A method and system for automatically cleaning photovoltaic panels

Abstract

The application provides a photovoltaic panel automatic cleaning method and system, and belongs to the technical field of new energy; the method comprises the following steps: S1, a spraying cleaning unit is configured, and the spraying cleaning unit is used for cleaning and dust removing the surface of the photovoltaic panel; S2, the geographical position and weather condition of the place where the photovoltaic panel is located are acquired, a prediction output model of the daily power generation of the photovoltaic panel is constructed, and the prediction value of the daily power generation of each photovoltaic panel is obtained through the prediction output model of the daily power generation of the photovoltaic panel; S3, the actual daily power generation of each photovoltaic panel is acquired; S4, the difference degree between the prediction value and the actual daily power generation of the daily power generation of each photovoltaic panel in a period of time is compared, and the photovoltaic panel needing cleaning is screened out; and S5, after the spraying cleaning unit performs operation on the photovoltaic panel needing cleaning, whether the difference degree is improved is re-evaluated, if the difference degree is improved, steps S3-S4 are repeated in the next comparison period, and if the difference degree is not improved after continuous cleaning reaches a specified number of times, an alarm prompt information is sent.

Description

Technical Field

[0001] This invention relates to the field of new energy technology, and in particular to an automatic cleaning method and system for photovoltaic panels. Background Technology

[0002] With continuous economic development, people's demand for clean energy is increasing. Photovoltaic panels and power supply components store the electricity generated by photovoltaic power generation, then boost and invert it to supply the grid, meeting the power needs of electrical equipment. In specific outdoor scenarios, such as specialized agriculture in mountainous areas, tracked vehicles are often used in conjunction with photovoltaic power generation systems to automatically power the vehicles, meeting their energy needs for transporting materials at different altitudes without relying on grid power.

[0003] However, the outdoor environment is changeable. With the continuous use of photovoltaic modules, dust, debris or snow often accumulate on the surface, which seriously affects the stability of the photovoltaic module's power generation capacity. Therefore, it is necessary to determine the type and severity of the debris and clean it in a timely manner to maintain the stability of the power generation capacity.

[0004] Therefore, it is essential to provide an automatic cleaning method and system for photovoltaic panels, which regularly cleans the photovoltaic panels in accordance with weather changes to maintain their reliable working condition and power generation capacity. Summary of the Invention

[0005] In view of this, the present invention proposes an automatic photovoltaic panel cleaning method and system that improves power generation stability by predicting the difference between the power generation and the predicted discharge on a corresponding date based on weather conditions, and identifying photovoltaic panels that may need cleaning based on the difference.

[0006] On one hand, the present invention provides an automatic cleaning method for photovoltaic panels, comprising the following steps:

[0007] S1: A spray cleaning unit is configured on the array structure of photovoltaic panels of the same specifications. The spray cleaning unit is used to clean and remove dust from the surface of the photovoltaic panels.

[0008] S2: Obtain the geographical location and weather conditions of the photovoltaic panel's location, construct a prediction output model for the daily power generation of the photovoltaic panel, and obtain the predicted value of the daily power generation of each photovoltaic panel through the prediction output model;

[0009] S3: Obtain the actual daily power generation of each photovoltaic panel;

[0010] S4: Compare the difference between the predicted daily power generation of each photovoltaic panel and the actual daily power generation over a period of time, and screen out the photovoltaic panels that need to be cleaned;

[0011] S5: After the spray cleaning unit has performed the cleaning operation on the photovoltaic panels that need to be cleaned, it repeats step S4 to reassess whether the difference has improved. If it has improved, it repeats steps S3-S4 in the next comparison cycle. If the difference still does not improve after continuous cleaning for a specified number of times, it issues an alarm message.

[0012] Based on the above technical solution, preferably, step S2 involves obtaining the total daily atmospheric top irradiance. H 0. Atmospheric transparency factor f atm Photovoltaic system efficiency η sys and the effective light-receiving area of ​​a single photovoltaic panel A Construct a predictive output model for the daily power generation of photovoltaic panels. E ideal , in kilowatt-hours.

[0013] Preferably, the total daily atmospheric top horizontal radiation H 0 is calculated based on the duration corresponding to the sunrise and sunset angles, solar declination, and daily average atmospheric top normal irradiance; atmospheric transparency factor. f atm It divides the daytime into equal-length periods. t The efficiency of the photovoltaic system is obtained by weighted summation of the normalized weather values ​​and the weather weights for each equal-length period; η sys It is a linear function based on the initial efficiency of the photovoltaic system.

[0014] Preferably, step S3 involves adjusting the photovoltaic panels during the daytime period. T 1, T 2] Within the sampling interval △ t Divide into sections, sampling interval Δ t ≤ Equal Length Period t Instantaneous power within each sampling interval is obtained by measuring with a digital power meter. P t The instantaneous power is accumulated at sampling intervals, and after adding a correction term, a temperature correction coefficient is used to compensate for temperature loss, thus obtaining the actual daily power generation of each photovoltaic panel. E actual .

[0015] More preferably, the correction term is the power of non-photovoltaic panel equipment in the photovoltaic system during daytime hours.

[0016] More preferably, step S4 includes the following:

[0017] S41: Set a first threshold and a second threshold, where the first threshold is less than the second threshold, and compare them over a continuous period of time.k The actual daily power generation of each photovoltaic panel within the unit E actual The sum of the predicted outputs of the daily power generation of each photovoltaic panel E ideal The deviation of the sum PR i sequence values, , , m The total number of photovoltaic panels; 1) If the deviation PR i The sequence values ​​exhibit discontinuous, occasional deviations. These discontinuous, occasional deviations are either less than the first threshold or greater than the second threshold, and the deviations... PR i If more than 95% of the sequence values ​​are within the [first threshold, second threshold] range, proceed to step S44; 2) If the deviation... PR i If at least three consecutive singular values ​​in the sequence are all less than the first threshold, it indicates a serious fault or severe weather, and step S42 is executed; 3) If the deviation PR i If at least three consecutive singular values ​​in the sequence are greater than the second threshold, it indicates that there is a measurement error or environmental factors. Proceed to step S43.

[0018] S42: For situations with serious faults, check the following faults: inaccurate installation tilt angle or azimuth angle of photovoltaic panels; string mismatch between photovoltaic panels; excessive performance degradation of photovoltaic panels due to reaching or exceeding the maximum operating limit; excessive transmission line loss; abnormal reactive power compensation; obstruction by buildings, trees, or adjacent photovoltaic panels; photovoltaic panel backsheet temperature higher than the preset temperature; after correcting the faults, repeat step S3; for situations affected by severe weather, remove the coverings on the photovoltaic surface promptly after the severe weather ends, execute step S5, use the spray cleaning unit to clean all photovoltaic panels, and then repeat step S3;

[0019] S43: In cases where there are measurement errors or environmental factors, check for the following errors: Verify the actual ground irradiance and the total daily irradiance at the top of the atmosphere. H Deviations between 0 and 1; confirm the presence of environmental reflections in the photovoltaic panel installation environment; verify the efficiency of the photovoltaic system. η sys After adjusting the parameters of the prediction output model that affect the daily power generation of photovoltaic panels, repeat steps S2 and S3.

[0020] S44: Eliminate sporadic, intermittent biases to determine the actual daily power generation of missing photovoltaic panels. Eactual Interpolation is performed to fill in the missing data; then the actual daily power generation of each photovoltaic panel is calculated after interpolation. E actual The daily power generation is sorted from largest to smallest to obtain a sequence. The number of sequence values ​​corresponds one-to-one with the number of photovoltaic panels. Using the average of the top 5% of the daily power generation sequences as a reference, a third threshold is set. Photovoltaic panels corresponding to sequence values ​​that deviate from the reference reference by more than the third threshold are marked. This process is repeated for a continuous period of time. k The actual daily power generation of the photovoltaic panels inside E actual Repeat the above interpolation, time-sequence, and labeling process, exceeding the number of times the labeling is performed. k / 3 of the photovoltaic panels are those that require cleaning.

[0021] A further preferred embodiment is that if the difference does not improve after a specified number of consecutive cleaning cycles, an alarm message is issued to alert management personnel that the number of times the marking has been exceeded. k / 3 of the photovoltaic panels were inspected for defects, including the following: back panel temperature, local deformation or hot spots, local shading, damage, loose connection of output terminals, or excessive performance degradation. The photovoltaic panels that do not fall under the defect category are those that need cleaning, while the photovoltaic panels with defects are maintained or replaced.

[0022] Further preferred, the first threshold is the predicted output result of the daily power generation of each photovoltaic panel. E ideal The sum of 93%-95%; the first threshold is the predicted output of the daily power generation of each photovoltaic panel. E ideal The sum is 110%-112%; the third threshold is more than 3% of the reference benchmark.

[0023] More preferably, the spray cleaning unit includes several brush rollers and a water inlet pipe. The rotating shafts of the several brush rollers are located on the same virtual plane, and the rotating shafts of adjacent brush rollers are set at an angle. The water inlet pipe is used to connect to a water source. The spray cleaning unit performs surface cleaning on all photovoltaic panels in the column where the photovoltaic panel to be cleaned is located and all photovoltaic panels in adjacent columns.

[0024] On the other hand, the present invention also provides an automatic photovoltaic panel cleaning system for implementing the above-mentioned automatic photovoltaic panel cleaning method, comprising:

[0025] A spray cleaning unit is configured on the array structure of photovoltaic panels;

[0026] The photovoltaic panel daily power generation prediction output model construction unit is used to construct a photovoltaic panel daily power generation prediction output model based on the geographical location and weather conditions of the photovoltaic panel location, and output the predicted value of the photovoltaic panel daily power generation.

[0027] The photovoltaic panel actual daily power generation acquisition unit is used to acquire the actual daily power generation of each photovoltaic panel;

[0028] The difference comparison unit is connected to the photovoltaic panel daily power generation prediction output model construction unit, the photovoltaic panel actual daily power generation acquisition unit, and the spray cleaning unit. It is used to acquire and compare the degree of difference between the predicted value and the actual daily power generation of each photovoltaic panel over a period of time, screen out the photovoltaic panels that need to be cleaned, and drive the spray cleaning unit to perform cleaning on the photovoltaic panels that need to be cleaned. Then, it re-evaluates the degree of difference between the predicted value and the actual daily power generation of each photovoltaic panel. If the degree of difference decreases, it indicates that the cleaning of the photovoltaic panels is effective. If the difference does not improve after a specified number of consecutive cleanings, an alarm message is issued.

[0029] The automatic cleaning method and system for photovoltaic panels provided by this invention have the following advantages compared with the prior art:

[0030] (1) This invention dynamically identifies photovoltaic panels that need cleaning by comparing the difference between the predicted power generation and the actual power generation of the photovoltaic panels, avoiding the waste of resources caused by blind cleaning and reducing the frequency and scope of spraying operations; after cleaning, a closed-loop verification is performed to see if the difference is improved. If it is not improved, an alarm is triggered to ensure the effectiveness of cleaning and eliminate performance degradation caused by non-dust factors, thereby improving the stability of photovoltaic output.

[0031] (2) Introducing factors such as daily atmospheric top irradiance and atmospheric transparency factor, combined with the efficiency and light-gathering area of ​​the photovoltaic system, provides accurate estimation results for subsequent difference analysis; adopting dynamic weather weighting method to divide the daytime into different equal-length periods and assigning different weights can more accurately reflect the impact of instantaneous weather on irradiance and improve the accuracy of power generation prediction.

[0032] (3) High-frequency sampling is used to obtain the instantaneous power of each photovoltaic panel and the power consumption of auxiliary equipment is accumulated to eliminate the influence of environmental factors and auxiliary equipment power consumption, so as to ensure the authenticity of the actual power generation data and improve the reliability of subsequent difference comparison.

[0033] (4) Comparison of differences: First, the difference between the predicted power generation and the actual power generation of the entire photovoltaic panel is compared. Then, for the data within the threshold range, the actual power generation of each photovoltaic panel is sorted and compared again. Only non-faulty photovoltaic panels are cleaned, combining regular cleaning with equipment health management. Attached Figure Description

[0034] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0035] Figure 1 This is an overall flowchart of an automatic cleaning method and system for photovoltaic panels according to the present invention;

[0036] Figure 2 This is a schematic diagram of a spray cleaning unit of an automatic photovoltaic panel cleaning method and system according to the present invention.

[0037] Figure 3 This is a schematic diagram of the spray cleaning unit moving along the photovoltaic panel in an automatic cleaning method and system for photovoltaic panels according to the present invention. Detailed Implementation

[0038] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0039] With continuous use, photovoltaic modules often accumulate dust, debris, or snow on their surfaces, severely affecting the stability of their power generation capacity. Therefore, it is necessary to determine the type and severity of the debris and clean it promptly to maintain stable power generation. In view of this, if... Figure 1 As shown, the present invention provides an automatic cleaning method for photovoltaic panels, comprising the following steps:

[0040] S1: A spray cleaning unit is configured on the array structure of photovoltaic panels of the same specifications. The spray cleaning unit is used to clean and remove dust from the surface of the photovoltaic panels.

[0041] like Figure 2As shown, the spray cleaning unit includes several brush rollers and a water inlet pipe. The rotating shafts of the brush rollers are located on the same virtual plane, and the rotating shafts of adjacent brush rollers are set at an angle. The water inlet pipe is used to connect to a water source. The spray cleaning unit cleans the surface of all photovoltaic panels in the column containing the photovoltaic panel to be cleaned and all adjacent columns. As an embodiment of the present invention, the brush rollers are arranged in a polygonal shape. A circular brush disk and a water spray pipe are arranged at the center of the virtual polygon. The circular brush disk is rotated relative to the rotating shaft. The brush rollers rotate on their own axis and also revolve around the circular brush disk. There are gaps between adjacent brush rollers and between the brush rollers and the circular brush disk. On one hand, the virtual polygon containing the brush rollers moves linearly along the photovoltaic surface; on the other hand, the brush rollers and the brush disk also rotate for cleaning. The spray cleaning unit can be configured at the end of a robot arm or the movable end of an XY-axis linear movement mechanism. The XY-axis linear movement mechanism drives the spray cleaning unit to move within the plane containing the photovoltaic panel. When cleaning is not required, the spray cleaning unit is moved to an edge position that will not obstruct the photovoltaic panel.

[0042] S2: Obtain the geographical location and weather conditions of the photovoltaic panel's location, construct a predictive output model for the daily power generation of the photovoltaic panel, and obtain the predicted value of the daily power generation of each photovoltaic panel through the predictive output model for the daily power generation of the photovoltaic panel.

[0043] Step S2 involves obtaining the total daily atmospheric top glacier irradiance. H 0. Atmospheric transparency factor f atm Photovoltaic system efficiency η sys and the effective light-receiving area of ​​a single photovoltaic panel A Construct a predictive output model for the daily power generation of photovoltaic panels. E ideal Unit: kilowatt-hour The total daily atmospheric top horizontal radiation H 0 is based on the sunrise and sunset angles. Corresponding duration, solar declination Calculated from the daily average normal irradiance at the top of the atmosphere: Sunrise and sunset angle The unit is radians, and the total hour angle from sunrise to sunset is twice the radian. Divide by 2π and multiply by 24 to get the duration of sunshine; The latitude of the location of the photovoltaic panel. E 0 is the Earth-Sun distance correction factor; S c The solar constant is used. The daily average atmospheric top normal irradiance is the product of the Earth-Sun distance correction factor and the solar constant. It represents the irradiance reaching the atmospheric top in a single day, assuming the Sun is always in the zenith direction. It should be noted that solar declination... This is the angle between the sun's rays and the Earth's equatorial plane, which varies from day to day but is considered the same each day. Dividing the denominator by 3600 is to output the predicted daily power generation of the photovoltaic panels into the model. E ideal The output unit is converted to kilowatt-hours for later comparison.

[0044] Atmospheric transparency factor f atm It divides the daytime into equal-length periods. t The result is obtained by weighted summation of the normalized weather values ​​and the weather weights for each equal-length time period: ,in To ensure consistent calculations and comparisons, the summer solstice is used as the baseline, with 10-15 minutes serving as an equal-length interval. t The sunrise and sunset periods are divided into consecutive segments.

[0045] The weather normalization value for each equal-length period is S ( t The normalized value for sunny weather is 0.7-0.85, with higher values ​​for longer periods of consecutive sunny days; the normalized value for cloudy weather is 0.2-0.7, depending on cloud thickness and the degree of sky obstruction around the photovoltaic panel area; the normalized value for rainy weather is 0.02-0.25, with lower values ​​for greater precipitation and longer duration; and the normalized value for snowy weather is 0.1-0.2, with lower values ​​for heavier snowfall.

[0046] For sunny days, the number of consecutive sunny days d It is an integer, and d ≥1 indicates the normalized value of the weather over an equal-length period on a sunny day. , The parameter representing the growth rate on sunny days is a positive real number.

[0047] For cloudy weather, the normalized value of cloud thickness is... C A value of 1 indicates that the clouds are thick enough to completely block sunlight, while a value of 0 indicates that there are thin clouds or no clouds. The degree of sky obstruction around the area where the photovoltaic panels are located is... B Taking the photovoltaic panel as the center radius r The proportion of the sky obscured by clouds is the normalized value of the equal-length period of cloudy weather. , ;

[0048] For rainy days, let the normalized value of precipitation over an equal period be... R Longer periods t The normalized value range for precipitation within the area is [0, 1] for 0-10 mm, and 1 for amounts exceeding 10 mm; the normalized value for precipitation duration is...D The duration of precipitation is equal to the length of the same period. t If the normalized value interval corresponding to the proportion is [0, 1], then the normalized value of the equal-length period of rainy days... , p As the attenuation factor, and R + D The sum is inversely proportional;

[0049] For snowy days, let the normalized value of snowfall over an equal period be... S w Longer periods t The value is 1 when the snowfall is less than 0.02 mm and 1 when it is greater than or equal to 0.02 mm; then the normalized value for the equal-length period of snowy days is... .

[0050] Weather weights for equal-length periods are Calculate according to the following formula: ,in For solar altitude angle, molecule The sine of the solar altitude angle for the current equal-length period represents the radiation intensity received per unit area of ​​the Earth's surface during the equal-length period. The denominator represents the normalized baseline of the total solar radiation intensity throughout the day. The significance of weather weighting is to dynamically assign different weights to different equal-length periods. For example, around noon, when the solar altitude angle is close to 90°, the weather weight is greater than at sunrise and sunset. This dynamic adjustment of weather weights for equal-length periods can automatically adapt to seasonal and latitudinal changes. For instance, in winter, when days are shorter and nights are longer, the denominator becomes smaller, and the weather weight in winter is greater than in summer under the same weather conditions. Similarly, lower radiation levels in high-latitude regions result in a smaller denominator, and the weather weight in high-latitude regions is greater than in low-latitude regions under the same weather conditions.

[0051] Photovoltaic system efficiency η sys It is a linear function based on the initial efficiency of the photovoltaic system: , The initial efficiency of the photovoltaic system. Y As the attenuation factor, , T 0 represents the current cumulative number of days the photovoltaic panels have been used. This is a rounding operation. It means that the photovoltaic system has a relatively rapid degradation rate in the first three years, and the degradation rate is relatively gradual after three years.

[0052] S3: Obtain the actual daily power generation of each photovoltaic panel.

[0053] Step S3 involves adjusting the photovoltaic panels during the daytime period. T 1, T 2] Within the sampling interval △ tDivide into sections, sampling interval Δ t ≤ Equal Length Period t Instantaneous power within each sampling interval is obtained by measuring with a digital power meter. P t The instantaneous power is accumulated at sampling intervals, and after adding a correction term, a temperature correction coefficient is used to compensate for temperature loss, thus obtaining the actual daily power generation of each photovoltaic panel. E actual : ,definition T 1 and T 2 represents sunrise and sunset times, respectively, with the time span between sunrise and sunset times determined by the sampling interval Δ. t Divide into sections. Correction terms. E aux This refers to the power demand of non-PV panel equipment in a photovoltaic system during daytime hours, such as inverters and fans that directly utilize photovoltaic power generation. Temperature correction factor. It is a negative real number, and its value does not exceed -0.005 / ℃; The average temperature for the current day. The operating temperature of the photovoltaic panel has a certain impact on the output capacity; for every degree Celsius increase in temperature, the output capacity decreases by 0.35%-0.5%. Through the above calculations, the actual daily power generation can be obtained relatively accurately. The digital power meter can be a precise electricity meter or power metering device.

[0054] S4: Compare the difference between the predicted and actual daily power generation of each photovoltaic panel over a period of time, and select the photovoltaic panels that need to be cleaned.

[0055] Step S4 includes the following:

[0056] S41: Set a first threshold and a second threshold, where the first threshold is less than the second threshold, and compare them over a continuous period of time. k The actual daily power generation of each photovoltaic panel within the unit E actual The sum of the predicted outputs of the daily power generation of each photovoltaic panel E ideal The deviation of the sum PR i sequence values, , , m The total number of photovoltaic panels; 1) If the deviation PR i The sequence values ​​exhibit discontinuous, occasional deviations. These discontinuous, occasional deviations are either less than the first threshold or greater than the second threshold, and the deviations... PR i If more than 95% of the sequence values ​​are within the [first threshold, second threshold] range, proceed to step S44; 2) If the deviation... PR i If at least three consecutive singular values ​​in the sequence are all less than the first threshold, it indicates a serious fault or severe weather, and step S42 is executed; 3) If the deviation PR i If at least three consecutive singular values ​​in the sequence are greater than the second threshold, it indicates that there is a measurement error or environmental factors. Proceed to step S43.

[0057] S42: For situations with serious faults, check the following faults: inaccurate installation tilt angle or azimuth angle of photovoltaic panels; string mismatch between photovoltaic panels; excessive performance degradation of photovoltaic panels due to reaching or exceeding the maximum operating limit; excessive transmission line loss; abnormal reactive power compensation; obstruction by buildings, trees, or adjacent photovoltaic panels; photovoltaic panel backsheet temperature higher than the preset temperature; after correcting the faults, repeat step S3; for situations affected by severe weather, remove the coverings on the photovoltaic surface promptly after the severe weather ends, execute step S5, use the spray cleaning unit to clean all photovoltaic panels, and then repeat step S3;

[0058] If the actual daily power generation is generally low, it can be verified from the following aspects: replacing the power meter measuring instrument, checking the degree of degradation corresponding to the service life of the photovoltaic panels, changes in the posture and installation direction of the photovoltaic panels, abnormal heating of the lines leading to power loss, poor soldering of the lines or unreasonable number of strings, and the presence of large areas of shading, etc.

[0059] S43: In cases where there are measurement errors or environmental factors, check the following errors: 1. Verify the actual ground irradiance and the total daily irradiance at the top of the atmosphere. H 1. Deviation between 0 and 1; 2. Confirm whether there is environmental reflection in the photovoltaic panel installation environment; 3. Verify the efficiency of the photovoltaic system. η sys After adjusting the parameters of the prediction output model that affect the daily power generation of photovoltaic panels, repeat steps S2 and S3.

[0060] To address potential shortcomings in the theoretical model, adjustments can be made from the following aspects: temperature correction coefficient. Photovoltaic system efficiency needs to be measured and calibrated. η sys It needs to be recalibrated; whether there is significant environmental reflection due to seasonal factors, such as the impact of snowfall in high-latitude regions; and whether the actual ground irradiance at the current latitude matches the total daily irradiance at the top of the atmosphere. H There is a significant deviation. After adjustment, the predicted and actual daily power generation of each photovoltaic panel are recalculated, and the comparison in S41 is executed again.

[0061] S44: Eliminate sporadic, intermittent biases to determine the actual daily power generation of missing photovoltaic panels. E actual Interpolation is performed to fill in the missing data; then the actual daily power generation of each photovoltaic panel is calculated after interpolation. E actual The daily power generation is sorted from largest to smallest to obtain a sequence. The number of sequence values ​​corresponds one-to-one with the number of photovoltaic panels. Using the average of the top 5% of the daily power generation sequences as a reference, a third threshold is set. Photovoltaic panels corresponding to sequence values ​​that deviate from the reference reference by more than the third threshold are marked. This process is repeated for a continuous period of time. k The actual daily power generation of the photovoltaic panels inside E actual Repeat the above interpolation, time-sequence, and marking process until the cumulative number of markings exceeds [number missing]. k / 3 t The photovoltaic panels are those that require cleaning.

[0062] like Figure 3 As shown, the spray cleaning unit cleans the surface of all photovoltaic panels in the column containing the photovoltaic panel that needs cleaning, as well as all photovoltaic panels in adjacent columns. Figure 3 The rectangular area in the diagram represents the photovoltaic panel area, and the area marked with "L" represents the photovoltaic panel that needs to be cleaned. Figure 3 The diagram shows the area for cleaning the surface of all photovoltaic panels in the column and its adjacent columns. The cleaning can proceed from one side of the area to be cleaned to the other, or from the two outer columns to the middle column, or from the middle column to the two outer columns. Further details will not be provided here.

[0063] Actual daily power generation of missing photovoltaic panels E actual Interpolation can be performed by using the average of the actual daily power generation of the current photovoltaic panel on adjacent dates that are missing actual daily power generation, or by using the previous value of the current photovoltaic panel as the interpolation, or by sampling the average of the actual daily power generation of multiple normally functioning photovoltaic panels adjacent to the current photovoltaic panel and using it as the interpolation of the actual daily power generation of the missing photovoltaic panel.

[0064] In this embodiment, the first threshold is the predicted output result of the daily power generation of each photovoltaic panel. E ideal The sum of 93%-95%; the second threshold is the predicted output of the daily power generation of each photovoltaic panel. E ideal The sum is 110%-112%; the third threshold is more than 3% of the reference benchmark.

[0065] S5: After the spray cleaning unit has performed the cleaning operation on the photovoltaic panels that need to be cleaned, it repeats step S4 to reassess whether the difference has improved. If it has improved, it repeats steps S3-S4 in the next comparison cycle. If the difference still does not improve after continuous cleaning for a specified number of times, it issues an alarm message.

[0066] If the difference does not improve after a specified number of consecutive cleaning cycles, an alarm message will be issued. After generating the alarm message, it will be sent to the management terminal device, such as a mobile handheld terminal like a mobile phone or PAD, prompting the management personnel to check for defects in photovoltaic panels that have been marked more than k / 3t. The defects to be checked include: photovoltaic panel backsheet temperature, photovoltaic panel local deformation or hot spots, local shading, photovoltaic panel damage or loose connection of output terminals (loose, oxidation, poor contact, etc.), or excessive performance degradation of photovoltaic panels. Photovoltaic panels that do not fall under the defect category are marked as photovoltaic panels that need to be cleaned, while photovoltaic panels with defects are maintained or replaced.

[0067] Reaching the specified number of continuous cleaning attempts refers to reaching the maximum number of cleaning retries, typically an odd number of times (3 or more). For maintenance or replacement of defective photovoltaic panels, please refer to step S42.

[0068] Of course, in addition to the above situations, you can also set an automatic cleaning time cycle, or a strategy to clean immediately after severe weather.

[0069] On the other hand, the present invention also provides an automatic photovoltaic panel cleaning system for implementing the above-mentioned automatic photovoltaic panel cleaning method, comprising:

[0070] A spray cleaning unit is configured on the array structure of photovoltaic panels;

[0071] The photovoltaic panel daily power generation prediction output model construction unit is used to construct a photovoltaic panel daily power generation prediction output model based on the geographical location and weather conditions of the photovoltaic panel location, and output the predicted value of the photovoltaic panel daily power generation.

[0072] The photovoltaic panel actual daily power generation acquisition unit is used to acquire the actual daily power generation of each photovoltaic panel;

[0073] The difference comparison unit is connected to the photovoltaic panel daily power generation prediction output model construction unit, the photovoltaic panel actual daily power generation acquisition unit, and the spray cleaning unit. It is used to acquire and compare the degree of difference between the predicted value and the actual daily power generation of each photovoltaic panel over a period of time, screen out the photovoltaic panels that need to be cleaned, and drive the spray cleaning unit to perform cleaning on the photovoltaic panels that need to be cleaned. Then, it re-evaluates the degree of difference between the predicted value and the actual daily power generation of each photovoltaic panel. If the degree of difference decreases, it indicates that the cleaning of the photovoltaic panels is effective. If the difference does not improve after a specified number of consecutive cleanings, an alarm message is issued.

[0074] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An automatic cleaning method for photovoltaic panels, characterized in that, Includes the following steps: S1: A spray cleaning unit is configured on the array structure of photovoltaic panels of the same specifications. The spray cleaning unit is used to clean and remove dust from the surface of the photovoltaic panels. S2: Obtain the geographical location and weather conditions of the photovoltaic panel's location, construct a prediction output model for the daily power generation of the photovoltaic panel, and obtain the predicted value of the daily power generation of each photovoltaic panel through the prediction output model; The content of step S2 is to obtain the total daily atmospheric top horizontal irradiance. H 0. Atmospheric transparency factor f atm Photovoltaic system efficiency η sys and the effective light-receiving area of ​​a single photovoltaic panel A Construct a predictive output model for the daily power generation of photovoltaic panels. E ideal Unit: kilowatt-hour; Total daily atmospheric top horizontal radiation H 0 is calculated based on the duration corresponding to the sunrise and sunset angles, solar declination, and daily average atmospheric top normal irradiance; atmospheric transparency factor. f atm It divides the daytime into equal-length periods. t The efficiency of the photovoltaic system is obtained by weighted summation of the normalized weather values ​​and the weather weights for each equal-length period; η sys It is a linear function based on the initial efficiency of the photovoltaic system; S3: Obtain the actual daily power generation of each photovoltaic panel; The content of step S3 is to [process the photovoltaic panels during the daytime period]. T 1, T 2] Within the sampling interval △ t Divide into sections, sampling interval Δ t ≤ Equal Length Period t Instantaneous power within each sampling interval is obtained by measuring with a digital power meter. P t The instantaneous power is accumulated at sampling intervals, and after adding a correction term, a temperature correction coefficient is used to compensate for temperature loss, thus obtaining the actual daily power generation of each photovoltaic panel. E actual The correction term refers to the power of non-photovoltaic panel equipment in the photovoltaic system during daytime hours. S4: Compare the difference between the predicted daily power generation of each photovoltaic panel and the actual daily power generation over a period of time, and screen out the photovoltaic panels that need to be cleaned; The content of step S4 includes the following: S41: Set a first threshold and a second threshold, where the first threshold is less than the second threshold, and compare them over a continuous period of time. k The actual daily power generation of each photovoltaic panel within the unit E actual The sum of the predicted outputs of the daily power generation of each photovoltaic panel E ideal The deviation of the sum PR i sequence values, , , m The total number of photovoltaic panels; 1) If the deviation PR i The sequence values ​​exhibit discontinuous, occasional deviations. These discontinuous, occasional deviations are either less than the first threshold or greater than the second threshold, and the deviations... PR i If more than 95% of the sequence values ​​are within the range of [first threshold, second threshold], proceed to step S44; 2) If the deviation PR i If at least three consecutive singular values ​​in the sequence are all less than the first threshold, it indicates a serious fault or severe weather, and step S42 is executed; 3) If the deviation PR i If at least three consecutive singular values ​​in the sequence are greater than the second threshold, it indicates that there is a measurement error or environmental factors. Proceed to step S43. S42: For situations with serious faults, check the following faults: inaccurate installation tilt angle or azimuth angle of photovoltaic panels; string mismatch between photovoltaic panels; excessive performance degradation of photovoltaic panels due to reaching or exceeding the maximum operating limit; excessive transmission line loss; abnormal reactive power compensation. There is obstruction from buildings, trees, or adjacent photovoltaic panels; The temperature of the photovoltaic panel backsheet is higher than the preset temperature; after the fault is corrected, repeat step S3; in the case of severe weather, remove the covering on the photovoltaic surface in time after the severe weather ends, execute step S5, use the spray cleaning unit to clean all photovoltaic panels, and then repeat step S3. S43: In cases where there are measurement errors or environmental factors, check for the following errors: Verify the actual ground irradiance and the total daily irradiance at the top of the atmosphere. H Deviations between 0 and 1; confirm the presence of environmental reflections in the photovoltaic panel installation environment; verify the efficiency of the photovoltaic system. η sys After adjusting the parameters of the prediction output model that affect the daily power generation of photovoltaic panels, repeat steps S2 and S3. S44: Eliminate sporadic, intermittent biases to determine the actual daily power generation of missing photovoltaic panels. E actual Interpolation is performed to fill in the missing data; then the actual daily power generation of each photovoltaic panel is calculated after interpolation. E actual The daily power generation is sorted from largest to smallest to obtain a sequence. The number of sequence values ​​corresponds one-to-one with the number of photovoltaic panels. Using the average of the top 5% of the daily power generation sequences as a reference, a third threshold is set. Photovoltaic panels corresponding to sequence values ​​that deviate from the reference benchmark by more than the third threshold are marked. This process is repeated for a continuous period of time. k The actual daily power generation of the photovoltaic panels inside E actual Repeat the above interpolation, time-sequence, and labeling process, exceeding the number of times the labeling is performed. k / 3 of the photovoltaic panels are photovoltaic panels that need to be cleaned; S5: After the spray cleaning unit has performed the cleaning operation on the photovoltaic panels that need to be cleaned, it repeats step S4 to reassess whether the difference has improved. If it has improved, it repeats steps S3-S4 in the next comparison cycle. If the difference still does not improve after continuous cleaning for a specified number of times, it issues an alarm message.

2. The automatic cleaning method for photovoltaic panels according to claim 1, characterized in that, If the difference does not improve after a specified number of consecutive cleaning cycles, an alarm message will be issued to alert management personnel that the number of times the marking has been exceeded. k / 3 of the photovoltaic panels were inspected for defects, including the following: back panel temperature, local deformation or hot spots, local shading, damage, loose connection of output terminals, or excessive performance degradation. The photovoltaic panels that do not fall under the defect category are those that need cleaning, while the photovoltaic panels with defects are maintained or replaced.

3. The automatic cleaning method for photovoltaic panels according to claim 1, characterized in that, The first threshold is the predicted output result of the daily power generation of each photovoltaic panel. E ideal The sum of 93%-95%; the first threshold is the predicted output of the daily power generation of each photovoltaic panel. E ideal The sum is 110%-112%; the third threshold is more than 3% of the reference benchmark.

4. The automatic cleaning method for photovoltaic panels according to claim 1, characterized in that, The spray cleaning unit includes several brush rollers and a water inlet pipe. The rotating shafts of the brush rollers are located on the same virtual plane, and the rotating shafts of adjacent brush rollers are set at an angle. The water inlet pipe is used to connect to a water source. The spray cleaning unit performs surface cleaning on all photovoltaic panels in the column where the photovoltaic panel to be cleaned is located and all photovoltaic panels in adjacent columns.

5. An automatic photovoltaic panel cleaning system for implementing the automatic photovoltaic panel cleaning method according to any one of claims 1-4, characterized in that, include: A spray cleaning unit is configured on the array structure of photovoltaic panels; The photovoltaic panel daily power generation prediction output model construction unit is used to construct a photovoltaic panel daily power generation prediction output model based on the geographical location and weather conditions of the photovoltaic panel location, and output the predicted value of the photovoltaic panel daily power generation. The photovoltaic panel actual daily power generation acquisition unit is used to acquire the actual daily power generation of each photovoltaic panel; The difference comparison unit is connected to the photovoltaic panel daily power generation prediction output model construction unit, the photovoltaic panel actual daily power generation acquisition unit, and the spray cleaning unit. It is used to acquire and compare the degree of difference between the predicted value and the actual daily power generation of each photovoltaic panel over a period of time, screen out the photovoltaic panels that need to be cleaned, and drive the spray cleaning unit to perform cleaning on the photovoltaic panels that need to be cleaned. Then, it re-evaluates the degree of difference between the predicted value and the actual daily power generation of each photovoltaic panel. If the degree of difference decreases, it indicates that the cleaning of the photovoltaic panels is effective. If the difference does not improve after a specified number of consecutive cleanings, an alarm message is issued.

Images