 4
- 239000001301 oxygen Substances 0.000 description 4
- 238000011161 development Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000001960 triggered effect Effects 0.000 description 3
- 238000005406 washing Methods 0.000 description 3
- 241000282414 Homo sapiens Species 0.000 description 2
- 230000036626 alertness Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 230000001186 cumulative effect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000003628 erosive effect Effects 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 244000062645 predators Species 0.000 description 2
- 230000001737 promoting effect Effects 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011109 contamination Methods 0.000 description 1
- 239000006059 cover glass Substances 0.000 description 1
- 239000005357 flat glass Substances 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Water Supply & Treatment (AREA)
- Tourism & Hospitality (AREA)
- Public Health (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Human Resources & Organizations (AREA)
- Artificial Intelligence (AREA)
- Primary Health Care (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a data-driven identification method and device for optimizing the cleaning time of a photovoltaic array, and belongs to the field of optimizing the cleaning time of the photovoltaic array. The method of the invention comprises the following steps: collecting relevant data of the accumulated dust density and the output power of the photovoltaic array under a preset condition; based on the relevant data, performing principle analysis and process description on the model identification of the hunting deer optimization algorithm; the model parameter identification method based on the hunting deer optimization algorithm is used for respectively identifying the photovoltaic system dust accumulation attenuation model and the peak hour number prediction model; and carrying out rationality evaluation and quantitative analysis on the predicted time of the dust deposition cleaning of the photovoltaic array. According to the invention, the optimal cleaning time interval can be indirectly obtained through the established photovoltaic system dust accumulation attenuation model and the peak hour prediction model, and a reference is provided for dynamic optimization of cleaning time. The power generation benefit changes of the photovoltaic system before and after cleaning are compared through benefit quantification so as to effectively evaluate the rationality of the obtained cleaning time scheme.
Description
Technical Field
The invention belongs to the field of photovoltaic array cleaning time optimization, and particularly relates to a data-driven identification method and device for photovoltaic array cleaning time optimization.
Background
With the transformation of energy worldwide, solar energy is actively used in a plurality of fields such as power generation and heating, and the application of the solar energy in the power generation process shows a trend of increasing prosperity, and the attention of various countries to the solar energy is gradually increased. The focus of photovoltaic systems is on improving the power generation efficiency of photovoltaic arrays or restoring their efficiency losses due to various internal and external factors. In this context, as one of the main causes affecting the energy efficiency of the photovoltaic system, a phenomenon of dust deposition on the surface of the cover glass of the photovoltaic module or contamination of the plate glass becomes a problem which cannot be ignored. The economic loss caused by the change of the problems of the dust deposition of the photovoltaic array and the like along with the time needs to be evaluated, and then the reasonable cleaning time interval is determined. However, existing studies have had limited exploration for this problem. The importance of the method in improving the efficiency and economy of photovoltaic power generation is yet to be further explored.
The performance of a photovoltaic module is affected by factors such as conversion efficiency, weather patterns, soil, photovoltaic tilt angle, photovoltaic cell temperature, ambient temperature and humidity. In order to relieve the influence of the problems of dust deposition or pollution on the surface of the component caused by the operation of the component in the outdoor environment, the cleaning and maintenance are key means for ensuring the safe and efficient operation of the component. The choice of the cleaning interval is an important factor in determining the cost of operating and maintaining the cleaning chamber. Although the generated energy can be effectively improved through frequent cleaning operation, the generated cost of water, electricity, equipment, labor and the like can directly reduce the income and even avoid the condition of reverse lifting. On the contrary, if the cleaning interval is too long, the power generation efficiency of the photovoltaic system is directly influenced, and further the income is reduced. Therefore, photovoltaic array cleaning is essentially an economic optimization problem that accounts for power generation and cleaning costs. Although the methods used to optimize the photovoltaic array cleaning time or cycle are different, the prediction of cleaning time and optimization of benefits is mostly achieved by modeling the deposition rate and cleaning cost.
Therefore, in consideration of the rapid development of big data technology, the invention provides a data-driven identification method and device for optimizing the cleaning time of a photovoltaic array, wherein the data-driven identification method has great potential in the attempt and application of the method in the optimization of the cleaning time of the photovoltaic array.
Disclosure of Invention
The present invention is directed to at least one of the technical problems of the prior art, and provides a method and an apparatus for data-driven identification with optimized photovoltaic array cleaning time.
In one aspect of the present invention, a data-driven identification method for optimizing a photovoltaic array cleaning time is provided, which includes the following specific steps:
collecting relevant data of the accumulated dust density and the output power of the photovoltaic array under a preset condition;
performing principle analysis and process description on model identification of the hunting deer optimization algorithm based on the related data;
the model parameter identification method based on the hunting deer optimization algorithm is used for respectively identifying the photovoltaic system dust accumulation attenuation model and the peak hour number prediction model;
and carrying out rationality evaluation and quantitative analysis on the predicted time of the dust deposition cleaning of the photovoltaic array.
Optionally, the acquiring data related to the accumulated dust density and the output power of the photovoltaic array under the preset condition includes:
collecting data related to the accumulated dust density and the output power of the photovoltaic system by taking weather, environment and terrain factors as references;
and selecting a plurality of climate, environment and terrain condition combinations for data acquisition.
Optionally, the performing principle analysis and process description on model identification of the hunting deer optimization algorithm based on the relevant data includes:
establishing a preset form for the model to be identified;
regarding the optimal position of the hunting deer as the optimal model parameter vector theta*Identifying and optimizing a hunting deer optimization algorithm model;
obtaining an optimal hunting position H according to the hunting deer optimization algorithm model identificationbestAnd theta*＝Hbest。
Optionally, the optimal position of the hunting deer is regarded as the optimal model parameter vector theta*And optimizing the model identification of the optimization algorithm of the hunting deer, comprising the following steps:
taking s hunters as a population H, and randomly initializing the hunter population to H ═ H1,H2,…,Hs}；
Determining an initial wind direction angle and a position angle of a hunting male deer;
calculating to obtain a position closest to the optimal solution according to the fitness function;
and stopping position updating of the hunter after the optimal hunting position is determined.
Optionally, the determination of the initialized wind direction angle and the position angle of the hunting male deer is obtained by adopting the following formula:
φk＝2πl
wherein phi represents a wind direction angle,
represents a position angle; l is [0,1]]The random number in the range, k, represents the current iteration step.
Optionally, the calculating the position closest to the optimal solution according to the fitness function includes:
the propagation process based on the leader's location is such that when an optimal location is determined, each hunter in the population attempts to reach the optimal location, at which point a location update process is triggered, wherein hunter surround behavior can be represented by:
Hk+1＝Hlead-M·c|R×Hlead-Hk|
wherein: hleadRepresenting the position of the leader, c is a random number considering the wind direction angle, and the value range of c is (0, 2)]M and R both refer to parameter vectors;
the propagation process based on the position angle is realized by considering the position angle in the position updating process of hunters
To improve the search space, in the hunting process of deer, a new parameter ds is introduced, assuming that the hunting process is valid in a certain position anglekAn update of the position angle is made and the parameter is developed from the variance between the wind angles, the viewing angle of which is as follows:
dsk＝φk-vsk
and, visual angle vs of preykThe following formula:
wherein rd is a random number taken from [0,1], and the parameter a is used for measuring a coefficient vector in an equation;
the position angle is updated by using the following relational expression:
updating the hunter location by calling using the following relationship:
Hk+1＝Hlead-c|cos(w)×Hlead-Hk|；
wherein: hk+1To use hunter position vector H at time kkThe hunter position at the updated k +1 moment, w represents a position angle;
propagation process based on the position of the successor: then, assuming that the value of the vector R is less than 1, the updating process of the hunter position depends on the position of the successor, as follows:
Hk+1＝Hsuccessor-M·c|R×Hsuccessor-Hk|
wherein HsuccessorRepresenting the location of the successor.
Optionally, the model parameter identification method based on the hunting deer optimization algorithm is used for identifying a photovoltaic system soot deposition attenuation model and a peak hour number prediction model respectively, and includes:
selecting the time of cleaning the photovoltaic module as a sampling starting point under different seasons and weather conditions, and establishing a power attenuation trend model capable of accurately describing the time of the photovoltaic module along with dust deposition by taking the change condition of rated peak power of the photovoltaic module along with the time lapse after cleaning as a reference;
and selecting environmental temperature, air humidity, cloud layer thickness and air density data from the sampling data as modeling input, taking solar irradiance as output, identifying unknown parameters of the model through a DHO algorithm based on a preset model form, and calculating the peak hours.
Optionally, the following relation is adopted for calculating the peak hour number:
wherein, the peak hour number H refers to the condition that the total solar radiation in a certain time period is converted into standard test, namely the irradiance is 1000W/m2At a temperature of 25 DEG CThe number of hours of (d);
t1and t2Respectively, a start time point and an end time point of the calculated time period, and r (t) is the irradiance of the photovoltaic array at time t.
Optionally, the performing rationality evaluation and quantitative analysis on the predicted time for cleaning the accumulated dust of the photovoltaic array includes:
drawing a rated peak power change curve of the following photovoltaic system after dust deposition and cleaning, and obtaining a power generation increment delta Q generated by the photovoltaic system after cleaning;
the method comprises the following steps of quantitatively calculating the yield increment delta E of the power station, caused by cleaning operation, for photovoltaic power generation through a formula as follows:
ΔE＝Eclean-Edust-C
Eclean＝r×Qclean
Edust＝r×Qdust
wherein E iscleanAnd EdustRespectively indicating the power generation income of whether the photovoltaic array is cleaned, C is the cleaning cost, QcleanRepresenting the t of the photovoltaic array over the Tc period1Cumulative power generation amount, Q, cleaned at all timesdustThe accumulated power generation amount of the photovoltaic array which is not cleaned in the Tc time period is represented, and r represents the grid-connected electricity price of photovoltaic power generation;
and, whether Δ E is larger than 0 is judged to obtain the rationality of the cleaning time and to seek the optimum cleaning time interval.
In another aspect of the present invention, a data-driven identification device for optimizing a cleaning time of a photovoltaic array is provided, which includes: the device comprises an acquisition module, an analysis module, an identification module and an evaluation module; wherein,
the acquisition module is used for acquiring relevant data of the accumulated dust density and the output power of the photovoltaic array under a preset condition;
the analysis module is used for carrying out principle analysis and process description on the model identification of the hunting deer optimization algorithm based on the relevant data;
the identification module is used for identifying a photovoltaic system dust accumulation attenuation model and a peak hour number prediction model respectively based on the model parameter identification method of the hunting deer optimization algorithm;
and the evaluation module is used for carrying out rationality evaluation and quantitative analysis on the predicted time of the accumulated dust cleaning of the photovoltaic array.
The invention aims to provide a data-driven identification method for optimizing the cleaning time of a photovoltaic array, so as to predict the optimal cleaning time of a photovoltaic system and improve the power generation efficiency and the economy of the photovoltaic system. The method constructs a data-driven model identification method through a hunting deer optimization algorithm and uses the data-driven model identification method in modeling of a photovoltaic system dust accumulation attenuation model and a peak hour number prediction model. Meanwhile, the feasibility and the rationality of the cleaning strategy are evaluated by taking cost constraint into consideration and taking economic benefit as a performance index in the process. Effective reference is provided for obtaining the optimal cleaning time of the photovoltaic array, and the method plays a great role in promoting the economic, efficient, safe and stable operation of the photovoltaic power generation system.
Drawings
FIG. 1 is a schematic diagram of a data-driven identification method for optimizing a cleaning time of a photovoltaic array according to an embodiment of the present invention;
FIG. 2 is a block diagram of a data-driven identification method for optimizing a cleaning time of a photovoltaic array according to another embodiment of the present invention;
FIG. 3 is a plot of peak power rating of a photovoltaic system after ash deposition and cleaning in accordance with another embodiment of the present invention;
FIG. 4 is a schematic diagram of a data-driven identification device for optimizing a cleaning time of a photovoltaic array according to another embodiment of the present invention.
Detailed Description
In order to make the technical solutions of the present invention better understood, the present invention will be described in further detail with reference to the accompanying drawings and specific embodiments. It is to be understood that the embodiments described are only a few embodiments of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the described embodiments of the invention without any inventive step, are within the scope of protection of the invention.
As shown in fig. 1 and fig. 2, in one aspect of the present invention, a data-driven identification method S100 for optimizing a cleaning time of a photovoltaic array is provided, which includes the following steps S110 to S140:
and S110, collecting relevant data of the accumulated dust density and the output power of the photovoltaic array under a preset condition.
It should be noted that different climatic, environmental and topographic factors can cause large differences in air humidity, oxygen content and dust density, which in turn affects the dust deposition of the photovoltaic array. For example, in the season that the rainwater is more, the dust on the surface of the photovoltaic array can be effectively removed through the erosion of the rainwater to achieve the cleaning effect, and the power generation efficiency of the photovoltaic system is effectively improved at the moment. Based on this, step S110 can be embodied as S1101 to S1102 described below.
And S1101, collecting data related to the accumulated dust density and the output power of the photovoltaic system by taking the climate, environment and terrain factors as references.
And S1102, selecting a plurality of climate, environment and terrain condition combinations for data acquisition in order to ensure the comprehensiveness of the sampled data. If the variable which can not be directly measured exists in the accumulated dust density and the output power of the photovoltaic system, sampling data related to the variable, and calculating to obtain an indirectly measured target. The sampling step length is set to be T, the number of climate, environment and terrain condition combinations is set to be N, the number of data vectors sampled in each combination is set to be N, and the dimensionality of the data vectors is set to be m.
Based on the data related to the output power and the deposition density of the photovoltaic system obtained in step S110, a model identification based on a hunting deer optimization (DHO) algorithm is specifically described in step S120.
And S120, performing principle analysis and process description on the model identification of the hunting deer optimization algorithm based on the related data.
Specifically, S1201, a preset form, such as a transfer function or subspace model form, is set for the model to be identified, and the parameter vector to be identified is set to be θ, the optimal value of which is θ*。
S1202, a hunting mode of the human beings for the deer provides a sense of inspiration for developing the DHO algorithm. A key objective of the DHO algorithm is to determine the best or effective deer hunterLocation. The male deer (i.e. deer) has certain specific features that protect itself from predators or hunters, such as excellent vision, extraordinary smell and alertness to ultra-high frequency sounds. Based on the above analysis, the optimal position of the hunting deer is regarded as the optimal model parameter vector theta*The model identification based on the DHO algorithm is mainly divided into four stages.
First, a population is initialized. Taking s hunters as a population H, and randomly initializing the hunter population to H ═ H1,H2,…,Hs}：
Second, the wind and position angle are initialized. The initialization of position and wind direction angle is considered as a key process for a hunter to determine the optimal ideal position to hunt a male deer. Wind direction angle phi and position angle based on circumference
Are given by the following equations, respectively:
φk＝2πl (1)
wherein: l is a random number in the range of [0,1], and k represents the current iteration step.
And thirdly, position propagation. In the initial stage, the position of the optimal or ideal space is uncertain, and therefore, the position closest to the optimal solution calculated according to the fitness function is regarded as the optimal space. Two location update schemes are generally considered, namely a successor location and a leader location.
Propagation process based on leader location: once the optimal location is determined, each hunter in the population attempts to reach the optimal location, at which point a location update process is triggered. Hunter surround behavior can be represented by the following formula:
Hk+1＝Hlead-M·c|R×Hlead-Hk| (3)
wherein: representing the position of the leader, c is a random number considering the wind direction angle, the value range of c is (0, 2), M and R refer to parameter vectors, and the calculation formula is as follows:
R＝2rd (5)
wherein: rd is derived from [0,1]Random number of (k)maxFor the maximum number of iterations, parameter a is used to measure the coefficient vector in the equation.
Position angle based propagation process: in this case, the position angle is taken into account in the position updating process of the hunter
To improve the search space. In the hunting process for deer, it is assumed that the hunting process is valid under a certain position angle. In addition, a new parameter ds is introducedkAn update of the position angle is made and this parameter is developed from the variance between the wind angles, the angle of visibility of which is expressed in equation (6). At the same time, the visual angle vs of the preykIs defined by formula (7).
dsk＝φk-vsk (6)
The position angle is updated by equation (8), and the hunter position is updated based on the position angle shown in equation (9):
Hk+1＝Hlead-c|cos(w)×Hlead-Hk| (9)
wherein: hk+1To use hunter position vector H at time kkThe hunter position at the updated time k + 1, w represents the position angle.
Propagation process based on the position of the successor: the bounding mechanism is used to change the vector R of the exploration phase. However, in the initial state, a random search is first considered, and then the value of the vector R is assumed to be less than 1. The hunter location update process depends on the successor location, as shown in equation (10). If the value of the vector R is less than 1, then the search agent is randomly selected, otherwise the optimal solution is selected to update the search agent's location.
Hk+1＝Hsuccessor-M·c|R×Hsuccessor-Hk| (10)
And fourthly, terminating the process. And when the optimal hunting position is determined, stopping position updating of the hunter, and ending the optimization process.
S1203, for the data-driven modeling of the present embodiment, the optimal hunting position H obtained at this timebestIs the optimal solution of the model parameter vector, i.e. θ*＝Hbest。
According to the DHO-based model parameter identification method designed in step S120, identification of the photovoltaic system soot deposition attenuation model and the peak hour number prediction model is performed in step S130.
S130, identifying a photovoltaic system dust accumulation attenuation model and a peak hour number prediction model respectively based on the model parameter identification method of the hunting deer optimization algorithm.
Specifically, S1301, identifying a photovoltaic system dust accumulation attenuation model. Selecting the time of the photovoltaic module which is just cleaned as a sampling starting point under different seasons and weather conditions, and taking the change situation of the rated peak power of the photovoltaic module along with the time lapse after the photovoltaic module is cleaned as a reference, thereby establishing a power attenuation trend model which can accurately describe the time of the photovoltaic module along with dust deposition, and providing a reference for reasonably planning the cleaning time of the photovoltaic module.
The gray deposition attenuation characteristic of the photovoltaic system is related to solar irradiance, photovoltaic module temperature and output power, so that corresponding variables are selected from collected data in the modeling process, the attenuation model form is preset, and the optimization identification of unknown parameters of the model is carried out through the DHO-based modeling method.
And S1302, identifying a peak hour prediction model. The peak hour number H refers to the solar radiation in a certain time periodThe total irradiation is converted into standard test conditions, i.e. the irradiance is 1000W/m2The hours at 25 ℃ can be generally calculated by the following formula:
wherein: t is t1And t2Respectively, a start time point and an end time point of the calculated time period, and r (t) is the irradiance of the photovoltaic array at time t.
As can be seen from the above equation, the number of peak hours is directly related to the solar irradiance, and therefore, the influence of the solar irradiance variation needs to be considered. Passing through t1～t2The change of the solar irradiance is closely related to the clear degree of the weather, the thickness of the cloud layer, the oxygen content of air, namely the humidity and the like.
Therefore, based on five weather types of sunny, cloudy, light rain and heavy to heavy rain, data such as ambient temperature, air humidity, cloud layer thickness and air density are selected from the sampling data to be used as modeling input, solar irradiance is output, and unknown parameters of the model are identified through a DHO algorithm based on a preset model form. The peak hours are then calculated using equation (11).
And S1303, obtaining the optimal interval of the cleaning time of the photovoltaic array by combining the economy of the photovoltaic system according to the established photovoltaic system soot deposition attenuation model and the peak hour prediction model.
S140, carrying out rationality assessment and quantitative analysis on the predicted time of accumulated dust cleaning of the photovoltaic array.
Specifically, in step S140, the photovoltaic array cleaning effectiveness is quantitatively analyzed to achieve a dynamic assessment of the resulting soot cleaning time rationality. First, the rated peak power change curves of the following photovoltaic systems after dust deposition and cleaning are plotted, as shown in fig. 3, wherein P is0Rated peak power before cleaning for accumulated dust, Tc represents cleaning time interval, t1Representing the specific time of a cleaning, as can be seen from FIG. 3, the photovoltaic system can produce Δ Q of the hair after cleaningAn electrical increment.
Based on this fig. 3, the change in the plant profit due to the cleaning operation of the plant is calculated quantitatively by a formula. Firstly, a power generation formula of a photovoltaic system is given:
Q(t)＝P(t)×H(t) (12)
wherein: h (t) represents the number of peak hours on day t calculated by equation (11).
If the photovoltaic array is not being cleaned during the Tc time period, the cumulative power generation can be expressed as:
if the photovoltaic array is at t within the Tc time period1The washing is carried out at any moment, and the accumulated power generation amount is changed into:
and r represents the grid-connected electricity price of the photovoltaic power generation, the obtained income E is as follows:
ΔE＝Eclean-Edust-C
Eclean＝r×Qclean (15)
Edust＝r×Qdust
wherein: ecleanAnd EdustAnd C, respectively representing the power generation income of whether the photovoltaic array is cleaned or not, wherein C is the cleaning cost. In this case, the rationality of the resulting cleaning time can be judged by whether Δ E is larger than 0 and the optimum cleaning time interval can be sought.
In order to predict the optimal cleaning time of a photovoltaic system to improve the power generation efficiency and economy of the photovoltaic system, the embodiment provides a data-driven identification method based on a hunting deer optimization algorithm, and the method takes the influences of different climates, environments and terrain factors on the dust density and the output power of a photovoltaic array into account and collects relevant data. Then, based on the rapid development of a big data technology and a swarm intelligence optimization algorithm, a data-driven model identification method fused with a hunting deer optimization algorithm is provided and is respectively applied to modeling of a photovoltaic system soot deposition attenuation model and a peak hour prediction model. The optimal cleaning time interval can be indirectly obtained through the established photovoltaic system dust accumulation attenuation model and the peak hour prediction model, and reference is provided for dynamic optimization of cleaning time. Finally, the power generation benefit changes of the photovoltaic system before and after cleaning are compared through benefit quantification, and the rationality of the obtained cleaning time scheme can be effectively evaluated.
As shown in FIG. 4, in another aspect of the present invention, a data-driven identification device 200 for optimizing the cleaning time of a photovoltaic array is provided, which includes: an acquisition module 210, an analysis module 220, a recognition module 230, and an evaluation module 240; the acquisition module 210 is configured to acquire data related to the accumulated dust density and the output power of the photovoltaic array under a preset condition; the analysis module 220 is configured to perform principle analysis and process description on model identification of the hunting deer optimization algorithm based on the relevant data; the identification module 230 is configured to identify a photovoltaic system soot deposition attenuation model and a peak hour prediction model respectively based on the model parameter identification method of the hunting deer optimization algorithm; the evaluation module 240 is used for performing rationality evaluation and quantitative analysis on the predicted time of the accumulated dust cleaning of the photovoltaic array.
It should be noted that the optimization process of the apparatus of the present embodiment on the cleaning time of the photovoltaic array is based on the method described above, and is not described herein again.
The data-driven identification device for optimizing the cleaning time of the photovoltaic array is used for predicting the optimal cleaning time of the photovoltaic system so as to improve the power generation efficiency and the economy of the photovoltaic system. The device constructs a data-driven model identification method through a hunting deer optimization algorithm and uses the data-driven model identification method in modeling of a photovoltaic system dust accumulation attenuation model and a peak hour number prediction model. Meanwhile, the feasibility and the rationality of the cleaning strategy are evaluated by taking cost constraint into consideration and taking economic benefit as a performance index in the process. Effective reference is provided for obtaining the optimal cleaning time of the photovoltaic array, and the method plays a great role in promoting the economic, efficient, safe and stable operation of the photovoltaic power generation system.
The following will specifically describe the data-driven identification method for optimizing the cleaning time of the photovoltaic array by using an embodiment:
referring to fig. 1, fig. 1 is a schematic diagram of a data driving identification method for optimizing a cleaning time of a photovoltaic array according to the present embodiment. The embodiment is completed by relying on a Matlab software platform, and specifically comprises the following 4 steps:
s1: collecting data related to the density and the output power of the photovoltaic array accumulated dust under different climates, environments and terrain factors;
s2: analyzing a model identification principle and describing a process based on a hunting deer optimization algorithm;
s3: identifying a photovoltaic system dust accumulation attenuation model and a peak hour number prediction model;
s4: and (4) evaluating the reasonability of the predicted time of the ash deposition cleaning and carrying out quantitative analysis.
Different climatic, environmental and topographic factors can cause great differences in air humidity, oxygen content and dust density, and further affect the dust deposition condition of the photovoltaic array. For example, in the season that the rainwater is more, the dust on the surface of the photovoltaic array can be effectively removed through the erosion of the rainwater to achieve the cleaning effect, and the power generation efficiency of the photovoltaic system is effectively improved at the moment. Based on this, step S1 can be embodied as:
s1.1: and collecting data related to the accumulated dust density and the output power of the photovoltaic system by taking climate, environment and terrain factors as references.
S1.2: in order to ensure the comprehensiveness of the sampled data, a plurality of climate, environment and terrain condition combinations are selected for data acquisition. If the variable which can not be directly measured exists in the accumulated dust density and the output power of the photovoltaic system, sampling data related to the variable, and calculating to obtain an indirectly measured target. And setting the sampling step length as T-3 min, the number of the climate, environment and terrain condition combinations as N-8, and the number of the data vectors sampled by each combination as N-8000.
Based on the data on the output power and the deposition density of the photovoltaic system obtained in step S1, model identification based on the hunting deer optimization (DHO) algorithm is specifically described in step S2.
S2.1: setting a preset form, such as a transfer function or a subspace model form, for the model to be identified, and setting the parameter vector to be identified as theta, wherein the optimal value of the parameter vector is theta*。
S2.2: the hunting way of the deer by the human provides a sense of inspiration for developing the DHO algorithm. A key goal of the DHO algorithm is to determine the optimal or effective location of the hunting deer. The male deer (i.e. deer) has certain specific features that protect itself from predators or hunters, such as excellent vision, extraordinary smell and alertness to ultra-high frequency sounds. Based on the above analysis, the optimal position of the hunting deer is regarded as the optimal model parameter vector theta*The model identification based on the DHO algorithm is mainly divided into four stages.
S2.2.1: and (5) initializing a population. Taking s hunters as a population H, and randomly initializing the hunter population to H ═ H1,H2,…,Hs}：
S2.2.2: wind and position angles are initialized. The initialization of position and wind direction angle is considered as a key process for a hunter to determine the optimal ideal position to hunt a male deer. Wind direction angle phi and position angle based on circumference
Are given by the following equations, respectively:
φk＝2πl (1)
wherein: l is a random number in the range of [0,1], and k represents the current iteration step.
S2.2.3: and (4) position propagation. In the initial stage, the position of the optimal or ideal space is uncertain, and therefore, the position closest to the optimal solution calculated according to the fitness function is regarded as the optimal space. Two location update schemes are generally considered, namely a successor location and a leader location.
Propagation process based on leader location: once the optimal location is determined, each hunter in the population attempts to reach the optimal location, at which point a location update process is triggered. Hunter surround behavior can be represented by the following formula:
Hk+1＝Hlead-M·c|R×Hlead-Hk| (3)
wherein: representing the position of the leader, c is a random number considering the wind direction angle, the value range of c is (0, 2), M and R refer to parameter vectors, and the calculation formula is as follows:
R＝2rd (5)
wherein: rd is derived from [0,1]Random number of (k)maxThe parameter a is used to measure the coefficient vector in the equation, here 0.1, for a maximum number of iterations of 50.
Position angle based propagation process: in this case, the position angle is taken into account in the position updating process of the hunter
To improve the search space. In the hunting process for deer, it is assumed that the hunting process is valid under a certain position angle. In addition, a new parameter ds is introducedkAn update of the position angle is made and this parameter is developed from the variance between the wind angles, the angle of visibility of which is expressed in equation (6). At the same time, the visual angle vs of the preykIs defined by formula (7).
dsk＝φk-vsk (6)
The position angle is updated by equation (8), and the hunter position is updated based on the position angle shown in equation (9):
Hk+1＝Hlead-c|cos(w)×Hlead-Hk| (9)
propagation process based on the position of the successor: the bounding mechanism is used to change the vector R of the exploration phase. However, in the initial state, a random search is first considered, and then the value of the vector R is assumed to be less than 1. The hunter location update process depends on the successor location, as shown in equation (10). If the value of the vector R is less than 1, then the search agent is randomly selected, otherwise the optimal solution is selected to update the search agent's location.
Hk+1＝Hsuccessor-M·c|R×Hsuccessor-Hk| (10)
S2.2.4: the process is terminated. And when the optimal hunting position is determined, stopping position updating of the hunter, and ending the optimization process.
S2.3: for the data-driven modeling of the present invention, the optimal hunting position H has been obtained at this timebestIs the optimal solution of the model parameter vector, i.e. θ*＝Hbest。
According to the DHO-based model parameter identification method designed in step S2, identification of the photovoltaic system soot deposition decay model and the peak hour number prediction model is performed in step S3.
S3.1: and identifying the gray deposition attenuation model of the photovoltaic system. Selecting the time of the photovoltaic module which is just cleaned as a sampling starting point under different seasons and weather conditions, and taking the change situation of the rated peak power of the photovoltaic module along with the time lapse after the photovoltaic module is cleaned as a reference, thereby establishing a power attenuation trend model which can accurately describe the time of the photovoltaic module along with dust deposition, and providing a reference for reasonably planning the cleaning time of the photovoltaic module.
The gray deposition attenuation characteristic of the photovoltaic system is related to solar irradiance, photovoltaic module temperature and output power, so that corresponding variables are selected from collected data in the modeling process, the attenuation model form is preset, and the optimization identification of unknown parameters of the model is carried out through the DHO-based modeling method.
S3.2: identifying a peak hour prediction model. The peak hour number H refers to the conversion of the total amount of solar radiation to standard test conditions within a certain time periodI.e. irradiance of 1000W/m2The hours at 25 ℃ can be generally calculated by the following formula:
wherein: t is t1And t2Respectively, a start time point and an end time point of the calculated time period, and r (t) is the irradiance of the photovoltaic array at time t.
As can be seen from the above equation, the number of peak hours is directly related to the solar irradiance, and therefore, the influence of the solar irradiance variation needs to be considered. Passing through t1～t2The change of the solar irradiance is closely related to the clear degree of the weather, the thickness of the cloud layer, the oxygen content of air, namely the humidity and the like.
Therefore, based on five weather types of sunny, cloudy, light rain and heavy to heavy rain, data such as ambient temperature, air humidity, cloud layer thickness and air density are selected from the sampling data to be used as modeling input, solar irradiance is output, and unknown parameters of the model are identified through a DHO algorithm based on a preset model form. The peak hours are then calculated using equation (11).
S3.3: and obtaining the optimal interval of the cleaning time of the photovoltaic array by combining the economy of the photovoltaic system according to the established photovoltaic system ash deposition attenuation model and the peak hour number prediction model.
The photovoltaic array cleaning effectiveness is then quantitatively analyzed in step S4 to enable a dynamic assessment of the resulting soot cleaning time rationality.
First, a rated peak power change curve of the following photovoltaic system after dust deposition and cleaning is drawn, as shown in fig. 3, P0Rated peak power before cleaning for accumulated dust, Tc represents cleaning time interval, t1Representing the specific time of a certain cleaning, it can be seen from the figure that the photovoltaic system can generate delta power generation of delta Q after cleaning.
For this fig. 3, the plant revenue change due to the cleaning operation of the plant is calculated quantitatively by a formula. Firstly, a power generation formula of a photovoltaic system is given:
Q(t)＝P(t)×H(t) (12)
wherein: h (t) represents the number of peak hours on day t calculated by equation (11).
If the photovoltaic array is not being cleaned during the Tc time period, the cumulative power generation can be expressed as:
if the photovoltaic array is at t within the Tc time period1The washing is carried out at any moment, and the accumulated power generation amount is changed into:
and r represents the grid-connected electricity price of the photovoltaic power generation, the obtained income E is as follows:
E＝r×Q (15)
the gain increase Δ E for photovoltaic power generation from the washing operation is then:
ΔE＝Eclean-Edust-C
Eclean＝r×Qclean (16)
Edust＝r×Qdust
wherein: ecleanAnd EdustAnd C, respectively representing the power generation income of whether the photovoltaic array is cleaned or not, wherein C is the cleaning cost. In this case, the rationality of the resulting cleaning time can be judged by whether Δ E is larger than 0 and the optimum cleaning time interval can be sought.
The invention provides a data-driven identification method and device for optimizing photovoltaic array cleaning time, which have the following beneficial effects compared with the prior art:
firstly, the invention provides a data-driven identification method for optimizing the cleaning time of a photovoltaic array by combining with a hunting deer optimization algorithm so as to predict the optimal cleaning time of a photovoltaic system and improve the power generation efficiency and the economy of the photovoltaic system.
Secondly, the invention relates to the rapid development of big data technology and swarm intelligence optimization algorithm, provides a data-driven model identification method based on the hunting deer optimization algorithm, enables the modeling process to be more intelligent and flexible, and is used for modeling a photovoltaic system soot deposition attenuation model and a peak hour number prediction model.
Thirdly, based on the established photovoltaic system soot deposition attenuation model and the peak hour number prediction model, the invention compares the power generation income change of the photovoltaic system before and after the cleaning operation through income quantification, thereby verifying the reasonability of the obtained cleaning time and providing effective reference for obtaining the optimal cleaning time of the photovoltaic array.
It will be understood that the above embodiments are merely exemplary embodiments taken to illustrate the principles of the present invention, which is not limited thereto. It will be apparent to those skilled in the art that various modifications and improvements can be made without departing from the spirit and substance of the invention, and these modifications and improvements are also considered to be within the scope of the invention.
Claims (10)
1. A data-driven identification method for optimizing photovoltaic array cleaning time is characterized by comprising the following specific steps:
collecting relevant data of the accumulated dust density and the output power of the photovoltaic array under a preset condition;
performing principle analysis and process description on model identification of the hunting deer optimization algorithm based on the related data;
the model parameter identification method based on the hunting deer optimization algorithm is used for respectively identifying the photovoltaic system dust accumulation attenuation model and the peak hour number prediction model;
and carrying out rationality evaluation and quantitative analysis on the predicted time of the dust deposition cleaning of the photovoltaic array.
2. The method of claim 1, wherein collecting data relating to the density and output power of the photovoltaic array under predetermined conditions comprises:
collecting data related to the accumulated dust density and the output power of the photovoltaic system by taking weather, environment and terrain factors as references;
and selecting a plurality of climate, environment and terrain condition combinations for data acquisition.
3. The method of claim 1, wherein the performing principle analysis and process description on model identification of the hunting deer optimization algorithm based on the relevant data comprises:
establishing a preset form for the model to be identified;
regarding the optimal position of the hunting deer as the optimal model parameter vector theta*Identifying and optimizing a hunting deer optimization algorithm model;
obtaining an optimal hunting position H according to the hunting deer optimization algorithm model identificationbestAnd theta*＝Hbest。
4. The method of claim 3, wherein said considering the optimal position of the hunting deer as the optimal model parameter vector θ*And optimizing the model identification of the optimization algorithm of the hunting deer, comprising the following steps:
taking s hunters as a population H, and randomly initializing the hunter population to H ═ H1,H2,…,Hs}；
Determining an initial wind direction angle and a position angle of a hunting male deer;
calculating to obtain a position closest to the optimal solution according to the fitness function;
and stopping position updating of the hunter after the optimal hunting position is determined.
5. The method of claim 4, wherein the determination of the initialized wind and position angles for a hunting male deer is obtained using the following formula:
φk＝2πl
wherein phi represents a wind direction angle,
represents a position angle; l is [0,1]]The random number in the range, k, represents the current iteration step.
6. The method of claim 4, wherein calculating the location closest to the optimal solution based on the fitness function comprises:
propagation process based on leader location: when the optimal position is determined, each hunter in the population attempts to reach the optimal position, which triggers a position update process, wherein hunter surround behavior can be represented by:
Hk+1＝Hlead-M·c|R×Hlead-Hk|
wherein: hleadRepresenting the position of the leader, c is a random number considering the wind direction angle, and the value range of c is (0, 2)]M and R both refer to parameter vectors;
position angle based propagation process: by taking into account the position angle during the hunter's position update
To improve the search space, in the hunting process of deer, a new parameter ds is introduced, assuming that the hunting process is valid in a certain position anglekAn update of the position angle is made and the parameter is developed from the variance between the wind angles, the viewing angle of which is as follows:
and, visual angle vs of preykThe following formula:
wherein rd is a random number taken from [0,1], and the parameter a is used for measuring a coefficient vector in an equation;
the position angle is updated by using the following relational expression:
updating the hunter position using the following relationship:
Hk+1＝Hlead-c|cos(w)×Hlead-Hk|；
wherein: hk+1To use hunter position vector H at time kkThe hunter position at the updated k +1 moment, w represents a position angle;
propagation process based on the position of the successor: assuming that the value of the vector R is less than 1, the updating process of the hunter position depends on the position of the successor, as follows:
Hk+1＝Hsuccessor-M·c|R×Hsuccessor-Hk|
wherein HsuccessorRepresenting the location of the successor.
7. The method according to any one of claims 1 to 6, wherein the model parameter identification method based on the hunting deer optimization algorithm is used for respectively identifying a photovoltaic system soot deposition attenuation model and a peak hour number prediction model, and comprises the following steps:
selecting the time of cleaning the photovoltaic module as a sampling starting point under different seasons and weather conditions, and establishing a power attenuation trend model describing the photovoltaic module along with the dust deposition duration by taking the change condition of rated peak power of the photovoltaic module along with the time lapse after cleaning as a reference;
and selecting environmental temperature, air humidity, cloud layer thickness and air density data from the sampling data as modeling input, taking solar irradiance as output, identifying unknown parameters of the model through a DHO algorithm based on a preset model form, and calculating the peak hours.
8. The method of claim 7, wherein the calculating the peak hours uses the following relationship:
wherein, the peak hour number H refers to the condition that the total solar radiation in a certain time period is converted into standard test, namely the irradiance is 1000W/m2Hours at a temperature of 25 ℃;
t1and t2Respectively, a start time point and an end time point of the calculated time period, and r (t) is the irradiance of the photovoltaic array at time t.
9. The method according to any one of claims 1 to 6, wherein the rationality assessment and quantitative analysis of the predicted time for soot cleaning of the photovoltaic array comprises:
drawing a rated peak power change curve of the following photovoltaic system after dust deposition and cleaning, and obtaining a power generation increment delta Q generated by the photovoltaic system after cleaning;
the method comprises the following steps of quantitatively calculating the yield increment delta E of the power station, caused by cleaning operation, for photovoltaic power generation through a formula as follows:
ΔE＝Eclean-Edust-C
Eclean＝r×Qclean
Edust＝r×Qdust
wherein E iscleanAnd EdustRespectively indicating the power generation income of whether the photovoltaic array is cleaned, C is the cleaning cost, QcleanRepresenting the t of the photovoltaic array over the Tc period1Cumulative power generation amount, Q, cleaned at all timesdustThe accumulated power generation amount of the photovoltaic array which is not cleaned in the Tc time period is represented, and r represents the grid-connected electricity price of photovoltaic power generation;
and, whether Δ E is larger than 0 is judged to obtain the rationality of the cleaning time and to seek the optimum cleaning time interval.
10. A photovoltaic array cleaning time optimized data driven identification device, comprising: the device comprises an acquisition module, an analysis module, an identification module and an evaluation module; wherein,
the acquisition module is used for acquiring relevant data of the accumulated dust density and the output power of the photovoltaic array under a preset condition;
the analysis module is used for carrying out principle analysis and process description on the model identification of the hunting deer optimization algorithm based on the relevant data;
the identification module is used for identifying a photovoltaic system dust accumulation attenuation model and a peak hour number prediction model respectively based on the model parameter identification method of the hunting deer optimization algorithm;
and the evaluation module is used for carrying out rationality evaluation and quantitative analysis on the predicted time of the accumulated dust cleaning of the photovoltaic array.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111405477.4A CN114169229A (en) | 2021-11-24 | 2021-11-24 | Data-driven identification method and device for optimizing cleaning time of photovoltaic array |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111405477.4A CN114169229A (en) | 2021-11-24 | 2021-11-24 | Data-driven identification method and device for optimizing cleaning time of photovoltaic array |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114169229A true CN114169229A (en) | 2022-03-11 |
Family
ID=80480360
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111405477.4A Pending CN114169229A (en) | 2021-11-24 | 2021-11-24 | Data-driven identification method and device for optimizing cleaning time of photovoltaic array |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114169229A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116232222A (en) * | 2023-05-10 | 2023-06-06 | 山东科技大学 | Cloud edge cooperative dust accumulation degree monitoring method and system for distributed photovoltaic system |
| CN117978081A (en) * | 2024-04-01 | 2024-05-03 | 誉金新能源科技(山东)有限公司 | Photovoltaic array self-cleaning system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160034627A1 (en) * | 2014-07-29 | 2016-02-04 | Samsung Electronics Co., Ltd. | Method and program for designing integrated circuit |
| CN107133713A (en) * | 2017-03-13 | 2017-09-05 | 华电电力科学研究院 | A kind of photovoltaic plant intelligently cleans the method for building up of decision system |
| EP3719689A1 (en) * | 2019-04-05 | 2020-10-07 | Sunpower Corporation | Generating a photovoltaic system design for a building |
| CN112001119A (en) * | 2020-08-14 | 2020-11-27 | 中国电力科学研究院有限公司 | Electric vehicle charging simulation method and system |
| CN112257331A (en) * | 2020-09-21 | 2021-01-22 | 郑州迈拓信息技术有限公司 | Cleaning method and system for icing of photovoltaic cell panel based on artificial intelligence |
| CN112561190A (en) * | 2020-12-23 | 2021-03-26 | 宁夏中科嘉业新能源研究院（有限公司） | Photovoltaic power station cleaning model prediction method based on discrete particle swarm algorithm |
-
2021
- 2021-11-24 CN CN202111405477.4A patent/CN114169229A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160034627A1 (en) * | 2014-07-29 | 2016-02-04 | Samsung Electronics Co., Ltd. | Method and program for designing integrated circuit |
| CN107133713A (en) * | 2017-03-13 | 2017-09-05 | 华电电力科学研究院 | A kind of photovoltaic plant intelligently cleans the method for building up of decision system |
| EP3719689A1 (en) * | 2019-04-05 | 2020-10-07 | Sunpower Corporation | Generating a photovoltaic system design for a building |
| CN112001119A (en) * | 2020-08-14 | 2020-11-27 | 中国电力科学研究院有限公司 | Electric vehicle charging simulation method and system |
| CN112257331A (en) * | 2020-09-21 | 2021-01-22 | 郑州迈拓信息技术有限公司 | Cleaning method and system for icing of photovoltaic cell panel based on artificial intelligence |
| CN112561190A (en) * | 2020-12-23 | 2021-03-26 | 宁夏中科嘉业新能源研究院（有限公司） | Photovoltaic power station cleaning model prediction method based on discrete particle swarm algorithm |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116232222A (en) * | 2023-05-10 | 2023-06-06 | 山东科技大学 | Cloud edge cooperative dust accumulation degree monitoring method and system for distributed photovoltaic system |
| CN116232222B (en) * | 2023-05-10 | 2023-09-08 | 山东科技大学 | Cloud edge cooperative dust accumulation degree monitoring method and system for distributed photovoltaic system |
| CN117978081A (en) * | 2024-04-01 | 2024-05-03 | 誉金新能源科技(山东)有限公司 | Photovoltaic array self-cleaning system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Liu et al. | An improved photovoltaic power forecasting model with the assistance of aerosol index data | |
| CN108399493B (en) | Method for predicting photovoltaic power generation loss caused by dust deposition and method for cleaning and judging photovoltaic module | |
| US20200057175A1 (en) | Weather dependent energy output forecasting | |
| Tao et al. | Forecasting power output for grid-connected photovoltaic power system without using solar radiation measurement | |
| CN113919232A (en) | Photovoltaic power station power prediction method and system based on recurrent neural network | |
| CN114169229A (en) | Data-driven identification method and device for optimizing cleaning time of photovoltaic array | |
| de Guia et al. | Using stacked long short term memory with principal component analysis for short term prediction of solar irradiance based on weather patterns | |
| CN107766986A (en) | Leak integral form echo state network on-line study photovoltaic power Forecasting Methodology | |
| CN112418346B (en) | Numerical weather forecast total radiation system error classification calculation method | |
| Mohammed et al. | Hourly solar radiation prediction based on nonlinear autoregressive exogenous (Narx) neural network | |
| CN115392494A (en) | Intelligent photovoltaic ash removal method and system | |
| CN116341613A (en) | Ultra-short-term photovoltaic power prediction method based on Informar encoder and LSTM | |
| CN107230003A (en) | A kind of power forecasting method of grid-connected power generation system | |
| CN110796292A (en) | Photovoltaic power short-term prediction method considering haze influence | |
| CN111814399A (en) | Model parameter optimization extraction method and measurement data prediction method for solar photovoltaic cell system | |
| CN114861555A (en) | Regional comprehensive energy system short-term load prediction method based on Copula theory | |
| CN117787551A (en) | Photovoltaic module cleaning decision method and system based on photovoltaic power generation amount and illumination radiation amount | |
| Tai et al. | Power prediction of photovoltaic power generation based on LSTM model with additive Attention mechanism | |
| CN112132344A (en) | Short-term wind power prediction method based on similar day and FRS-SVM | |
| CN117810987A (en) | Prediction method, prediction device, prediction terminal and prediction storage medium for residential payload section | |
| CN117728386A (en) | Prediction method for renewable energy power generation capacity of green building | |
| Wang et al. | Two-tier prediction of solar power generation with limited sensing resource | |
| Gerards et al. | Pv predictions made easy: Flexibility through simplicity | |
| CN116167508A (en) | Short-term photovoltaic output rapid prediction method and system based on meteorological factor decomposition | |
| CN116050592A (en) | Multi-dimensional photovoltaic power prediction method and system |
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 |