Hybrid control method for photovoltaic tracking system based on double-layer nested irradiance prediction

Abstract

The invention discloses a hybrid control method for a photovoltaic tracking system based on double-layer nested irradiance prediction, including: in sunny and dust-free weather conditions, a closed-loop control algorithm of an ultraviolet sensor is used to control the azimuth of a photovoltaic panel; in cloudy, dusty or rainy days, an open-loop control algorithm containing tracking motor energy consumption constraints is used to obtain historical data of irradiation coefficient; a double-layer nested irradiance prediction model based on GM (1,1)-DMC is constructed, the irradiation coefficient is input, the irradiation coefficient prediction value is obtained using the GM (1,1) model, and the obtained irradiation coefficient prediction value is optimized using the DMC algorithm; an irradiance model is constructed, and the predicted total irradiance data of a photovoltaic panel is obtained by using the irradiance model for the optimized irradiance coefficient prediction value; the predicted total irradiance data of a photovoltaic panel is used as the input of the open-loop control algorithm, and the optimal azimuth of a photovoltaic panel is output. The invention can effectively improve the energy absorption efficiency of a photovoltaic panel and extend the service life of a photovoltaic tracking bracket.

Description

Technical Field

[0001] The present invention belongs to the technical field of photovoltaic tracking, and more specifically relates to a hybrid control method of a photovoltaic tracking system based on double-layer nested irradiance prediction. Background Art

[0002] Due to the huge carbon dioxide emissions from countries around the world, global temperatures are gradually rising, and the Arctic Ocean is facing the problem of melting glaciers and acidification of seawater. In order to meet these challenges, how to efficiently, cleanly and stably develop and utilize new energy has become the focus of scholars. In recent years, the development of new energy has been increasing, and solar energy has been widely promoted due to its easy access and convenient use. In photovoltaic applications, the use of photovoltaic tracking systems is one of the methods to improve photovoltaic utilization. The current photovoltaic tracking system mainly relies on open-loop tracking control technology based on solar position prediction and solar irradiance prediction. Its accuracy is directly limited by the effectiveness of the solar position prediction algorithm and the accuracy of the irradiance prediction data; in addition, a small number of photovoltaic tracking systems use traditional closed-loop tracking control technology to track the sun's position, using light-dependent resistors to detect visible light spectrum transmission signals to achieve tracking of the sun, but it will greatly reduce the tracking performance of the photovoltaic system under weather conditions of saturated light intensity and low visibility. None of the above photovoltaic tracking systems fully consider the tracking performance of the system. Summary of the invention

[0003] In view of the deficiencies in the prior art, an object of the present invention is to provide a hybrid control method for a photovoltaic tracking system based on double-layer nested irradiance prediction.

[0004] In order to achieve the above object, the present invention provides the following technical solution: a hybrid control method for a photovoltaic tracking system based on double-layer nested irradiance prediction, the control method comprising the following steps:

[0005] Select a tracking control algorithm based on weather conditions; if it is sunny and dusty, choose a closed-loop control algorithm using a UV sensor; if it is cloudy, dusty or rainy, choose an open-loop control algorithm that includes tracking motor energy consumption constraints;

[0006] Adopting an open-loop control algorithm that includes tracking the motor energy consumption constraint includes the following steps:

[0007] Obtain historical data of irradiance coefficient;

[0008] A double-layer nested irradiance prediction model based on GM(1,1)-DMC is constructed. The irradiation coefficient is input, and the irradiation coefficient prediction value is obtained using the GM(1,1) model. The irradiation coefficient prediction value obtained by the GM(1,1) model is optimized using the dynamic matrix predictive control DMC algorithm to obtain the optimized irradiation coefficient prediction value.

[0009] Construct an irradiance model for the photovoltaic tracking system, and use the optimized irradiance coefficient prediction value through the irradiance model to obtain the predicted total irradiance data of the photovoltaic panel;

[0010] The predicted total irradiance data of the photovoltaic panel is used as the input of the open-loop control algorithm to output the optimal azimuth angle of the photovoltaic panel;

[0011] The closed-loop control algorithm using the UV sensor consists of the following steps:

[0012] An ultraviolet sensor is used to detect the real-time irradiance on the left and right sides of the photovoltaic panel, and the real-time irradiance difference signal on the left and right sides of the photovoltaic panel is used as the input of the closed-loop control algorithm to dynamically adjust the azimuth angle of the photovoltaic panel.

[0013] Furthermore, the irradiance coefficient includes: the normal irradiance coefficient received by the photovoltaic panel , the irradiance diffusion coefficient of sunlight on the front and back of photovoltaic panels , , the overall irradiance coefficient of sunlight on the ground , ground area , , The reflected irradiance coefficient of sunlight , , , , The area on the ground directly reached by sunlight. The shadow area of ​​the ground.

[0014] Furthermore, the radiation coefficient is input, and the radiation coefficient prediction value is obtained by using the GM (1,1) model. The radiation coefficient prediction value obtained by the GM (1,1) model is optimized by using the dynamic matrix predictive control DMC algorithm to obtain the optimized radiation coefficient prediction value, including the following steps:

[0015] Input the original data of irradiance coefficient and generate the irradiance coefficient data matrix :

[0016] ,

[0017] For each row vector in the irradiance coefficient data matrix , , , , , , The elements in the matrix are averaged and used as known data to generate the radiation coefficient sequence matrix. ;

[0018] For the irradiance coefficient sequence matrix Perform 1-AGO accumulation processing to generate the irradiance coefficient whitening matrix , the processing expression is:

[0019]

[0020] In the formula, is the row vector of the irradiance coefficient sequence matrix No. elements, u is the number of column vectors in the irradiance coefficient sequence matrix, is the row vector of the irradiance whitening matrix No. elements, is the type of irradiation coefficient;

[0021] Perform sliding average processing on the irradiance coefficient whitening matrix to generate the irradiance coefficient grayscale matrix , the processing expression is:

[0022] ,

[0023] In the formula, is the row vector of the irradiance gray matrix No. elements;

[0024] The least square method is used to solve the development coefficient m and gray generation amount n of the GM (1,1) model. The expression is:

[0025] ,

[0026] ,

[0027] ,

[0028] In the formula, , is the constructed data matrix, which is used to solve the development coefficient m and the grey generation amount n; , They are respectively the irradiance coefficient gray matrix and the irradiance coefficient sequence matrix row vector No. elements;

[0029] Substitute the solved development coefficient m and gray generation n into the following matrix equation to solve and obtain the predicted value of irradiation coefficient:

[0030] ,

[0031] ,

[0032] ,

[0033] In the formula, is the first whitening matrix of the irradiance coefficient column vector, is the first column vector of the irradiance whitening matrix;

[0034] The dynamic matrix predictive control DMC algorithm is used to optimize the irradiation coefficient prediction value obtained by the GM (1,1) model to obtain the optimized irradiation coefficient prediction value, as follows:

[0035] Step 1: Predict the overall radiation coefficient at time t+T in the future ,Right now

[0036] ,

[0037] Step 2: predict the overall radiation coefficient at time t+1+T in the future ,Right now

[0038] ,

[0039] Step 3: For the interval [t+2, t+T], the predicted overall irradiance coefficient at time t+1 is subtracted from the predicted overall irradiance coefficient at time t to obtain the prediction error ,Right now

[0040] ,

[0041] Step 4: convert the prediction error Input into the GM(1,1) model, predict the error at the future time [t+2, t+T], and update the predicted value of the overall irradiance coefficient at time t+2 ,Right now

[0042] ,

[0043] In the formula is the predicted value of the overall irradiance coefficient at time t+2 before the update, is the prediction error at time t+2;

[0044] Step 5, repeat steps 2 to 4 to complete the prediction value optimization of a complete cycle T.

[0045] Furthermore, the irradiance model is as follows:

[0046] ,

[0047] In the formula, represents the total irradiance of the photovoltaic panel, , , They represent the direct irradiance, diffuse irradiance and ground reflected irradiance received by the photovoltaic panels respectively.

[0048] Furthermore, the direct irradiance received by the photovoltaic panel is calculated as follows:

[0049] ,

[0050] in, represents the normal irradiance coefficient received by the photovoltaic panel, is the incident angle of the sun's rays.

[0051] Furthermore, the calculation formula for the diffuse irradiance received by the photovoltaic panel is as follows:

[0052] ,

[0053] ,

[0054] ,

[0055] in, , Respectively represent the diffuse irradiance received by the front and back sides of the photovoltaic panel; is the positive conversion factor; , are the irradiance diffusion coefficients of sunlight on the front and back of the photovoltaic panel respectively; is the length of the photovoltaic panel, is the distance between the photovoltaic arrays; is the angle between the photovoltaic panel and the horizontal plane.

[0056] Furthermore, the ground reflected irradiance received by the photovoltaic panel is calculated as follows:

[0057] ,

[0058] ,

[0059] ,

[0060] in, , Respectively represent the ground reflected radiation received by the front and back of the photovoltaic panel, is the overall irradiance coefficient of sunlight on the ground, is the height of the photovoltaic panel from the ground, is the ground reflectivity, , , Represents the ground area , , The reflected irradiance of sunlight.

[0061] Furthermore, the azimuth angle of the photovoltaic tracking system and the angle relationship in the irradiance model are as follows:

[0062] ,

[0063] ,

[0064] In the formula, , are the azimuth and altitude angles of the sun’s rays relative to the photovoltaic panel, Represents the angle of the photovoltaic panel relative to the plane on which it is installed.

[0065] Furthermore, the predicted total irradiance data of the photovoltaic panel is used as the input of the open-loop control algorithm to output the optimal azimuth angle of the photovoltaic panel, including the following steps:

[0066] The open-loop control algorithm aims to solve the maximum output power of the photovoltaic panel under the predicted total irradiance of the photovoltaic panel. The objective function expression of the open-loop control algorithm is as follows:

[0067] ,

[0068] In the formula, represents the output power of the photovoltaic panel under the predicted total irradiance of the photovoltaic panel; Indicates the output power of the photovoltaic panel under standard operating conditions; Indicates the irradiance of photovoltaic panels under standard operating conditions; is the temperature coefficient, is the temperature of the photovoltaic panel under standard operating conditions, is the operating temperature of the photovoltaic panel;

[0069] Introducing constraints, the maximum output power of the photovoltaic panel obtained is Subtract the tracking motor movement to the azimuth The power consumption is calculated to obtain the predicted net power under the total irradiance of the photovoltaic panel. If the net power is greater than the output power of the tracking motor in the current state, the tracking motor works, otherwise the tracking motor does not work. The constraint expression is as follows:

[0070] ,

[0071] To track the energy consumption of the motor; To track the output power of the motor in its current state.

[0072] Furthermore, a closed-loop control algorithm using the UV sensor is used, including:

[0073] Define the constraints of the closed-loop control algorithm for adaptive tracking. If the constraints are met, the photovoltaic panel will adaptively adjust the azimuth angle, otherwise it will maintain the position, as follows:

[0074] ,

[0075] In the formula, , Respectively represent the ultraviolet irradiance signals on the left and right sides of the photovoltaic panel, is the working critical value.

[0076] Compared with the prior art, the present invention achieves the following beneficial effects:

[0077] The present invention constructs a photovoltaic tracking system based on the GM(1,1)-DMC double-layer nested irradiance prediction model to solve the problem of poor tracking performance caused by inaccurate irradiance prediction;

[0078] The present invention proposes an open-loop control algorithm with the maximum output power of the photovoltaic panel as the goal, including tracking motor consumption constraints, and obtains the predicted maximum output power of the photovoltaic panel under the total radiation condition and the corresponding optimal azimuth angle of the photovoltaic panel;

[0079] Taking into account specific weather conditions, the present invention innovatively proposes a hybrid control method that combines a closed-loop control algorithm of a UV sensor with an open-loop control algorithm. This method comprehensively considers the impact of factors such as the energy consumption and tracking performance of the photovoltaic tracking system on its output power under various weather conditions, effectively improves the energy conversion efficiency of the photovoltaic tracking system, and significantly improves the accuracy of the tracking system. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Figure 1 It is a schematic diagram of the process of the present invention;

[0081] Figure 2 It is a schematic diagram of the flow of the GM(1,1)-DMC double-layer nested irradiance prediction model of the present invention. DETAILED DESCRIPTION

[0082] The present invention is further described below in conjunction with embodiments. It is necessary to point out that the following embodiments are only used to further illustrate the present invention and cannot be understood as limiting the scope of protection of the present invention. Some non-essential improvements and adjustments made by technical personnel in this field based on the above invention content still fall within the scope of protection of the present invention.

[0083] like Figure 1 As shown, this embodiment provides a hybrid control method for a photovoltaic tracking system based on double-layer nested irradiance prediction, and the control method includes the following steps:

[0084] Select a tracking control algorithm based on weather conditions; if it is sunny and dusty, choose a closed-loop control algorithm using a UV sensor; if it is cloudy, dusty or rainy, choose an open-loop control algorithm that includes tracking motor energy consumption constraints.

[0085] The definition of weather conditions in this patent is shown in Table 1 below:

[0086] Table 1 Weather conditions classification table

[0087]

[0088] Adopting an open-loop control algorithm that includes tracking the motor energy consumption constraint includes the following steps:

[0089] Get historical data of irradiance coefficient, including: normal irradiance coefficient received by photovoltaic panels , the irradiance diffusion coefficient of sunlight on the front and back of photovoltaic panels , , the overall irradiance coefficient of sunlight on the ground , ground area , , The reflected irradiance coefficient of sunlight , , , , The area on the ground directly reached by sunlight. The shadow area of ​​the ground.

[0090] Construct a double-layer nested irradiance prediction model based on GM(1,1)-DMC, such as Figure 2 As shown, the irradiation coefficient is input, and the irradiation coefficient prediction value is obtained by using the GM (1,1) model. The irradiation coefficient prediction value obtained by the GM (1,1) model is optimized by using the dynamic matrix predictive control DMC algorithm to obtain the optimized irradiation coefficient prediction value, including the following steps:

[0091] Input the original data of irradiance coefficient and generate the irradiance coefficient data matrix :

[0092] ,

[0093] For each row vector in the irradiance coefficient data matrix , , , , , , The elements in the matrix are averaged and used as known data to generate the radiation coefficient sequence matrix. ;

[0094] For the irradiance coefficient sequence matrix Perform 1-AGO accumulation processing to generate the irradiance coefficient whitening matrix , the processing expression is:

[0095]

[0096] In the formula, is the row vector of the irradiance coefficient sequence matrix No. elements, u is the number of column vectors in the irradiance coefficient sequence matrix, is the row vector of the irradiance whitening matrix No. elements, is the type of irradiation coefficient;

[0097] Perform sliding average processing on the irradiance coefficient whitening matrix to generate the irradiance coefficient grayscale matrix , the processing expression is:

[0098] ,

[0099] In the formula, is the row vector of the irradiance gray matrix No. elements;

[0100] The least square method is used to solve the development coefficient m and gray generation amount n of the GM (1,1) model. The expression is:

[0101] ,

[0102] ,

[0103] ,

[0104] In the formula, , is the constructed data matrix, which is used to solve the development coefficient m and the grey generation amount n; , They are respectively the irradiance coefficient gray matrix and the irradiance coefficient sequence matrix row vector The u-th element of ;

[0105] Substitute the solved development coefficient m and gray generation n into the following matrix equation to solve and obtain the predicted value of irradiation coefficient:

[0106] ,

[0107] ,

[0108] ,

[0109] In the formula, is the first whitening matrix of the irradiance coefficient column vector, is the first column vector of the irradiance whitening matrix.

[0110] Since the irradiation coefficient of the photovoltaic tracking system will fluctuate due to some random factors, and the GM (1,1) model is extremely sensitive to random fluctuations in data, the dynamic matrix predictive control DMC algorithm is used to optimize the irradiation coefficient prediction value obtained by the GM (1,1) model to obtain the optimized irradiation coefficient prediction value, as follows:

[0111] Step 1: Predict the overall radiation coefficient at time t+T in the future ,Right now

[0112] ,

[0113] Step 2: predict the overall radiation coefficient at time t+1+T in the future ,Right now

[0114] ,

[0115] Step 3: For the interval [t+2, t+T], the predicted overall irradiance coefficient at time t+1 is subtracted from the predicted overall irradiance coefficient at time t to obtain the prediction error ,Right now

[0116] ,

[0117] Step 4: convert the prediction error Input into the GM(1,1) model, predict the error at the future time [t+2, t+T], and update the predicted value of the overall irradiance coefficient at time t+2 ,Right now

[0118] ,

[0119] In the formula is the predicted value of the overall irradiance coefficient at time t+2 before the update, is the prediction error at time t+2;

[0120] Step 5, repeat steps 2 to 4 to complete the prediction value optimization of a complete cycle T and obtain the optimized irradiance coefficient prediction value.

[0121] Construct an irradiance model for the photovoltaic tracking system, and use the optimized irradiance coefficient prediction value through the irradiance model to obtain the predicted total irradiance data of the photovoltaic panel;

[0122] The irradiance model is as follows:

[0123] ,

[0124] In the formula, represents the total irradiance of the photovoltaic panel, , , They represent the direct irradiance, diffuse irradiance and ground reflected irradiance received by the photovoltaic panels respectively.

[0125] The direct irradiance received by the photovoltaic panel is calculated as follows:

[0126] ,

[0127] in, represents the normal irradiance coefficient received by the photovoltaic panel, is the incident angle of the sun's rays.

[0128] The formula for calculating the diffuse irradiance received by the photovoltaic panel is as follows:

[0129] ,

[0130] ,

[0131] ,

[0132] in, , Respectively represent the diffuse irradiance received by the front and back sides of the photovoltaic panel; is the positive conversion factor, whose value is provided by the photovoltaic panel manufacturer; , are the irradiance diffusion coefficients of sunlight on the front and back of the photovoltaic panel respectively; is the length of the photovoltaic panel, is the distance between the photovoltaic arrays; is the angle between the photovoltaic panel and the horizontal plane.

[0133] The calculation formula for the ground reflected irradiance received by the photovoltaic panel is as follows:

[0134] ,

[0135] ,

[0136] ,

[0137] in, , Respectively represent the ground reflected radiation received by the front and back of the photovoltaic panel, is the overall irradiance coefficient of sunlight on the ground, is the height of the photovoltaic panel from the ground, is the ground reflectivity, , , Represents the ground area , , The reflected irradiance of sunlight.

[0138] The relationship between the azimuth angle of the photovoltaic tracking system and the angle in the irradiance model is as follows:

[0139] ,

[0140] ,

[0141] In the formula, , are the azimuth and altitude angles of the sun’s rays relative to the photovoltaic panel, Represents the angle of the photovoltaic panel relative to the plane on which it is installed.

[0142] The predicted total irradiance data of the photovoltaic panel is used as the input of the open-loop control algorithm to output the optimal azimuth angle of the photovoltaic panel;

[0143] (1) Open-loop control algorithm objective function;

[0144] The open-loop control algorithm aims to solve the maximum output power of the photovoltaic panel under the predicted total irradiance of the photovoltaic panel. The specific expression is as follows:

[0145] ,

[0146] In the formula, represents the output power of the photovoltaic panel under the predicted total irradiance of the photovoltaic panel; Indicates the output power of the photovoltaic panel under standard operating conditions; Indicates the irradiance of photovoltaic panels under standard operating conditions; is the temperature coefficient, is the temperature of the photovoltaic panel under standard operating conditions, is the operating temperature of the photovoltaic panel.

[0147] Due to the forced convection effect of wind, a large amount of heat will be lost from the surface of photovoltaic panels. The randomness and volatility of the heat dissipation will lead to extremely unstable temperature prediction. Therefore, a temperature sensor is used to measure the working temperature of photovoltaic panels. Conduct testing.

[0148] 2) Open-loop control algorithm constraints;

[0149] Solving the objective function of the open-loop control algorithm aims to obtain the maximum output power of the photovoltaic panel under the predicted total irradiance of the photovoltaic panel and the corresponding azimuth angle of the photovoltaic panel. ; In order to consider the energy consumption of the tracking motor, the present invention introduces a constraint condition, which solves the maximum output power of the photovoltaic panel Subtract the tracking motor movement to the azimuth The power consumption is calculated to obtain the predicted net power under the total irradiance of the photovoltaic panel. If the net power is greater than the output power of the tracking motor in the current state, the tracking motor works, otherwise the tracking motor does not work. The constraint expression is as follows:

[0150] ,

[0151] To track the energy consumption of the motor, its value is obtained by testing; To track the output power of the motor in its current state.

[0152] Under certain weather conditions, the present invention uses an ultraviolet sensor to achieve closed-loop control of the azimuth angle of the photovoltaic panel. The closed-loop control algorithm using the ultraviolet sensor includes the following steps:

[0153] Ultraviolet sensors are used to detect the real-time irradiance on the left and right sides of the photovoltaic panel, and the real-time irradiance difference signal on the left and right sides of the photovoltaic panel is used as the input of the closed-loop control algorithm to dynamically adjust the azimuth of the photovoltaic panel. In addition, considering the energy consumption of closed-loop control information decision processing, in order to optimize the energy conversion efficiency, closed-loop control is only used under relatively bad weather conditions. The closed-loop control algorithm implements regulation based on the real-time irradiance difference signal transmitted by the ultraviolet sensors installed on the left and right sides of the photovoltaic panel. In order to reduce the over-sensitivity of the closed-loop control algorithm and avoid misoperation, the constraints of the closed-loop control algorithm adaptive tracking are defined. If the constraints are met, the photovoltaic panel will adaptively adjust the azimuth, otherwise it will maintain the position, as follows:

[0154] ,

[0155] In the formula, , Respectively represent the ultraviolet irradiance signals on the left and right sides of the photovoltaic panel, is the working critical value, obtained from actual tests.

[0156] The present invention can switch the tracking control method according to weather conditions, and comprehensively considers the energy consumption and tracking performance of the photovoltaic tracking system. At present, the photovoltaic tracking system mainly relies on the open-loop tracking control technology that predicts the position of the sun, without considering the energy consumption of the tracking motor; and the traditional closed-loop tracking control algorithm usually uses light-dependent resistors to detect visible light spectrum transmission signals, which will greatly reduce the tracking performance of the photovoltaic system under weather conditions with saturated light intensity and low visibility (rainy or dusty). In contrast, the hybrid tracking control method that combines an open-loop control algorithm that includes tracking motor energy consumption constraints with a closed-loop control algorithm that uses real-time feedback signals from ultraviolet sensors can effectively improve the energy conversion efficiency of the photovoltaic tracking system and significantly improve the accuracy of the photovoltaic tracking system.

[0157] The above only expresses the preferred implementation of the present invention, and does not limit the present invention in other forms. Any technician familiar with the profession may use the above disclosure to modify or modify it into an equivalent embodiment with equivalent changes. However, any simple modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still belongs to the protection scope of the technical solution of the present invention.

Claims

1. A hybrid control method for photovoltaic tracking system based on double-layer nested irradiance prediction, characterized in that: The control method comprises the following steps: Select a tracking control algorithm based on weather conditions; if it is sunny and dusty, choose a closed-loop control algorithm using a UV sensor; if it is cloudy, dusty or rainy, choose an open-loop control algorithm that includes tracking motor energy consumption constraints; Adopting an open-loop control algorithm that includes tracking the motor energy consumption constraint includes the following steps: Obtain historical data of irradiance coefficient; A double-layer nested irradiance prediction model based on GM(1,1)-DMC is constructed. The irradiation coefficient is input, and the irradiation coefficient prediction value is obtained using the GM(1,1) model. The irradiation coefficient prediction value obtained by the GM(1,1) model is optimized using the dynamic matrix predictive control DMC algorithm to obtain the optimized irradiation coefficient prediction value. Construct an irradiance model for the photovoltaic tracking system, and use the optimized irradiance coefficient prediction value through the irradiance model to obtain the predicted total irradiance data of the photovoltaic panel; The predicted total irradiance data of the photovoltaic panel is used as the input of the open-loop control algorithm to output the optimal azimuth angle of the photovoltaic panel; The closed-loop control algorithm using the UV sensor consists of the following steps: An ultraviolet sensor is used to detect the real-time irradiance on the left and right sides of the photovoltaic panel, and the real-time irradiance difference signal on the left and right sides of the photovoltaic panel is used as the input of the closed-loop control algorithm to dynamically adjust the azimuth angle of the photovoltaic panel.

2. The hybrid control method for photovoltaic tracking system based on double-layer nested irradiance prediction according to claim 1 is characterized in that: The irradiance coefficient includes: the normal irradiance coefficient received by the photovoltaic panel , the irradiance diffusion coefficient of sunlight on the front and back of photovoltaic panels , , the overall irradiance coefficient of sunlight on the ground , ground area , , The reflected irradiance coefficient of sunlight , , , , The area on the ground directly reached by sunlight. The shadow area of ​​the ground.

3. The hybrid control method for photovoltaic tracking system based on double-layer nested irradiance prediction according to claim 2 is characterized in that: Input the radiation coefficient, use the GM (1,1) model to obtain the radiation coefficient prediction value, use the dynamic matrix predictive control DMC algorithm to optimize the radiation coefficient prediction value obtained by the GM (1,1) model, and obtain the optimized radiation coefficient prediction value, including the following steps: Input the original data of irradiance coefficient and generate the irradiance coefficient data matrix : ， For each row vector in the irradiance coefficient data matrix , , , , , , The elements in the matrix are averaged and used as known data to generate the radiation coefficient sequence matrix. ; For the irradiance coefficient sequence matrix Perform 1-AGO accumulation processing to generate the irradiance coefficient whitening matrix , the processing expression is: ， In the formula, is the row vector of the irradiance coefficient sequence matrix No. elements, u is the number of column vectors in the irradiance coefficient sequence matrix, is the row vector of the irradiance whitening matrix No. elements, is the type of irradiation coefficient; Perform sliding average processing on the irradiance coefficient whitening matrix to generate the irradiance coefficient grayscale matrix , the processing expression is: ， In the formula, is the row vector of the irradiance gray matrix No. elements; The least square method is used to solve the development coefficient m and gray generation amount n of the GM (1,1) model. The expression is: ， ， ， In the formula, , is the constructed data matrix, which is used to solve the development coefficient m and the grey generation amount n; , They are respectively the irradiance coefficient gray matrix and the irradiance coefficient sequence matrix row vector No. elements; Substitute the solved development coefficient m and gray generation n into the following matrix equation to solve and obtain the predicted value of irradiation coefficient: ， ， ， In the formula, is the first whitening matrix of the irradiance coefficient column vector, is the first column vector of the irradiance whitening matrix; The dynamic matrix predictive control DMC algorithm is used to optimize the irradiation coefficient prediction value obtained by the GM (1,1) model to obtain the optimized irradiation coefficient prediction value, as follows: Step 1: Predict the overall radiation coefficient at time t+T in the future ,Right now ， Step 2: predict the overall radiation coefficient at time t+1+T in the future ,Right now ， Step 3: For the interval [t+2, t+T], the predicted overall irradiation coefficient at time t+1 is subtracted from the predicted overall irradiation coefficient at time t to obtain the prediction error ,Right now ， Step 4: Convert the prediction error Input into the GM(1,1) model, predict the error at the future time [t+2, t+T], and update the predicted value of the overall irradiance coefficient at time t+2 ,Right now ， In the formula is the predicted value of the overall irradiance coefficient at time t+2 before the update, is the prediction error at time t+2; Step 5, repeat steps 2 to 4 to complete the prediction value optimization of a complete cycle T.

4. The hybrid control method for photovoltaic tracking system based on double-layer nested irradiance prediction according to claim 2 is characterized in that: The irradiance model is as follows: ， In the formula, represents the total irradiance of the photovoltaic panel, , , They represent the direct irradiance, diffuse irradiance and ground reflected irradiance received by the photovoltaic panels respectively.

5. The hybrid control method for photovoltaic tracking system based on double-layer nested irradiance prediction according to claim 4 is characterized in that: The direct irradiance received by the photovoltaic panel is calculated as follows: ， in, represents the normal irradiance coefficient received by the photovoltaic panel, is the incident angle of the sun's rays.

6. The hybrid control method for photovoltaic tracking system based on double-layer nested irradiance prediction according to claim 5 is characterized in that: The formula for calculating the diffuse irradiance received by the photovoltaic panel is as follows: ， ， ， in, , Respectively represent the diffuse irradiance received by the front and back sides of the photovoltaic panel; is the positive conversion factor; , are the irradiance diffusion coefficients of sunlight on the front and back of the photovoltaic panel respectively; is the length of the photovoltaic panel, is the distance between the photovoltaic arrays; is the angle between the photovoltaic panel and the horizontal plane.

7. The hybrid control method for photovoltaic tracking system based on double-layer nested irradiance prediction according to claim 6 is characterized in that: The calculation formula for the ground reflected irradiance received by the photovoltaic panel is as follows: ， ， ， in, , Respectively represent the ground reflected radiation received by the front and back of the photovoltaic panel, is the overall irradiance coefficient of sunlight on the ground, is the height of the photovoltaic panel from the ground, is the ground reflectivity, , , Represents the ground area , , The reflected irradiance of sunlight.

8. The hybrid control method for photovoltaic tracking system based on double-layer nested irradiance prediction according to claim 7 is characterized in that: The relationship between the azimuth angle of the photovoltaic tracking system and the angle in the irradiance model is as follows: ， ， In the formula, , are the azimuth and altitude angles of the sun’s rays relative to the photovoltaic panel, Represents the angle of the photovoltaic panel relative to the plane on which it is installed.

9. The hybrid control method for photovoltaic tracking system based on double-layer nested irradiance prediction according to claim 8 is characterized in that: The predicted total irradiance data of the photovoltaic panel is used as the input of the open-loop control algorithm to output the optimal azimuth angle of the photovoltaic panel, including the following steps: The open-loop control algorithm aims to solve the maximum output power of the photovoltaic panel under the predicted total irradiance of the photovoltaic panel. The objective function expression of the open-loop control algorithm is as follows: ， In the formula, represents the output power of the photovoltaic panel under the predicted total irradiance of the photovoltaic panel; Indicates the output power of the photovoltaic panel under standard operating conditions; Indicates the irradiance of photovoltaic panels under standard operating conditions; is the temperature coefficient, is the temperature of the photovoltaic panel under standard operating conditions, is the operating temperature of the photovoltaic panel; Introducing constraints, the maximum output power of the photovoltaic panel obtained is Subtract the tracking motor movement to the azimuth The power consumption is calculated to obtain the predicted net power under the total irradiance of the photovoltaic panel. If the net power is greater than the output power of the tracking motor in the current state, the tracking motor works, otherwise the tracking motor does not work. The constraint expression is as follows: ， To track the energy consumption of the motor; To track the output power of the motor in its current state.

10. The hybrid control method for photovoltaic tracking system based on double-layer nested irradiance prediction according to claim 1, characterized in that: Closed-loop control algorithm using UV sensor, including: Define the constraints of the closed-loop control algorithm for adaptive tracking. If the constraints are met, the photovoltaic panel will adaptively adjust the azimuth angle, otherwise it will maintain the position, as follows: ， In the formula, , Respectively represent the ultraviolet irradiance signals on the left and right sides of the photovoltaic panel, is the working critical value.

Images