Photovoltaic power generation performance prediction device and implementation method

Abstract

The invention discloses a photovoltaic power generation performance prediction device, which is arranged in a photovoltaic power generation field, and comprises a weather acquisition device, a dust retention degree acquisition module, a same-condition glass plate, a power supply module, a power supply module and a power supply module, the inclination angle and orientation of the photovoltaic panel are equal to the supporting angles and orientations of other photovoltaic panels in the photovoltaic power generation field; the first silicon photocell is fixed on one side of the partition plate at the first darkroom and is positioned above the partition plate opening; the second silicon photocell is arranged on one side of the same-condition glass plate facing the ground; the invention further discloses an implementation method of the photovoltaic power generation performance prediction device, according to the photovoltaic power generation performance prediction device, a silicon photocell is used for completing real-time testing of the dust density of the glass plate under the same condition, the problem that data cannot be collected in time is solved, and the photovoltaic power generation performance prediction device has the advantages that the real-time testing of the dust density of the glass plate under the same condition is achieved. The data can be timely fed back and collected and can participate in the prediction of the photovoltaic performance.

Description

Technical Field

[0001] The present invention relates to the field of photovoltaic performance prediction. More specifically, the present invention relates to a device and an implementation method for predicting the power generation performance of a photovoltaic power generation system. Background Art

[0002] With the continuous development of photovoltaic technology and the gradual decline of the cost of photovoltaic power generation, solar power generation has become the clean energy with the fastest growth rate. In recent years, the global photovoltaic installed capacity has maintained a stable upward trend. Since photovoltaic power stations are strongly correlated with meteorological factors, they have large fluctuations. In order to ensure the safe and reliable operation of the power system, it is crucial to accurately predict the output power and power generation of photovoltaic power stations.

[0003] At present, prediction models can be established through methods such as the above-mentioned multiple linear regression method, artificial neural network method, and support vector machine method to predict the output power of a photovoltaic power generation system. Photovoltaic power prediction can improve the utilization efficiency of solar photovoltaic power generation resources. The above prediction methods are all based on historical meteorological data of the same cycle and the same day type.

[0004] Currently, the influence of dust accumulation on solar panels is rarely considered in most prediction models. In many prediction models, since it is difficult to collect the on-site dust accumulation situation in a timely manner and match it with the meteorological data collected in real time, for the consideration of dust accumulation in the prediction method, the number of dust accumulation days and the cleaning time or rainfall are often combined to obtain a value as a training set through correction or other means for training the prediction model. The Chinese patent with the authorization announcement number CN113393046B discloses a photovoltaic power prediction method and its application device. After determining the rainfall from the current moment to the target prediction moment, the initial power prediction value of the photovoltaic power station at the target prediction moment, and the cleaning information of the cleaning robot at the current moment, it determines the dust accumulation duration of the photovoltaic power station at the target prediction moment and determines the dust accumulation rate of the photovoltaic power station based on the rainfall and the dust accumulation duration. Finally, it corrects the initial power prediction value based on the dust accumulation rate to obtain the final power prediction value of the photovoltaic power station at the target prediction moment. The Chinese patent with the authorization announcement number CN108880470B discloses a method for calculating the influence of dust accumulation on the output power and power generation of photovoltaic modules. According to the relationship between different environments and the dust accumulation rate, it establishes the relationship between the environment and the surface cleanliness of the module and proposes a model for calculating the dust accumulation loss of the module based on the step method; the above two methods are applicable to photovoltaic fields that are often in a natural dust accumulation situation, but in the case of large sand and dust amounts and complex sand and wind conditions, it is difficult to calculate the dust density on the panel area based on the rainfall time and cleaning time, and the above devices and methods cannot be used.

[0005] In summary, dust density is an important factor affecting photovoltaic performance. Since it is difficult to obtain the actual situation in real time, dust density is often used as an input layer parameter of the photovoltaic prediction model through indirect parameters such as the time the panel is not cleaned. It is involved in the photovoltaic performance prediction. In complex wind and sand conditions, this method is difficult to be accurate. Therefore, it is urgent to propose a photovoltaic power generation performance prediction device and implementation method. Summary of the invention

[0006] One object of the present invention is to provide a photovoltaic power generation performance prediction device and implementation method. The device can be directly installed in an existing photovoltaic power field, and the prediction of photovoltaic power generation or power can be completed by conducting dust density tests on glass plates with the same setting angle and orientation as photovoltaic panels, in conjunction with a meteorological data collection device.

[0007] In order to achieve these purposes and other advantages according to the present invention, in the first aspect, the present invention provides a photovoltaic power generation performance prediction device, which is arranged in a photovoltaic power plant, including: a meteorological collection device, which records various meteorological data in the photovoltaic power plant in real time; a dust accumulation acquisition module, which includes a dust accumulation measurement box, the dust accumulation measurement box is a closed box, a partition plate is provided inside the dust accumulation measurement box, the inclination angle of the partition plate is equal to the support angle of the photovoltaic panel in the photovoltaic power plant, a partition opening is provided in the middle of the partition plate, the partition plate divides the interior of the dust accumulation measurement box into a first dark room and a second dark room, and the first dark room is equipped with a direct light source, one side wall of the dust accumulation measuring box of the second darkroom is an opening and closing structure; a same-condition glass plate, the inclination angle and orientation of the same-condition glass plate are equal to the support angle and orientation of other photovoltaic panels in the photovoltaic power field; a first silicon photocell, which is fixed to one side of the partition plate in the first darkroom, and the first silicon photocell is located above the partition opening; a second silicon photocell, which is fixed to the side of the same-condition glass plate facing the ground; wherein, when the same-condition glass plate is close to the partition plate in the second darkroom, the second silicon photocell is aligned with the center of the partition opening, and the illumination range of the direct light source can cover the first silicon photocell and the second silicon photocell at the same time.

[0008] Preferably, a six-degree-of-freedom device is installed in the first darkroom, the direct light source is installed on the movable part of the six-degree-of-freedom device, and there are at least five partition openings on the partition plate, each of which corresponds to a first silicon photocell and a second silicon photocell.

[0009] Preferably, a linear drive device is provided under the dust accumulation measuring box and the same condition glass plate, and the linear drive device includes a servo motor and a sliding seat transmission-connected thereto, and the same condition glass plate is located outside the opening and closing structure, wherein the servo motor enables the sliding seat to drive the same condition glass plate to translate from the outside to the second dark chamber until it is close to the partition plate through a matching screw.

[0010] Preferably, the inner wall and the partition board of the dust accumulation degree measuring box are coated with black light-absorbing paint.

[0011] Preferably, the meteorological collection device includes a hygrometer, a barometer, a dust sensor, a wind speed sensor, and a thermometer. The meteorological data collected by the meteorological collection device includes parameters such as relative humidity, wind speed, PM2.5 concentration, PM10 concentration, temperature, air pressure, and solar radiation value.

[0012] Preferably, it further includes: a dust accumulation degree test module, which is electrically connected to the first silicon photocell and the second silicon photocell and records the voltage ratio of the two under the same-source light illumination, and calculates the dust accumulation density on the corresponding same-condition glass plate according to the voltage ratio; a short-term power generation prediction module, which includes a model training unit, an information receiving module, and a power generation prediction unit. The short-term power generation prediction module is communicatively connected to the meteorological collection device and the dust accumulation degree test module, and can receive external weather forecast data. The model training unit can store the data collected by the meteorological collection device periodically, and uses the meteorological data collected by the meteorological collection device as a training set through a neural network learning calculation software to obtain a solar radiation amount prediction model. The information receiving module is used to obtain the weather forecast data for the prediction period and the current dust accumulation density data obtained by the dust accumulation degree test module. The short-term power generation prediction module calculates the power generation for the prediction period according to the solar radiation amount prediction model, the meteorological data for the prediction period, and the dust accumulation density data in sequence.

[0013] In a second aspect, the present invention provides an implementation method of a photovoltaic power generation performance prediction device, including the following steps: S1. Calibrate the photovoltaic power generation performance prediction device: In the laboratory, manually and evenly distribute dust on the upper side of the same-condition glass plate provided with the second silicon photocell from scratch. After dusting, attach the same-condition glass plate to the lower side of the partition board provided with the first silicon photocell, and use a direct light source to irradiate the first silicon photocell and the second silicon photocell simultaneously from the upper side of the partition board, record the output voltage ratio of the two, and establish a fitting curve of the dust accumulation density and the voltage ratio in this device; S2. Place the photovoltaic power generation performance prediction device: Place the photovoltaic power generation performance prediction device in a photovoltaic power generation field, and clean the same-condition glass plate and the photovoltaic panel synchronously; S3. Establish a solar radiation amount prediction model: Use the historical meteorological data collected by the meteorological collection device as sample set data to construct a solar radiation amount prediction model, and the solar radiation amount prediction model uses a BP neural network including an input layer, a hidden layer, and an output layer; S4. Calculate the predicted value of solar radiation: Obtain the predicted values of meteorological data for the time period to be predicted, input the elements of the predicted values of meteorological data into the solar radiation prediction model, and obtain the predicted value of solar radiation for the ultra-short term; S5. Calculate the initial dust deposition density at the beginning of the time period: Drive the same-condition glass plate into the dust deposition measurement box, obtain the voltage ratio of the first silicon photocell and the second silicon photocell at this time after being irradiated by the direct sunlight source, and calculate the dust deposition density at this time; S6. Calculate the predicted value of power generation: Based on the obtained dust ash density, calculate the conversion degradation efficiency of the dust-deposited photovoltaic module, and calculate the predicted value of power generation. The calculation formula is as follows: Where: P—the predicted value of power generation, kWh; S—the total area of the battery panels in the power plant, m²; H—the predicted value of solar radiation, kWh / m²; E d —the efficiency of the dust-deposited photovoltaic module.

[0014] Preferably, the step S5 includes the following steps: A1: Open the opening and closing structure. Wait for the driving device to pull the same-condition glass plate into the second dark room and align it with the partition board, and then close the opening and closing structure; A2: The six-degree-of-freedom device drives the preheated direct sunlight source to align it with an opening in the partition board, and the light covers the first silicon photocell and the second silicon photocell corresponding to the opening in the partition board; A3: After the output voltages of the first silicon photocell and the second silicon photocell are stable, record the output voltage ratio at this time, and obtain the dust deposition density corresponding to this voltage ratio according to the fitting curve obtained in step S1.

[0015] Preferably, the efficiency E of the dust-deposited photovoltaic module in the step S6 d The calculation formula is: Where: E c —the efficiency of the clean photovoltaic module; p m —the dust deposition density, g / m 2 .

[0016] The present invention has at least the following beneficial effects: First, the photovoltaic power generation performance prediction device of the present invention uses a same-condition glass plate with dimensions, orientation, and angle similar to those of the photovoltaic panel, which can better simulate the dust accumulation situation of the photovoltaic panel under severe sandstorm conditions, and has a better dust accumulation simulation effect compared with the current conventional sediment collection devices; Second, the photovoltaic power generation performance prediction device of the present invention uses a silicon photovoltaic cell to complete the test of the dust accumulation density, avoiding the problem of untimely data collection. The data can be fed back and collected in a timely manner and can be involved in the prediction of photovoltaic performance. Third, in the photovoltaic power generation performance prediction device of the present invention, considering that in previous photovoltaic power generation predictions, continuous data collection of dust accumulation density was basically not carried out, and the dust accumulation density could not be directly used as sample set data for the prediction model. Therefore, the direct measurement and feedback of the dust accumulation situation of the photovoltaic panel are completed through the same-condition glass plate and dust density measurement box that can be directly installed in the photovoltaic power generation field, and the efficiency of the photovoltaic panel is corrected to accurately predict the power generation.

[0017] Other advantages, objectives and features of the present invention will be partially reflected by the following description and partially understood by those skilled in the art through the research and practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a schematic diagram of the photovoltaic power generation performance prediction device in a technical solution of the present invention; Figure 2 It is a side schematic diagram of the photovoltaic power generation performance prediction device in a technical solution of the present invention; Figure 3 It is a schematic diagram of the same-condition glass plate of the photovoltaic power generation performance prediction device in a technical solution of the present invention; Figure 4 It is a working schematic diagram of the first silicon photovoltaic cell and the second silicon photovoltaic cell in a technical solution of the present invention; Figure 5 It is a schematic diagram of light illumination in a technical solution of the present invention; Figure 6 It is a structural block diagram of the photovoltaic power generation performance prediction device in a technical solution of the present invention; Figure 7 It is a working flowchart of the photovoltaic power generation performance prediction device in a technical solution of the present invention.

[0019] Legend: 1 - same-condition glass plate, 10 - linear drive device, 100 - drive device base, 101 - servo motor, 102 - sliding seat, 103 - drive plate, 11 - second silicon photovoltaic cell, 12 - glass plate fixing box, 13 - dust accumulation layer, 2 - dust density measurement box, 201 - first dark room, 202 - second dark room, 21 - partition board, 210 - partition opening, 22 - six-degree-of-freedom device, 23 - direct illumination light source, 24 - first silicon photovoltaic cell, 25 - opening and closing structure, 3 - dust density test device, 4 - meteorological data acquisition device, 5 - power generation prediction device, 51 - model training unit, 52 - information receiving unit, 53 - power generation prediction unit. DETAILED DESCRIPTION OF THE INVENTION

[0020] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments, so that those skilled in the art can implement it according to the description in the specification.

[0021] It should be understood that terms such as "having", "comprising", and "including" as used herein do not exclude the presence or addition of one or more other elements or combinations thereof.

[0022] It should be noted that the experimental methods described in the following embodiments are all conventional methods unless otherwise specified, and the reagents and materials can be obtained from commercial channels unless otherwise specified; in the description of the present invention, it should be noted that unless otherwise clearly defined and limited, the terms "installed", "connected", and "set" should be understood in a broad sense. For example, they can be fixedly connected and set, or detachably connected and set, or integrally connected and set. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. The orientation or positional relationship indicated by terms such as "lateral", "longitudinal", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the accompanying drawings, and is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation to the present invention.

[0023] such as Figures 1 - 6As shown in the figure, the present invention provides a photovoltaic power generation performance prediction device, which is arranged in a photovoltaic power generation field and includes: a meteorological acquisition device 4 that records various meteorological data in the photovoltaic power generation field in real time; a dust accumulation degree acquisition module, which includes a dust accumulation degree measurement box 2. The dust accumulation degree measurement box 2 is a closed box body. A partition plate 21 is arranged inside the dust accumulation degree measurement box 2. The inclination angle of the partition plate 21 is equal to the support angle of the photovoltaic panels in the photovoltaic power generation field. A partition hole 210 is opened in the middle of the partition plate 21. The partition plate 21 divides the inside of the dust accumulation degree measurement box 2 into a first dark room 201 and a second dark room 202. A direct illumination light source 23 is installed in the first dark room 201. One side wall of the dust accumulation degree measurement box 2 in the second dark room 202 is an opening and closing structure 25; a same-condition glass plate 1, the inclination angle and orientation of the same-condition glass plate 1 are the same as those of other photovoltaic panels in the photovoltaic power generation field; a first silicon photovoltaic cell 24, which is fixed on one side of the partition plate 21 in the first dark room 201, and the first silicon photovoltaic cell 24 is located above the partition hole 210; a second silicon photovoltaic cell 11, which is fixed on the side of the same-condition glass plate 1 facing the ground; wherein, when the same-condition glass plate 1 is close to the partition plate in the second dark room 202, the second silicon photovoltaic cell 11 is aligned with the center of the partition hole 210, and the illumination range of the direct illumination light source 23 can cover both the first silicon photovoltaic cell 24 and the second silicon photovoltaic cell 11 at the same time.

[0024] In this technical solution, the meteorological acquisition device 4 is approximately equivalent to the combination of an anemometer, an atmospheric sensor, a dust sensor, and a solar radiation sensor in an existing photovoltaic power generation field. It collects data in real time and sends the data to an existing or specially made photovoltaic prediction system. The photovoltaic prediction system uses neural network calculation software, takes the data provided by the meteorological acquisition device 4 as a sample set, and trains a prediction model for short-term solar radiation. When making a prediction, the solar radiation intensity in the prediction period can be inferred only based on the period data provided in the meteorological report. The same-condition glass plate 1 has the same size, orientation, and stacking angle as the photovoltaic panels used. The same-condition glass plate 1 accumulates dust, gets rained on, and is cleaned together with other photovoltaic panels in the field. Each time a short-term photovoltaic performance prediction is carried out, the same-condition glass plate 1 is pushed into the dust accumulation degree measurement box 2 manually or mechanically to complete the detection of the dust accumulation density. Then, the efficiency of the photovoltaic panel can be corrected according to the dust accumulation density, and the power generation amount or power of the photovoltaic power generation is calculated using the corrected efficiency of the photovoltaic panel and the predicted solar radiation intensity. The dust accumulation degree measurement box 2 in this technical solution can be added to an existing photovoltaic power generation field, and the collected dust accumulation density is used for correcting the efficiency of the photovoltaic panel instead of being used as data in the sample set, solving the problem of data loss when the dust accumulation density is used as a neural network model due to the lack of continuous data on the dust accumulation density before.

[0025] In this technical solution, the direct illumination light source 23 is an LED surface light source or a xenon lamp source that is easily obtainable in the market. The first silicon photocell 24 and the second silicon photocell 11 are finished components that are easily obtainable in the market. Optionally, the first silicon photocell 24 and the second silicon photocell 11 can also be photoelectric induction sensors. The principle is that under the same illumination conditions, the first silicon photocell 24 receives the light that has not been scattered by the dust accumulation layer 13, while the second silicon photocell 11 receives the light that has been scattered by the dust accumulation layer 13. As the dust accumulation degree continuously increases, the voltage ratio generated by the two silicon photocells increases or decreases accordingly. Therefore, the dust accumulation density on the same-condition glass plate 1 at this time can be compared by the voltage ratio of the two silicon photocells. Before the dust accumulation degree measurement box 2 and the same-condition glass plate 1 are installed in the photovoltaic power station, since the dust accumulation components and particle sizes of each photovoltaic power station are different, and the performance of the silicon photocells also varies slightly, it is necessary to calibrate the first silicon photocell 24 and the second silicon photocell 11. The dust samples obtained from previous sampling are evenly spread on the surface of the same-condition glass plate 1 from low to high. At this time, the arrangement of the first silicon photocell 24 and the second silicon photocell 11 is the same as above. Through multiple groups of illumination experiments, the voltage ratio values of the two silicon photocells at each dust accumulation density are obtained, and a fitting curve of dust accumulation degree - voltage difference is fitted using a computer program. In the subsequent measurement process, this dust accumulation degree - voltage difference fitting curve is used as the basis for calculating the dust accumulation degree. The obtained dust accumulation degree data is used as a calculation parameter for calculating the efficiency of the dust-bearing photovoltaic panel.

[0026] In this technical solution, the size of the same-condition glass plate 1 is basically equivalent to that of a complete photovoltaic panel. Compared with the current conventional same-condition dust accumulation equipment, it can better simulate the dust accumulation effect. By using the silicon photocell to measure the dust accumulation density, the problem of untimely collection of dust accumulation density data is avoided. The data can be timely fed back and collected and can be involved in the prediction of photovoltaic performance. The collected dust accumulation density data can be used not only as calculation data for the efficiency of the photovoltaic panel, but also, after a certain period of collection and when the dust accumulation density data obtains sufficient data density, as meteorological parameters for the training sample set of the neural network model directly predicting the photovoltaic power generation and power generation.

[0027] In another technical solution, a six-degree-of-freedom device 22 is installed in the first darkroom, and the direct light source 23 is installed on the movable part of the six-degree-of-freedom device 22. There are at least five partition openings 210 on the partition plate 21, and each partition opening 210 corresponds to a first silicon photocell 24 and a second silicon photocell 11. In this technical solution, the same glass plate 1 under the same conditions is tested by a five-point sampling and averaging method. At this time, the position and angle of the light source need to be adjusted, and the direct light source 23 needs to be close to the first silicon photocell 24 and the second silicon photocell 11 to be illuminated. The six-degree-of-freedom device 22 is a manipulator or other device that can control the direct light source 23 to move in the X, Y, and Z axes and rotate in the XY, XZ, and YZ directions. Optionally, the six-degree-of-freedom device 22 is a Stewart structure composed of six actuators and universal joints at both ends. The six-degree-of-freedom device 22 controls the displacement and angle change of the direct light source 23.

[0028] In another technical solution, a linear drive device 10 is provided under the dust accumulation measuring box 2 and the same condition glass plate 1, and the linear drive device 10 includes a servo motor 101 and a sliding seat 102 connected thereto by transmission, and the same condition glass plate 1 is located outside the opening and closing structure 25, wherein the servo motor 101 enables the sliding seat 102 to drive the same condition glass plate 1 to move from the outside to the second darkroom 202 until it is close to the partition plate 21 through a matching screw rod. In this technical solution, in order to save manpower, a linear drive platform can be used to complete the transfer of the same condition glass plate 1, the servo motor 101 is connected to a rotating screw rod, and the sliding seat 102 is driven on the rotating screw rod and several guide structures. When the servo motor 101 drives the rotating screw rod to rotate, the sliding seat 102 drives the same condition glass plate 1 to complete linear motion. In order to avoid the linear drive device 10 being affected by wind and sand, such as Figure 1 As shown, the driving device base 100 adopts a semi-sealed structure with plates wrapped around all sides, and the top plate is provided with a through groove so that the sliding seat 102 can pass through the vertical plate structure fixed upward, and the sliding seat 102 is connected to the driving plate 103 on the upper side of the driving device base 100 through the vertical plate structure. Under the same conditions, a glass plate fixing box 12 can be set under the glass plate 1 to be installed on the driving plate 103.

[0029] In another technical solution, the inner wall and partition plate 21 of the dust accumulation measuring box 2 are painted with black light-absorbing paint. In this technical solution, in order to avoid scattering of the test light of the direct light source 23, the inner wall and partition plate 21 of the dust accumulation measuring box 2 are painted with black paint, and the inside of the glass plate fixing box 12 is also painted with black paint.

[0030] In another technical solution, the meteorological data collection device 4 includes a hygrometer, a barometer, a dust sensor, a wind speed sensor, and a thermometer. The meteorological data collected by the meteorological data collection device 4 includes parameters such as relative humidity, wind speed, PM2.5 concentration, PM10 concentration, temperature, air pressure, and solar radiation value.

[0031] In another technical solution, the photovoltaic power generation performance prediction device further includes: a dust accumulation degree test module 3, which is electrically connected to the first silicon photovoltaic cell 24 and the second silicon photovoltaic cell 11 and records the voltage ratio of the two under the same-source light illumination, and calculates the corresponding dust accumulation density on the same-condition glass plate 1 according to the voltage ratio; a short-term power generation prediction module 5, which includes a model training unit 51, an information receiving module 52, and a power generation prediction unit 53. The short-term power generation prediction module 5 is communicatively connected to the meteorological data collection device 4 and the dust accumulation degree test module 3, and can receive external meteorological forecast data. The model training unit 51 can store the data collected by the meteorological data collection device 4 periodically, and uses the meteorological data collected by the meteorological data collection device 4 as a training set through a neural network learning calculation software to obtain a solar radiation amount prediction model. The information receiving module 52 is used to obtain the meteorological forecast data for the prediction period and the current dust accumulation density data obtained by the dust accumulation degree test module 3. The short-term power generation prediction module 5 calculates the power generation for the prediction period according to the solar radiation amount prediction model, the meteorological data for the prediction period, and the dust accumulation density data in sequence.

[0032] In another technical solution, the implementation method of the photovoltaic power generation performance prediction device includes the following steps: S1. Calibrate the photovoltaic power generation performance prediction device: In the laboratory, manually dust the upper side of the same-condition glass plate 1 equipped with the second silicon photovoltaic cell 11 evenly from scratch. After dusting, attach the same-condition glass plate 1 to the lower side of the partition plate 21 equipped with the first silicon photovoltaic cell 24, and use a direct light source 23 to irradiate the first silicon photovoltaic cell 24 and the second silicon photovoltaic cell 11 simultaneously from the upper side of the partition plate 21, record the output voltage ratio of the two, and establish a fitting curve of the dust accumulation density and the voltage ratio in this device; specifically, in this technical solution, the dust used for manual dusting can be taken from the dust accumulated on the photovoltaic panels collected during the previous cleaning operations of this power plant. Before each test, the surface of the same-condition glass plate 1 needs to be cleaned, and manual uniform dusting is carried out step by step at a density of 1 g / ㎡ or less, and the voltage ratios generated by the first silicon photovoltaic cell 24 and the second silicon photovoltaic cell 11 at different dust accumulation densities are collected, and a dust accumulation density - voltage ratio fitting curve is drawn for all the data using a computer program.

[0033] S2. Place the photovoltaic power generation performance prediction device: Place the photovoltaic power generation performance prediction device in the photovoltaic power generation field, and clean the same-condition glass plate 1 and the photovoltaic panels synchronously.

[0034] S3. Establish a solar radiation prediction model: Use the historical meteorological data collected by the meteorological acquisition device 4 as the sample set data to construct a solar radiation prediction model. The solar radiation prediction model adopts a BP neural network including an input layer, a hidden layer, and an output layer; Specifically, in this technical solution, the solar radiation prediction model is built based on the BP neural network model. The BP neural network is a multi-layer feedforward network, and the three-layer neural network with a hidden layer is relatively commonly used. Research shows that any three-layer BP neural network can approximate any non-linear continuous function and can achieve any accuracy; The input layer of the BP neural network model includes multiple input neuron nodes. There are multiple hidden neuron nodes set on the hidden layer. Each input neuron node is respectively connected to each hidden neuron node; the hidden neuron nodes are all connected to the output layer. The relative humidity, wind speed, PM2.5 concentration, PM10 concentration, air temperature, air pressure, etc. are used as the input values of the input neuron nodes, and the output result is compared with the corresponding solar radiation amount. The BP neural network model is optimized according to the PSO algorithm to obtain the solar radiation prediction model; The six parameters of relative humidity, wind speed, PM2.5 concentration, PM10 concentration, air temperature, and air pressure can be used as influencing factors. The total number of parameter types is used as the number of neuron nodes m, and the solar radiation amount is used as the only output value, that is, the number of nodes in the output layer c = 1, and the number of nodes in the hidden layer c1 is , where a is a random constant between 1 and 10; Perform normalization processing on the sample set data, and its mathematical expression is: Among them, represents the sample data of the influencing factor, , are respectively the minimum and maximum values in the sample data, is the dimensionless processed influencing factor data; Initialize the mapping relationship between the population particles and the weights and thresholds of the BP neural network, including particle dimension, initial velocity, population size, learning factor, and inertia weight; Input the normalized input variables and output variables into the BP neural network model, calculate the fitness function value of the particles, and obtain the historical optimal fitness and global fitness of the particles. The fitness function value of the particles is the mean square error of the calculation result, and its function expression is: Among them, represents the predicted value of the i-th sample, is the true value of the i-th sample, and n is the total number of calculation results of the neural network; perform iterative calculation on the particle fitness, and update the historical optimal fitness and the global fitness according to the preset update conditions until the preset iteration end condition is met. Subsequently, update the weights and thresholds of the preset BP neural network model to obtain an optimized BP neural network model for calculating the current solar radiation amount.

[0035] S4. Calculate the predicted value of solar radiation: Obtain the predicted value of meteorological data for the time period to be predicted, and input the elements of the predicted value of meteorological data into the solar radiation prediction model to obtain the predicted value of ultra-short-term solar radiation. Specifically, the meteorological forecast work is carried out by the meteorological department in the area where the photovoltaic power generation site is located, and the forecast results are sent to the photovoltaic power generation site in a timely manner. The content of the meteorological forecast includes six elements of relative humidity, wind speed, PM2.5 concentration, PM10 concentration, temperature, and air pressure for the prediction period. Input them into the solar radiation prediction model, and the predicted value of solar radiation for the corresponding period can be obtained through calculation.

[0036] S5. Calculate the initial dust deposition density at the time period: Drive the same-condition glass plate 1 into the dust deposition measurement box 2, obtain the voltage ratio of the first silicon photocell 24 and the second silicon photocell 11 at this time after being irradiated by the direct light source 23, and calculate the dust deposition density at this time.

[0037] S6. Calculate the predicted value of power generation: Based on the obtained dust accumulation density, calculate the conversion efficiency reduction of the dust-accumulated photovoltaic module, and calculate the predicted value of power generation. The calculation formula is as follows: where: P - predicted value of power generation, kWh; S - total area of the battery panels in the power generation field, ㎡; H - predicted value of solar radiation, kWh / m²; E d — efficiency of the dust-accumulated photovoltaic module.

[0038] Specifically, the efficiency of the dust-accumulated photovoltaic module can be obtained according to the corresponding relationship between the dust deposition density and the module efficiency in industry standards or relevant industry achievements.

[0039] In another technical solution, the step S5 includes the following steps: A1: Open the opening and closing structure 25. Wait for the driving device 10 to pull the same-condition glass plate 1 into the second dark room 202 and align it with the partition plate 21, and then close the opening and closing structure 25; A2: The six-degree-of-freedom device 22 drives the preheated direct light source 23 to align it with a partition opening 210, and the light irradiation covers the first silicon photocell 24 and the second silicon photocell 11 corresponding to the partition opening 210; A3: After the voltages of the first silicon photovoltaic cell 24 and the second silicon photovoltaic cell 11 are stabilized, record the output voltage ratio at this time. According to the fitting curve obtained in step S1, obtain the dust deposition density corresponding to this voltage ratio.

[0040] In another technical solution, the efficiency E of the dust deposition photovoltaic module in step S6 d is calculated by the formula: where: E c —the efficiency of the clean photovoltaic module; p m —the dust deposition density, g / m 2 .

[0041] In this technical solution, considering that the influence of dust deposition on photovoltaic panels of different materials is different, separate calculations are required. In this solution, according to the photovoltaic panel materials, it is divided into three categories: monocrystalline silicon, polycrystalline silicon, and amorphous silicon. The calculation results are more accurate and are also applicable to different photovoltaic power generation fields.

[0042] It should be noted that although the above steps are described in a specific order, it does not mean that the steps must be executed in the above specific order. In fact, some of these steps can be executed concurrently or even the order can be changed, as long as the required functions can be achieved. The number of devices and the processing scale described here are used to simplify the description of the present invention, and the application, modification, and variation of the present invention are obvious to those skilled in the art.

[0043] Although the embodiments of the present invention have been disclosed above, it is not limited to only the applications listed in the specification and embodiments. It can be fully applied to various fields suitable for the present invention. For those familiar with the art, additional modifications can be easily made. Therefore, without departing from the general concept defined by the claims and the equivalent scope, the present invention is not limited to specific details and the illustrations shown and described here.

Claims

1. A photovoltaic power generation performance prediction device, which is arranged in a photovoltaic power generation field, and is characterized in that, include: A meteorological data collection device (4) for recording various meteorological data in the photovoltaic power plant in real time; A dust accumulation acquisition module, comprising a dust accumulation measurement box (2), the dust accumulation measurement box (2) being a closed box, a partition plate (21) being provided inside the dust accumulation measurement box (2), the inclination angle of the partition plate (21) being equal to the supporting angle of the photovoltaic panels in the photovoltaic power generation field, a partition plate opening (210) being provided in the middle of the partition plate (21), the partition plate (21) dividing the interior of the dust accumulation measurement box (2) into a first dark room (201) and a second dark room (202), a direct illumination light source (23) being installed in the first dark room (201), and a side wall of the dust accumulation measurement box (2) of the second dark room (202) being an opening and closing structure (25); A glass plate (1) with the same conditions, wherein the inclination angle and orientation of the glass plate (1) with the same conditions are equal to the support angle and orientation of other photovoltaic panels in the photovoltaic power generation field; A first silicon photocell (24), which is fixed to one side of the partition plate (21) in the first darkroom (201), and the first silicon photocell (24) is located above the partition plate opening (210); A second silicon photocell (11) is fixed on the side of the glass plate (1) facing the ground; When the same condition glass plate (1) is close to the partition plate in the second darkroom (202), the second silicon photocell (11) is aligned with the center of the partition plate opening (210), and the illumination range of the direct light source (23) can simultaneously cover the first silicon photocell (24) and the second silicon photocell (11).

2. The photovoltaic power generation performance prediction device according to claim 1, characterized in that, A six-degree-of-freedom device (22) is installed in the first darkroom, the direct illumination light source (23) is installed on the movable part of the six-degree-of-freedom device (22), and there are at least five partition openings (210) on the partition plate (21), and each partition opening (210) corresponds to a first silicon photocell (24) and a second silicon photocell (11).

3. The photovoltaic power generation performance prediction device according to claim 1, wherein A linear drive device (10) is provided under the dust accumulation measurement box (2) and the same condition glass plate (1), the linear drive device (10) comprising a servo motor (101) and a sliding seat (102) connected thereto in transmission, the same condition glass plate (1) being located outside the opening and closing structure (25), wherein the servo motor (101) enables the sliding seat (102) to drive the same condition glass plate (1) to move horizontally from the outside to the inside of the second darkroom (202) until it is close to the partition plate (21) through a matching screw rod.

4. The photovoltaic power generation performance prediction device according to claim 1, wherein The inner wall and the partition plate (21) of the dust accumulation measurement box (2) are painted with black light-absorbing paint.

5. The photovoltaic power generation performance prediction device according to claim 1, characterized in that The meteorological collection device (4) comprises a hygrometer, a barometer, a dust sensor, a wind speed sensor, a thermometer and a solar radiation sensor. The meteorological data collected by the meteorological collection device (4) include parameters such as relative humidity, wind speed, PM2.5 concentration, PM10 concentration, temperature, air pressure, solar radiation value, etc.

6. The photovoltaic power generation performance prediction device according to claim 1, wherein, Also includes: Dust accumulation degree test module (3), which is electrically connected to the first silicon photovoltaic cell (24) and the second silicon photovoltaic cell (11), records the voltage ratio of the two under the same-source light illumination, and calculates the dust accumulation density on the corresponding same-condition glass plate (1) according to the voltage ratio. Short-term power generation prediction module (5), which includes a model training unit (51), an information receiving module (52) and a power generation prediction unit (53). The short-term power generation prediction module (5) is communicatively connected to the meteorological acquisition device (4) and the dust accumulation degree test module (3), and can receive external weather forecast data. The model training unit (51) can store the data periodically collected by the meteorological acquisition device (4), and uses the meteorological data collected by the meteorological acquisition device (4) as a training set through neural network learning calculation software to obtain a solar radiation amount prediction model. The information receiving module (52) is used to obtain the weather forecast data for the prediction period and the current dust accumulation density data obtained by the dust accumulation degree test module (3). The short-term power generation prediction module (5) calculates the power generation amount for the prediction period according to the solar radiation amount prediction model, the meteorological data for the prediction period, and the dust accumulation density data in sequence.

7. The implementation method of the photovoltaic power generation performance prediction device according to any one of claims 1-6, characterized in that, Including the following steps: S1. Calibrate the photovoltaic power generation performance prediction device: In the laboratory, dust is evenly distributed manually from scratch on the upper side of the same-condition glass plate (1) equipped with the second silicon photovoltaic cell (11). After dusting, the same-condition glass plate (1) is attached to the lower side of the partition plate (21) equipped with the first silicon photovoltaic cell (24). A direct light source (23) is used to irradiate the first silicon photovoltaic cell (24) and the second silicon photovoltaic cell (11) simultaneously from the upper side of the partition plate (21), and the output voltage ratio of the two is recorded to establish a fitting curve between the dust accumulation density and the voltage ratio in this device. S2. Place the photovoltaic power generation performance prediction device: Place the photovoltaic power generation performance prediction device in the photovoltaic power generation field, and clean the same-condition glass plate (1) synchronously with the photovoltaic panel. S3. Establish a solar radiation amount prediction model: Use the historical meteorological data collected by the meteorological acquisition device (4) as sample set data to construct a solar radiation amount prediction model. The solar radiation amount prediction model uses a BP neural network including an input layer, a hidden layer and an output layer. S4. Calculate the predicted value of solar radiation amount: Obtain the predicted value of meteorological data for the time period to be predicted, and input the elements of the predicted value of meteorological data into the solar radiation amount prediction model to obtain the predicted value of ultra-short-term solar radiation amount. S5. Calculate the dust accumulation density at the initial time of the time period: Drive the same-condition glass plate (1) into the dust accumulation degree measurement box (2). After being irradiated by the direct light source (23), obtain the voltage ratio of the first silicon photovoltaic cell (24) and the second silicon photovoltaic cell (11) at this time, and calculate the dust accumulation density at this time. S6. Calculate the predicted value of power generation: Based on the obtained dust accumulation density, calculate the conversion degradation efficiency of the dust-accumulated photovoltaic module, and calculate the predicted value of power generation. The calculation formula is as follows: Where: P—the predicted value of power generation, kWh; S—the total area of the battery panels in the power generation field, ㎡; H—Estimated value of solar radiation, kWh / m²; E d — Efficiency of dust-accumulated photovoltaic module.

8. The implementation method of the photovoltaic power generation performance prediction device according to claim 7, characterized in that, The step S5 includes the following steps: A1: Open the opening and closing structure (25). After the driving device (10) pulls the same-condition glass plate (1) into the second darkroom (202) and aligns it with the partition plate (21), close the opening and closing structure (25); A2: The six-degree-of-freedom device (22) drives the preheated direct light source (23) to align it with a partition opening (210), and the light irradiation covers the first silicon photocell (24) and the second silicon photocell (11) corresponding to the partition opening (210); A3: After the voltages of the first silicon photocell (24) and the second silicon photocell (11) are stable, record the output voltage ratio at this time, and obtain the dust density corresponding to this voltage ratio according to the fitting curve obtained in step S1.

9. The implementation method of the photovoltaic power generation performance prediction device according to claim 7, characterized in that, The efficiency E of the dust-accumulated photovoltaic module in step S6 d is calculated by the formula: Wherein: E c — cleaning efficiency of the photovoltaic module; p m — Dust accumulation density, g / m 2 .

Images