Component temperature prediction method during photovoltaic grid-connected power generation

Abstract

The invention discloses a module temperature prediction method during photovoltaic grid-connected power generation, and the method comprises the steps: building a prediction model of the temperature of a photovoltaic module during the photovoltaic grid-connected power generation based on a thermodynamic law, a Joule law and a Fourier law; temperature data of a photovoltaic module and key environmental factors influencing the temperature of the module during grid-connected photovoltaic power generation are acquired, and a Levenberg-Marquardt iterative algorithm is utilized to perform nonlinear function fitting to obtain related parameters of a prediction model, so that a prediction algorithm for predicting the temperature of the module during grid-connected photovoltaic power generation is obtained. In practical application, the temperature of the photovoltaic module can be predicted by substituting the air temperature of the surrounding environment of the photovoltaic module and the total irradiation amount of the photovoltaic module into the prediction algorithm. The power and the performance of the photovoltaic module are improved, the service life of the photovoltaic module is prolonged, and the operation stability and the power generation efficiency of the whole photovoltaic system are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic grid-connected power generation, and in particular to a component temperature prediction method during photovoltaic grid-connected power generation. Background Art

[0002] Photovoltaic grid-connected power generation refers to a method of converting the direct current generated by the solar photovoltaic power generation system into alternating current compatible with the power grid through an inverter, and then connecting it to the public power grid for power supply. This system can be divided into two forms: centralized and distributed. Centralized photovoltaic power generation usually consists of large photovoltaic power stations, while distributed photovoltaic power generation is mainly small-scale systems, which are common in building-integrated photovoltaic projects. In short, photovoltaic grid-connected power generation is widely used for its advantages such as reducing greenhouse gas emissions, reducing system maintenance costs and environmental pollution risks, and helping to balance the load of the power grid and reduce line losses.

[0003] However, the photovoltaic grid-connected power generation system also faces some challenges and problems in actual operation. Among them, the temperature control of photovoltaic modules is a key factor. The output power of photovoltaic modules will decrease with the increase of temperature. In theory, the power generation will decrease by about 0.5% for every degree increase in temperature. High temperature not only affects the power and performance of the module, but may also cause hot spot effect, thus affecting the life of photovoltaic modules. Therefore, the prediction of photovoltaic module temperature is particularly important for the system design of photovoltaic grid-connected power generation.

[0004] At present, although some studies have been conducted on the prediction of photovoltaic power generation, most methods still do not consider the change of component temperature, but directly use the ambient temperature as the operating temperature, which limits the improvement of prediction accuracy. Summary of the invention

[0005] The purpose of the present invention is to solve the technical problem in the above-mentioned prior art that the temperature changes of photovoltaic modules during photovoltaic grid-connected power generation cannot be predicted, and to provide a component temperature prediction method during photovoltaic grid-connected power generation, which can predict the temperature of photovoltaic modules during photovoltaic grid-connected power generation and improve the stability of photovoltaic power generation system operation.

[0006] In order to solve the above problems, the present invention is implemented according to the following technical solutions:

[0007] The present invention provides a method for predicting component temperature during photovoltaic grid-connected power generation, the method comprising:

[0008] Step 100, obtaining temperature data of photovoltaic modules;

[0009] Step 200, obtaining key environmental factors that affect the temperature of components during photovoltaic grid-connected power generation, wherein the key environmental factors include solar radiation data and air temperature data of the environment surrounding the photovoltaic components;

[0010] Step 300: Based on the laws of thermodynamics, Joule's law and Fourier's law, a prediction model for the temperature of photovoltaic modules during photovoltaic grid-connected power generation is constructed. The prediction model is T=T0+A×H+B×H 2 , where T is the temperature of the photovoltaic module, T0 is the air temperature of the environment around the photovoltaic module, H is the total radiation received by the photovoltaic module, and A and B are parameters related to the physical characteristics of the photovoltaic module itself;

[0011] Step 400, based on the temperature data in step 100 and the key environmental factors in step 200, calculate the values ​​of A and B in step 300;

[0012] Step 500: Substitute the values ​​of A and B obtained in step 400 into the prediction model in step 300 to obtain a prediction algorithm that can be used to predict the component temperature during photovoltaic grid-connected power generation.

[0013] Preferably, in step 200, the solar radiation data is obtained by installing a total irradiance meter in the area where the photovoltaic system is located, and the air temperature data is obtained by installing a meteorological temperature monitor in the area where the photovoltaic system is located; wherein the total irradiance meter is installed at the same angle as the photovoltaic module.

[0014] Preferably, in step 300, the step of constructing a prediction model for the temperature of photovoltaic modules during photovoltaic grid-connected power generation based on the laws of thermodynamics, Joule's law and Fourier's law includes: step 301, establishing a mathematical model ΔT1=A×H of the temperature increase of photovoltaic modules caused by solar radiation, wherein ΔT1 is the temperature increase of the photovoltaic modules caused by solar radiation, H is the total radiation received by the photovoltaic modules, and A is the parameter related to the physical properties of the photovoltaic modules themselves.

[0015] Preferably, step 302, establish a mathematical model of the temperature rise of photovoltaic modules caused by photovoltaic grid-connected power generation ΔT2 = B × H 2 , wherein ΔT2 is the temperature rise of the photovoltaic module caused by photovoltaic grid-connected power generation, and B is the parameter related to the physical characteristics of the photovoltaic module itself.

[0016] Preferably, step 303, constructing a mathematical model T=T0+ΔT1+ΔT2 related to the temperature of the photovoltaic module during grid-connected power generation;

[0017] Preferably, based on step 301, step 302 and step 303, a prediction model T=T0+A×H+B×H for the temperature of the photovoltaic module during photovoltaic grid-connected power generation is obtained. 2 .

[0018] Preferably, in step 300, the calculation formula of A is A=S÷(c×m), wherein S is the surface area of ​​the photovoltaic module, c is the specific heat capacity of the photovoltaic module, and m is the mass of the photovoltaic module.

[0019] Preferably, in step 300, the calculation formula of B is B=k 2 ×R×δ÷(λ×S), where λ is the thermal conductivity of the photovoltaic module, S is the surface area of ​​the photovoltaic module, δ is the heat transfer distance, R is the resistance of the photovoltaic module, and k is the power generation efficiency of the photovoltaic module.

[0020] Preferably, in step 400, based on the temperature data in step 100 and the key environmental factors in step 200, nonlinear function fitting is performed by using the Levenberg-Marquardt iterative algorithm to calculate the values ​​of A and B in step 300.

[0021] Compared with the prior art, the present invention has the following beneficial effects:

[0022] The present invention provides a method for predicting the temperature of a component during photovoltaic grid-connected power generation. Based on the laws of thermodynamics, Joule's law and Fourier's law, a prediction model for the temperature of a photovoltaic component during photovoltaic grid-connected power generation is constructed. The temperature data of the photovoltaic component and the key environmental factors affecting the temperature of the component during photovoltaic grid-connected power generation are obtained, and the relevant parameters of the prediction model are obtained by performing nonlinear function fitting using the Levenberg-Marquardt iterative algorithm, so as to obtain a prediction algorithm that can be used to predict the temperature of the component during photovoltaic grid-connected power generation. In practical applications, the temperature of the photovoltaic component can be predicted by substituting the air temperature of the surrounding environment of the photovoltaic component and the total radiation amount received by the photovoltaic component into the prediction algorithm. The component temperature during photovoltaic grid-connected power generation is predicted, and decisions related to the photovoltaic system can be made based on the predicted temperature, which is beneficial to improving the power and performance of the photovoltaic component, extending the life of the photovoltaic component, and also beneficial to improving the stability of the operation of the entire photovoltaic system and the power generation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The specific embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings, wherein:

[0024] Figure 1 It is a principle block diagram of a method for predicting component temperature during photovoltaic grid-connected power generation according to the present invention;

[0025] Figure 2 It is an implementation flow chart of a method for predicting component temperature during photovoltaic grid-connected power generation according to the present invention. DETAILED DESCRIPTION

[0026] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0027] The terms used in this application are only for the purpose of describing specific embodiments and are not intended to limit this application. Unless otherwise defined, the technical terms or scientific terms used in this specification should be understood by people with ordinary skills in the field to which this application belongs. The words "first", "second" and similar words used in this specification and claims do not indicate any order, quantity or importance, but are only used to distinguish different technical features.

[0028] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0029] like Figure 2 As shown, the method for predicting component temperature during photovoltaic grid-connected power generation described in the present invention includes:

[0030] Step 100: Acquire temperature data of a photovoltaic module.

[0031] In this embodiment, the photovoltaic modules are installed at the optimal inclination angle of the region where they are located, and the photovoltaic grid-connected power generation is operated. By arranging temperature probes on the surface of the photovoltaic modules, the temperature data of the photovoltaic modules can be monitored and obtained in real time.

[0032] In a specific embodiment, high-precision thermocouple detection points are evenly attached to the surfaces of all photovoltaic modules in the photovoltaic system to monitor and collect module temperature data in real time, with a collection time interval of 1 minute for a set of data. The temperature collection data of some modules are shown in Table 1 below:

[0033] Table 1

[0034] Air temperature (℃) <![CDATA[Irradiation dose (kWh / m 2 )]]> Surface temperature(℃) 28.95095 5.41546 40.33505 29.16147 3.99118 39.18614 28.07591 3.05411 37.3561 31.22401 3.13544 41.15542 31.29969 2.64183 39.20668 28.26102 2.41138 38.72387 28.61314 2.26475 37.73192 29.44606 2.51874 39.23787 30.61732 3.23679 41.51799

[0035] Step 200: Acquire key environmental factors that affect the temperature of photovoltaic components during grid-connected power generation. The key environmental factors include solar radiation data and air temperature data of the environment surrounding the photovoltaic components.

[0036] It should be noted that the solar radiation data and the surrounding air temperature data used to construct the solar radiation data that affect the temperature of photovoltaic modules can be the radiation data and the surrounding air temperature at the time point of interest, or the average radiation data and the average air temperature data of the surrounding environment in the time domain of interest.

[0037] In this embodiment, a total irradiance meter is installed in the area where the photovoltaic system is located to obtain solar irradiance data, and a meteorological temperature monitor is installed in the area where the photovoltaic system is located to obtain air temperature data. Preferably, the total irradiance meter is installed at the same angle as the photovoltaic module.

[0038] Step 300: Based on the laws of thermodynamics, Joule's law and Fourier's law, a prediction model for the temperature of photovoltaic modules during photovoltaic grid-connected power generation is constructed. The prediction model is T = T0 + A × H + B × H 2 , where T is the temperature of the PV module, T0 is the air temperature around the PV module, H is the total radiation received by the PV module, and A and B are parameters related to the physical properties of the PV module itself.

[0039] Step 301, establish a mathematical model of the temperature increase of the photovoltaic module caused by solar radiation: ΔT1=A×H, where ΔT1 is the temperature increase of the photovoltaic module caused by solar radiation, H is the total radiation received by the photovoltaic module, and A is a parameter related to the physical characteristics of the photovoltaic module itself.

[0040] It should be noted that according to the laws of thermodynamics, the total radiation H received by a photovoltaic module causes its temperature to rise by ΔT1, which satisfies the following formula: H×S=c×m×ΔT1, that is, ΔT1=H×S÷(c×m), where S is the surface area of ​​the photovoltaic module, c is the specific heat capacity of the photovoltaic module, and m is the mass of the photovoltaic module. For the same type of photovoltaic modules, S, c, and m are constants, that is, A=S÷(c×m). Therefore, the mathematical model for the temperature rise of photovoltaic modules caused by solar radiation is ΔT1=A×H.

[0041] Step 302: Establish a mathematical model of the temperature rise of photovoltaic modules caused by photovoltaic grid-connected power generation: ΔT2 = B × H 2 , where ΔT2 is the temperature rise of the photovoltaic module caused by photovoltaic grid-connected power generation, and B is a parameter related to the physical characteristics of the photovoltaic module itself.

[0042] It should be noted that (1) the photovoltaic system power generation current I is proportional to the total radiation H received by the photovoltaic module, satisfying the following relationship: I = k × H. (2) According to Joule's law, the relationship between the heat generation Q and the power generation current I during photovoltaic grid-connected power generation satisfies the following formula: Q = I 2×R×t, where R is the resistance of the photovoltaic module and t is time. (3) According to Fourier's law, the relationship between the heat Q generated during photovoltaic grid-connected power generation and the temperature increase ΔT2 of the photovoltaic module satisfies the following formula: Q=λ×S×t×ΔT2÷δ, where λ is the thermal conductivity of the photovoltaic module, S is the surface area of ​​the photovoltaic module, t is time, and δ is the heat transfer distance. (4) According to the formulas (1)-(3), it can be deduced that the temperature increase ΔT2 of the photovoltaic module caused by photovoltaic grid-connected power generation and the total radiation H received by the photovoltaic module satisfy the following formula: ΔT2=H 2 ×k 2 ×R×δ÷(λ×S). For components of the same type, k, R, δ, λ, and S are constants, so B=k 2 ×R×δ÷(λ×S). Therefore, the mathematical model of the temperature rise of photovoltaic modules caused by photovoltaic grid-connected power generation is established as follows: ΔT2=B×H 2 .

[0043] Step 303: construct a mathematical model T=T0+ΔT1+ΔT2 related to the temperature of the photovoltaic modules during grid-connected power generation.

[0044] Step 304: Based on step 301, step 302 and step 303, a prediction model T=T0+A×H+B×H for the temperature of the photovoltaic module during photovoltaic grid-connected power generation is obtained. 2 .

[0045] Step 400 : Based on the temperature data in step 100 and the key environmental factors in step 200 , the values ​​of A and B in step 300 are calculated.

[0046] Preferably, the values ​​of A and B in step 300 are calculated by performing nonlinear function fitting using a Levenberg-Marquardt iterative algorithm.

[0047] It should be noted that the Levenberg-Marquardt iterative algorithm is an optimization algorithm widely used in nonlinear least squares problems. It was first proposed by Levenberg in 1944, and then re-proposed and theoretically explored by Marquardt in 1963. The algorithm combines the advantages of the Gauss-Newton method and the gradient descent method. By introducing an adjustment parameter λ (lambda), it is dynamically adjusted in each step of the algorithm to balance the convergence speed and stability of the algorithm. In the Levenberg-Marquardt algorithm, when λ is very small, the step size of the algorithm is close to the step size of the Newton method, which means that the algorithm can converge quickly; when λ is very large, the step size is close to the step size of the gradient descent method, which helps to ensure the stability of the algorithm. This feature makes the Levenberg-Marquardt algorithm very robust when dealing with nonlinear least squares problems. The use of the Levenberg-Marquardt iterative algorithm for nonlinear function fitting includes the following aspects: first, it can achieve fast convergence while ensuring stability; second, it can adapt to different nonlinear problems and work effectively even when the initial parameter estimates are inaccurate; third, by adjusting the λ value, a trade-off can be made between the convergence speed and stability of the algorithm to adapt to different optimization problems; fourth, by adjusting the λ value, a trade-off can be made between the convergence speed and stability of the algorithm to adapt to different optimization problems; in other aspects, the implementation of the algorithm is relatively simple. In general, the nonlinear function fitting using the Levenberg-Marquardt algorithm is efficient and stable.

[0048] In a specific embodiment, based on the data in Table 1 in the above step 100, a nonlinear function fitting is performed using the Levenberg-Marquardt iterative algorithm to obtain A=3.829 and B=-0.192.

[0049] Step 500: Substitute the values ​​of A and B obtained in step 400 into the prediction model in step 300 to obtain a prediction algorithm that can be used to predict the component temperature during photovoltaic grid-connected power generation.

[0050] In a specific embodiment, A=3.829, B=-0.192 are substituted into the prediction model in step 300, and the prediction algorithm for predicting the component temperature during photovoltaic grid-connected power generation is obtained as T=T0+3.829×H-0.192×H 2 .

[0051] In an actual application scenario, as shown in Table 2 below, the daily average solar radiation, daily average air temperature and corresponding surface temperature of a certain area for multiple days are substituted into the prediction algorithm T = T0 + 3.829 × H - 0.192 × H 2 The surface temperature and prediction accuracy of this type of photovoltaic module when it is connected to the grid every day in the area can be predicted, as shown in Table 3 below. The prediction accuracy of each group is above 95%, and the average prediction accuracy is 97.73%.

[0052] Table 2

[0053] Air temperature (℃) <![CDATA[Irradiation dose (kWh / m 2 )]]> Surface temperature(℃) 30.74226 4.05539 43.68095 30.60876 4.34137 44.01865 29.82515 3.29886 38.76868 29.30703 3.72653 41.00113 31.80914 4.87068 47.23879 32.1834 5.29411 45.16978

[0054] Table 3

[0055]

[0056]

[0057] In summary, the present invention provides a method for predicting the temperature of a component during photovoltaic grid-connected power generation. Based on the laws of thermodynamics, Joule's law and Fourier's law, a prediction model for the temperature of a photovoltaic component during photovoltaic grid-connected power generation is constructed. The temperature data of the photovoltaic component and the key environmental factors affecting the temperature of the component during photovoltaic grid-connected power generation are obtained, and the relevant parameters of the prediction model are obtained by performing nonlinear function fitting using the Levenberg-Marquardt iterative algorithm, so as to obtain a prediction algorithm that can be used to predict the temperature of the component during photovoltaic grid-connected power generation. In practical applications, the temperature of the photovoltaic component can be predicted by substituting the air temperature of the surrounding environment of the photovoltaic component and the total radiation received by the photovoltaic component into the prediction algorithm. The component temperature during photovoltaic grid-connected power generation is predicted, and decisions related to the photovoltaic system can be made based on the predicted temperature, which is beneficial to improving the power and performance of the photovoltaic component, extending the life of the photovoltaic component, and also beneficial to improving the stability of the operation of the entire photovoltaic system and the power generation efficiency.

[0058] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Therefore, any modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention without departing from the technical solution of the present invention shall still fall within the scope of the technical solution of the present invention.

Claims

1. A method for predicting component temperature during photovoltaic grid-connected power generation, characterized in that: The method comprises: Step 100, obtaining temperature data of photovoltaic modules; Step 200, obtaining key environmental factors that affect the temperature of components during photovoltaic grid-connected power generation, wherein the key environmental factors include solar radiation data and air temperature data of the environment surrounding the photovoltaic components; Step 300: Based on the laws of thermodynamics, Joule's law and Fourier's law, a prediction model for the temperature of photovoltaic modules during photovoltaic grid-connected power generation is constructed. The prediction model is T=T0+A×H+B×H 2 , where T is the temperature of the photovoltaic module, T0 is the air temperature of the environment around the photovoltaic module, H is the total radiation received by the photovoltaic module, and A and B are parameters related to the physical characteristics of the photovoltaic module itself; Step 400, based on the temperature data in step 100 and the key environmental factors in step 200, calculate the values ​​of A and B in step 300; Step 500: Substitute the values ​​of A and B obtained in step 400 into the prediction model in step 300 to obtain a prediction algorithm that can be used to predict the component temperature during photovoltaic grid-connected power generation.

2. The method for predicting component temperature during photovoltaic grid-connected power generation according to claim 1, characterized in that: In step 200, the solar radiation data is obtained by installing a total irradiance meter in the area where the photovoltaic system is located, and the air temperature data is obtained by installing a meteorological temperature monitor in the area where the photovoltaic system is located; wherein the total irradiance meter is installed at the same angle as the photovoltaic module.

3. The method for predicting component temperature during photovoltaic grid-connected power generation according to claim 1, characterized in that: In step 300, the step of constructing a prediction model for the temperature of photovoltaic modules during photovoltaic grid-connected power generation based on the laws of thermodynamics, Joule's law and Fourier's law includes: Step 301, establish a mathematical model of the temperature increase of the photovoltaic module caused by solar radiation: ΔT1=A×H, wherein ΔT1 is the temperature increase of the photovoltaic module caused by solar radiation, H is the total radiation received by the photovoltaic module, and A is the parameter related to the physical characteristics of the photovoltaic module itself.

4. The method for predicting component temperature during photovoltaic grid-connected power generation according to claim 3, characterized in that: Step 302: Establish a mathematical model of the temperature rise of photovoltaic modules caused by photovoltaic grid-connected power generation: ΔT2 = B × H 2 , wherein ΔT2 is the temperature rise of the photovoltaic module caused by photovoltaic grid-connected power generation, and B is the parameter related to the physical characteristics of the photovoltaic module itself.

5. The method for predicting component temperature during photovoltaic grid-connected power generation according to claim 4, characterized in that: Step 303, constructing a mathematical model T=T0+ΔT1+ΔT2 related to the temperature of the photovoltaic module during grid-connected power generation; Step 304: Based on step 301, step 302 and step 303, a prediction model T=T0+A×H+B×H for the temperature of the photovoltaic module during the photovoltaic grid-connected power generation is obtained. 2 .

6. The method for predicting component temperature during photovoltaic grid-connected power generation according to claim 1, characterized in that: In step 300, the calculation formula of A is A=S÷(c×m), where S is the surface area of ​​the photovoltaic module, c is the specific heat capacity of the photovoltaic module, and m is the mass of the photovoltaic module.

7. A method for predicting component temperature during photovoltaic grid-connected power generation according to claim 6, characterized in that: In step 300, the calculation formula of B is B=k 2 ×R×δ÷(λ×S), where λ is the thermal conductivity of the photovoltaic module, S is the surface area of ​​the photovoltaic module, δ is the heat transfer distance, R is the resistance of the photovoltaic module, and k is the power generation efficiency of the photovoltaic module.

8. The method for predicting component temperature during photovoltaic grid-connected power generation according to claim 1, characterized in that: In step 400, based on the temperature data in step 100 and the key environmental factors in step 200, nonlinear function fitting is performed by using the Levenberg-Marquardt iterative algorithm to calculate the values ​​of A and B in step 300.

Images