Photovoltaic power station clean optimization scheduling method and device

Abstract

The present invention provides a method and device for optimizing the cleaning and scheduling of a photovoltaic power station, and belongs to the field of photovoltaic power generation technology. The method of the present invention includes the following specific steps: constructing a prediction model for the amount of dust accumulated in photovoltaic cells, and predicting the amount of dust accumulated; obtaining the power generation according to the prediction model for the amount of dust accumulated in photovoltaic cells; proposing a total profit model for a photovoltaic power station according to the power generation and the cost of cleaning the dust accumulated; and obtaining the best cleaning scheduling plan for the photovoltaic power station within the scheduling cycle according to the environmental conditions and the amount of dust accumulated. The present invention proposes a prediction model for the amount of dust accumulated on photovoltaic panels within the cycle, which provides a reference for scheduling decisions of photovoltaic power stations, comprehensively considers the prediction of power generation, the cost of the dust cleaning plan, and the electricity price, and provides a total profit model for the power station, solving the problem of optimizing the scheduling of dust cleaning in photovoltaic power generation systems.

Description

Technical Field

[0001] The present invention belongs to the technical field of photovoltaic power generation, and in particular relates to a clean optimization scheduling method and device for a photovoltaic power station. Background Art

[0002] Photovoltaic power generation systems can avoid shading from trees and buildings through careful site selection and installation of PV systems, but shading from dust deposition has a lasting impact on PV power generation. Studies have shown that the degradation of photovoltaic performance caused by dust varies with location, environmental conditions and exposure time, and the average efficiency of PV modules after being exposed to dust for a period of time decreases from 4.5% to 40%. In addition, dust accumulation has been shown to cause hot spots on the surface of PV modules, which can cause damage to PV modules.

[0003] It can be seen that dust accumulation has a significant impact on the output power of photovoltaic modules. Predicting the periodic dust accumulation of the PV system and cleaning it in time is of great significance to the scheduling optimization of photovoltaic power stations. Based on this, the present invention proposes a photovoltaic power station cleaning optimization scheduling method and device. Summary of the invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art and provide a method and device for clean optimization scheduling of a photovoltaic power station.

[0005] In one aspect of the present invention, a method for optimizing clean scheduling of a photovoltaic power station is provided, comprising the following specific steps:

[0006] Construct a prediction model for the amount of dust accumulated on photovoltaic cells and predict the amount of dust accumulated;

[0007] According to the photovoltaic cell dust accumulation prediction model, the power generation is obtained;

[0008] Based on the power generation and dust cleaning cost, a total profit model of photovoltaic power station is proposed;

[0009] According to the environmental conditions and the obtained dust accumulation, the best cleaning scheduling plan for the photovoltaic power station within the scheduling period is obtained.

[0010] Optionally, the step of constructing a photovoltaic cell dust accumulation prediction model and predicting the dust accumulation includes:

[0011] The dust accumulation degree DAD of the photovoltaic panel within one day of the dusty photovoltaic cell in the time interval between two cleaning days is expressed as a Brownian motion model with drift to obtain the drift coefficient and diffusion coefficient of the dust accumulation;

[0012] The model is specifically as follows:

[0013]

[0014] B(t) is the standard Brownian motion, that is, B(t)~N(0,t); μ and σ d are the drift coefficient and diffusion coefficient, respectively;

[0015] is the dust accumulation degree DAD of photovoltaic panels in one day, where t is the number sequence in one day; n = [n1, n2, ... n m ] represents the cleaning plan; n1, n2…, n m is the numerical sequence of cleaning days, where m is a positive integer representing the number of cleaning times.

[0016] Optionally, on the day when the photovoltaic power station is cleaned, the dust accumulation degree of the photovoltaic panels in the corresponding day is 0.

[0017] Optionally, obtaining the power generation power according to the photovoltaic cell dust accumulation prediction model includes:

[0018] The power loss caused by dust accumulation on photovoltaic cells in one day is collected. The actual power generation on day t due to the influence of dust is in,

[0019] Define the ideal power generation of the PV string on day t as Et, according to The power loss caused by dust deposition on day t is

[0020] Optionally, a total profit model of a photovoltaic power station is proposed based on the power generation and dust cleaning cost, including:

[0021] The total profit of a photovoltaic power station is expressed as the difference between the income from selling electricity and the cleaning cost. The specific mathematical model is as follows:

[0022]

[0023] Among them, ξ t is the electricity price on day t, η n is the total profit of the PV string in the PV power station under cleaning scheme n; m is the number of cleaning times, and w is the cleaning cost of the PV equipment once.

[0024] Optionally, obtaining an optimal cleaning scheduling plan for a photovoltaic power station within a scheduling period based on environmental conditions and the obtained dust accumulation amount includes:

[0025] According to the environmental conditions and the dust accumulation, the expected total profit under different cleaning plans is calculated to optimize the power station scheduling. The model of DAD and cleaning plan in the scheduling period c is expressed as follows:

[0026] if n c ≠0

[0027]

[0028] otherwise

[0029]

[0030] Among them, the cleaning plan is n c (n c ∈{0,1,2,…,T s}); Let c (c = 1, 2, ..., C) be the digital sequence of scheduling periods; the periods numbered by c include the periods from (c-1)T s +1 to cT s The DAD of a photovoltaic power plant every day can be expressed in a cyclic form: The forecast time range T s ∈[7,10] days as the scheduling period.

[0031] Optionally, according to the cleaning plan model, in the cleaning plan n c Under this condition, the dispatching income of the photovoltaic power station in the dispatching period c is expressed by the following relationship:

[0032]

[0033]

[0034] Among them, u c is a label indicating whether the PV module has been cleaned in the scheduling period c; T S To predict the time;

[0035] The ideal power generation on day t is The electricity price is

[0036] Optionally, the profit loss of the photovoltaic power generation system due to dust accumulation is expressed as the following relationship:

[0037]

[0038] Where g is the conversion efficiency of the photovoltaic module; Indicates the length of time that the initial DAD of scheduling period c affects the period; represents the average radiation price in scheduling period c; represents the predicted electricity price value of the dispatch period c; n c Plan for cleaning; represents the DAD in scheduling period c; and,

[0039] The objective function model of the scheduling period c is as follows:

[0040]

[0041] Optionally, the ideal power generation of the photovoltaic power station in one day is as follows:

[0042]

[0043] Among them, p t-h is the ideal power generation power of the photovoltaic power station; P max is the maximum rated power of the photovoltaic panel, G t-h and G STC are actual solar radiation and standard radiation respectively; β is the power temperature coefficient, Tem t-h and NOTC are the actual temperature of the module and the normal operating temperature of the module respectively.

[0044] Another aspect of the present invention provides a photovoltaic power station cleaning optimization scheduling device, including a prediction module, a power calculation module, a profit calculation module and a cleaning scheduling module:

[0045] The prediction module is used to construct a photovoltaic cell dust accumulation prediction model and predict the dust accumulation;

[0046] The power calculation module is used to obtain the power generation power according to the photovoltaic cell dust accumulation prediction model;

[0047] The profit calculation module is used to propose a total profit model of the photovoltaic power station based on the power generation and dust cleaning cost;

[0048] The cleaning scheduling module is used to obtain the best cleaning scheduling plan for the photovoltaic power station within the scheduling period according to the environmental conditions and the obtained dust accumulation amount.

[0049] The present invention provides a method for optimizing the cleaning and scheduling of a photovoltaic power station, comprising the following specific steps: constructing a prediction model for the amount of dust accumulated on photovoltaic cells, and predicting the amount of dust accumulated; obtaining the power generation according to the prediction model for the amount of dust accumulated on photovoltaic cells; proposing a total profit model for photovoltaic power stations according to the power generation and the cost of cleaning the dust accumulated; and obtaining the best cleaning scheduling plan for the photovoltaic power station within the scheduling cycle according to environmental conditions and the amount of dust accumulated. The present invention proposes a prediction model for the amount of dust accumulated on photovoltaic panels within the cycle, which provides a reference for scheduling decisions of photovoltaic power stations, comprehensively considers the prediction of power generation, the cost of the dust cleaning plan, and the electricity price, and provides a total profit model for the power station, solving the problem of optimizing the scheduling of dust cleaning in photovoltaic power generation systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 A flowchart of a clean optimization scheduling method for a photovoltaic power station according to an embodiment of the present invention;

[0051] Figure 2A cleaning scheduling framework diagram of a dust-accumulated photovoltaic power generation system according to another embodiment of the present invention;

[0052] Figure 3 This is a schematic structural diagram of a photovoltaic power station clean optimization scheduling device according to another embodiment of the present invention. DETAILED DESCRIPTION

[0053] In order to enable those skilled in the art to better understand the technical solution of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific implementation methods. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the described embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative work are within the scope of protection of the present invention.

[0054] like Figure 1 and Figure 2 As shown, in one aspect of the present invention, a photovoltaic power station clean optimization scheduling method S100 is provided, comprising the following specific steps S110 to S140:

[0055] S110, constructing a photovoltaic cell dust accumulation prediction model, and predicting the dust accumulation.

[0056] Specifically, the power loss caused by dust is quantified by the power loss caused by dust accumulation on photovoltaic cells within one day. To describe the dust accumulation degree (DAD) of photovoltaic panels within a day, where t is the number sequence within the day; n = [n1, n2, ... n m ] represents the cleaning scheme. The DAD of the same day will be different under different cleaning schemes. n1, n2…, n m is the number sequence of cleaning days, where m is a positive integer representing the number of cleaning times. Considering that manual cleaning ensures the cleaning effect, on the day of cleaning the PV plant, the dust deposition is removed, that is, DAD is 0.

[0057]

[0058] Studies have shown that the average dust density of PV panels increases linearly with time over a period of time after cleaning. In addition, the increase in dust accumulation is linearly correlated with the decrease in power generation efficiency. It is assumed that the expected value of the power loss rate of dust accumulation on PV panels is linearly related to its exposure time after cleaning. Since the weather changes over time, the process of dust accumulation is uncertain, and the uncertainty increases with the increase in exposure time. Therefore, the DAD of the dusty PV panel within the time interval of two cleaning days is expressed as a Brownian motion model with drift:

[0059]

[0060] Among them, B(t) is the standard Brownian motion, that is, B(t) ~ N(0,t); μ and σd are the drift coefficient and diffusion coefficient respectively. With the help of the historical operation data of the photovoltaic power station, the ideal and actual power generation at the historical moment are calculated, that is, the DAD at the historical moment. Then, the drift coefficient and diffusion coefficient can be obtained by fitting these historical data. Since dust deposition monitoring is carried out every day, these coefficients will be corrected and updated according to the dust monitoring results and the changes in historical data over time.

[0061] S120, obtaining power generation power according to the photovoltaic cell dust accumulation prediction model.

[0062] Specifically, the ideal power generation of the PV string on day t is defined as Et. "Ideal" means that under the environmental conditions on day t, if the components are clean, the power provided by the PV string is Et, that is, according to The power generation loss caused by dust deposition on day t is Therefore, the actual power generation on day t considering the dust effect is expressed as

[0063] S130: Based on the power generation and dust cleaning cost, a total profit model of a photovoltaic power station is proposed.

[0064] Specifically, PV cleaning is a method of sacrificing a portion of profit to improve the long-term efficiency of PV power plants. For the same PV string, the number and difficulty of cleaning are stable over a long period of time, so its one-time cleaning cost w is constant. The total profit of the PV string over a long period of time can be expressed as the difference between the income from selling electricity and the cleaning cost. The total profit of the PV power plant can be maximized by properly arranging the cleaning time. Therefore, the mathematical model of the problem is expressed as follows:

[0065]

[0066] Among them, ξ t is the electricity price on day t, η n is the total profit of the PV string in the PV power station under cleaning scheme n.

[0067] S140, according to the environmental conditions and the obtained dust accumulation amount, to obtain the best cleaning scheduling plan for the photovoltaic power station within the scheduling period, that is, the photovoltaic power station periodic scheduling based on dust accumulation cleaning.

[0068] Specifically, in order to optimize the total profit without affecting the normal operation of the photovoltaic power station, we divide the photovoltaic cleaning process into reasonable small blocks and optimize them step by step. A non-periodic scheduling method based on environmental condition prediction, photovoltaic power prediction and dust accumulation prediction is adopted. The predicted time range T s∈[7,10] days as the "dispatching cycle". Within this time frame, the model is used to predict the daily dust accumulation and environmental conditions of the photovoltaic panels, and then the expected total profit under different cleaning plans is calculated to optimize the power station dispatch. From t = 1 to t = T s The cleaning plan during this period is to first optimize the first fixed scheduling period, then enter the next period, and then expand to the entire time range. The time length of each period is the same, so the measurement accuracy is the same every time. The model of DAD and cleaning plan in scheduling period c is expressed as:

[0069]

[0070] Wherein, let c (c = 1, 2, ..., C) be the digital sequence of the scheduling period, and the period numbered by c includes the period from (c-1)T s +1 to cT s Therefore, the DAD of a photovoltaic power plant for each day can be expressed in a cyclic form: The ideal power generation on day t is The electricity price is According to practical experience, a photovoltaic power plant is usually cleaned at most once in a scheduling cycle. Therefore, in the scheduling cycle numbered c, the cleaning plan n is simplified to n c (n c ∈{0,1,2,…,T s}). This is a sign of cleaning day. c =0 means that the PV has not been cleared during the scheduling cycle.

[0071] Furthermore, if the photovoltaic panels are cleaned on a certain day during this time period, that is, n c = 0, then from t = (c-1)T s +1 to t = c T s The time frame will be divided into two parts, is the DAD of the last day of the last cycle of the PV power station. It is also the initial DAD of this cycle. If the PV modules are not cleaned during this cycle, that is, n c = 0, dust will continue to accumulate. c The dispatching income of the photovoltaic power station in the next cycle c is:

[0072]

[0073]

[0074] Among them, u c Is a label that shows whether the PV module has been cleaned in the c cycle; T S For the prediction time.

[0075] Furthermore, the profit loss of the photovoltaic power generation system due to dust accumulation is:

[0076]

[0077] Where g is the conversion efficiency of the photovoltaic module. It indicates the length of time that the initial DAD of period c affects the period. represents the average radiation price in scheduling period c; represents the predicted electricity price value of the dispatch period c; n c Plan for cleaning; represents the DAD in the scheduling period c; the objective function model of the scheduling period c is as follows:

[0078]

[0079] Among them, the ideal power generation of a photovoltaic power station in one day is:

[0080]

[0081] Among them, P max is the maximum rated power of the photovoltaic panel, G t-h and G STC are actual solar radiation and standard radiation respectively. β is the power temperature coefficient and Tem t-h and NOTC are the actual module temperature and the normal operating temperature of the cell, respectively. max , G STC , β and NOTC parameters can be obtained according to the specifications of the photovoltaic panel.

[0082] It should be noted that the above process has completed the design of the cleaning scheduling plan for the dust-accumulated photovoltaic power generation system. Please refer to Figure 2 Moreover, the total profit of the power station can be increased according to the cleaning cost, power generation capacity and electricity price.

[0083] like Figure 2 and Figure 3 As shown, another aspect of the present invention provides a photovoltaic power station cleaning optimization scheduling device 200, including a prediction module 210, a power calculation module 220, a profit calculation module 230 and a cleaning scheduling module 240: wherein the prediction module 210 is used to construct a photovoltaic cell dust accumulation prediction model and predict the dust accumulation; the power calculation module 220 is used to obtain the power generation power according to the photovoltaic cell dust accumulation prediction model; the profit calculation module 230 is used to propose a total profit model of the photovoltaic power station according to the power generation power and the dust cleaning cost; the cleaning scheduling module 240 is used to obtain the best cleaning scheduling plan for the photovoltaic power station within the scheduling period according to the environmental conditions and the obtained dust accumulation.

[0084] It should be noted that the scheduling method implemented by the scheduling device of this embodiment can be referred to the previous description and will not be repeated here.

[0085] The following will further illustrate the photovoltaic power station clean optimization scheduling method with reference to specific embodiments:

[0086] This example is based on the historical operating data of a 20MW photovoltaic power station within 7 days for analysis.

[0087] The time range involved in the PV power plant cleaning plan is considered in a segmented manner, that is, segment t = 1, 2, ..., T, where T is a positive integer. The scheduling and forecasting time unit is set to one day. The PV power plant optimizes the daily PV cell cleaning according to the scheduling plan to maximize the total profit of the power plant over a period of time.

[0088] 1. Establishing a dust accumulation model for photovoltaic power generation systems

[0089] The power loss caused by dust accumulation on photovoltaic cells in one day is used to quantify the power loss caused by dust. The dust accumulation degree (DAD) of photovoltaic panels in one day is calculated based on historical operation data. The DAD is 0 on the day when the photovoltaic power station is cleaned. Since the weather changes over time, the process of dust accumulation is uncertain, and the uncertainty increases with the increase of exposure time. With the help of historical operation data in the past 7 days, the DAD model of dusty photovoltaic panels within the time interval of two cleaning days is established according to formula (10):

[0090]

[0091] Subsequently, the drift coefficient and diffusion coefficient can be obtained by fitting these historical data, and the coefficients can be corrected and updated according to the dust monitoring results and the changes in historical data over time.

[0092] 2. Photovoltaic power generation forecast

[0093] Based on the historical operation data of the photovoltaic power station within 7 days, the power generation loss caused by dust deposition on day t is calculated. Therefore, the actual power generation on day t considering the impact of dust is expressed as

[0094] 3. Calculate the cost of PV cleaning and the total profit of the power station

[0095] Since the number and difficulty of cleaning a photovoltaic power station are stable over a long period of time, its one-time cleaning cost w is constant. The total profit of a PV string over a long period of time can be expressed as the difference between the income from selling electricity and the cleaning cost. The total profit of a photovoltaic power station can be maximized by properly arranging the cleaning time, and the photovoltaic power station scheduling model is expressed as:

[0096]

[0097] 4. Design of a periodic scheduling scheme for photovoltaic power plants based on dust cleaning

[0098] In order to optimize the total profit without affecting the normal operation of the photovoltaic power station, we divide the photovoltaic cleaning process into reasonable small blocks and optimize them step by step. A non-periodic scheduling method based on environmental condition prediction, photovoltaic power prediction and dust accumulation prediction is adopted. The prediction time range T s ∈[7,10] days as the "dispatching cycle". Within this time frame, the model is used to predict the daily dust accumulation and environmental conditions of the photovoltaic panels, and then the expected total profit under different cleaning plans is calculated to optimize the power station dispatch. From t = 1 to t = T s The cleaning plan during this period is to first optimize the first fixed scheduling cycle, then enter the next cycle, until it expands to the entire time range. The time length of each cycle is the same, so the measurement accuracy is the same every time. According to formula (12), the DAD and cleaning plan model in the scheduling cycle c is described as:

[0099]

[0100] Here, let c (c = 1, 2, ..., C) be the numerical sequence of the scheduling period. The period numbered by c includes the period from (c-1)T s +1 to cT s Therefore, the DAD of a photovoltaic power plant for each day can be expressed in a cyclic form: The ideal power generation on day t is The electricity price is According to practical experience, a photovoltaic power plant is usually cleaned at most once in a scheduling cycle. Therefore, in the scheduling cycle numbered c, the cleaning plan n is simplified to n c (n c ∈{0,1,2,…,T s}). This is a sign of cleaning day, n c =0 means that the PV has not been cleared during the scheduling cycle.

[0101] If the photovoltaic panels are cleaned on a certain day during this time period, that is, n c = 0, then from t = (c-1)T s +1 to t = c T s The time frame will be divided into two parts, is the DAD of the last day of the last cycle of the PV power station. It is also the initial DAD of this cycle. If the PV modules are not cleaned during this cycle, that is, n c = 0, dust will continue to accumulate, calculate the cleaning plan n cThe dispatching income of the photovoltaic power station in the next cycle c is:

[0102]

[0103]

[0104] Among them, u c Is a label that shows whether the PV module has been cleaned during the c-cycle.

[0105] Calculate the loss of profits due to dust accumulation in a photovoltaic system:

[0106]

[0107] Where g is the conversion efficiency of the photovoltaic module. Indicates that the initial DAD of period c affects the length of time of the period, and gives the objective function model of scheduling period c:

[0108]

[0109] The ideal power generation of a photovoltaic power station in one day is predicted to be:

[0110]

[0111] Among them, P max is the maximum rated power of the photovoltaic panel, G t-h and G STC are actual solar radiation and standard radiation respectively. β is the power temperature coefficient and Tem t-h and NOTC are the actual temperature of the module and the normal operating temperature of the cell respectively. max , G STC , β and NOTC parameters can be obtained according to the specifications of the photovoltaic panel.

[0112] In addition, execute Figure 2 The cleaning scheduling scheme of the dust-accumulated photovoltaic power generation system shown in the figure achieves the goal of maximizing the total profit of the power station by comprehensively considering the cleaning cost, power generation and electricity price.

[0113] The present invention provides a method and device for optimizing the clean scheduling of a photovoltaic power station, which has the following beneficial effects compared with the prior art:

[0114] First, the present invention proposes a prediction model for the amount of dust accumulated on photovoltaic panels within a cycle, providing a reference for scheduling decisions of photovoltaic power stations.

[0115] Second, the present invention provides a multi-objective scheduling model that comprehensively considers photovoltaic cleaning costs, power generation and electricity prices, laying the foundation for designing photovoltaic cleaning solutions.

[0116] Third, the present invention obtains the optimal cleaning plan by solving the scheduling model based on the historical operation data of the photovoltaic power station within 7 days, thereby achieving the goal of optimizing the scheduling of the photovoltaic power station.

[0117] It is to be understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of the present invention, but the present invention is not limited thereto. For those of ordinary skill in the art, various modifications and improvements can be made without departing from the spirit and essence of the present invention, and these modifications and improvements are also considered to be within the scope of protection of the present invention.

Claims

1. A clean optimization scheduling method for a photovoltaic power station, characterized in that: The specific steps include: Construct a prediction model for the amount of dust accumulated on photovoltaic cells and predict the amount of dust accumulated, including: The dust accumulation degree DAD of the photovoltaic panel within one day of the dusty photovoltaic cell in the time interval between two cleaning days is expressed as a Brownian motion model with drift to obtain the drift coefficient and diffusion coefficient of the dust accumulation; The model is specifically as follows: t∈(n i ,n j ),i,j∈{1,2,…,m}&i<j; B(t) is the standard Brownian motion, that is, B(t)~N(0,t); μ and σ d are the drift coefficient and diffusion coefficient, respectively; is the dust accumulation degree DAD of photovoltaic panels in one day, where t is the number sequence in one day; n = [n1, n2, ... n m ] represents the cleaning plan; n1, n2…, n m is the number sequence of cleaning days, where m is a positive integer representing the number of cleaning times; on the day of cleaning the photovoltaic power station, the dust accumulation degree of the photovoltaic panels in the corresponding day is 0; According to the photovoltaic cell dust accumulation prediction model, the power generation is obtained, including: The power loss caused by dust accumulation on photovoltaic cells in one day is collected. The actual power generation on day t due to the influence of dust is in, Define the ideal power generation of the PV string on day t as Et, according to The power loss caused by dust deposition on day t is According to the power generation and dust cleaning cost, a total profit model of photovoltaic power station is proposed, including: The total profit of a photovoltaic power station is expressed as the difference between the income from selling electricity and the cleaning cost. The specific mathematical model is as follows: Among them, ξ t is the electricity price on day t, η n is the total profit of the PV string in the PV power station under cleaning scheme n; m is the number of cleaning times, and w is the cleaning cost of the PV equipment once; According to the environmental conditions and the dust accumulation, the best cleaning scheduling plan for the photovoltaic power station within the scheduling period is obtained, including: According to the environmental conditions and the dust accumulation, the expected total profit under different cleaning plans is calculated to optimize the power station scheduling. The model of DAD and cleaning plan in the scheduling period c is expressed as follows: if n c ≠0 otherwise Among them, the cleaning plan is n c (n c ∈{0,1,2,…,T s }); Let c (c = 1, 2, ..., C) be the digital sequence of scheduling periods; the periods numbered by c include the periods from (c-1)T s +1 to cT s The DAD of a photovoltaic power plant every day can be expressed in a cyclic form: The forecast time range T s ∈[7,10] days as the scheduling period.

2. The method according to claim 1, characterized in that According to the cleaning plan model, in the cleaning plan n c Under this condition, the dispatching income of the photovoltaic power station in the dispatching period c is expressed by the following relationship: Among them, u c is a label indicating whether the PV module has been cleaned in the scheduling period c; T S To predict the time; The ideal power generation capacity on day t is The electricity price is 3. The method according to claim 2, characterized in that The profit loss of the photovoltaic power station due to dust accumulation is expressed as the following relationship: Where g is the conversion efficiency of the photovoltaic module; Indicates the length of time that the initial DAD of scheduling period c affects the period; represents the average radiation price in scheduling period c; represents the predicted electricity price value of the dispatch period c; n c Plan for cleaning; represents the DAD in scheduling period c; and, The objective function model of the scheduling period c is as follows:

4. The method according to claim 3, characterized in that The ideal power generation of the photovoltaic power station in one day is as follows: Among them, p t-h is the ideal power generation power of the photovoltaic power station; P max is the maximum rated power of the photovoltaic panel, G t-h and G STC are actual solar radiation and standard radiation respectively; β is the power temperature coefficient, Tem t-h and NOTC are the actual temperature of the module and the normal operating temperature of the module respectively.

5. A photovoltaic power station clean optimization scheduling device using the photovoltaic power station clean optimization scheduling method according to any one of claims 1 to 4, characterized in that: Including prediction module, power calculation module, profit calculation module and cleaning scheduling module: The prediction module is used to construct a photovoltaic cell dust accumulation prediction model and predict the dust accumulation; The power calculation module is used to obtain the power generation power according to the photovoltaic cell dust accumulation prediction model; The profit calculation module is used to propose a total profit model of the photovoltaic power station based on the power generation and dust cleaning cost; The cleaning scheduling module is used to obtain the best cleaning scheduling plan for the photovoltaic power station within the scheduling period according to the environmental conditions and the obtained dust accumulation amount.

Images