Weather detection method, sweeping method, device, equipment and storage medium

Abstract

The invention provides a weather detection method and device, a cleaning method and device, equipment and a storage medium. According to the method, the real-time power generation data of the photovoltaic power station are converted into the line segment graphs according to the fixed time interval, and ordered matching is performed based on the graph shapes of the line segment graphs, so that the detection mode can not only ensure the accuracy of weather detection, but also ensure the integrity of weather detection, and the matching process is relatively simple.

Description

Technical Field

[0001] The present invention relates to the field of photovoltaic technology, and in particular, to a weather detection method, a cleaning method, a device, equipment, and a storage medium. Background Art

[0002] With the increasing global energy demand and the attention to renewable energy, solar power generation has been widely used as a clean and sustainable energy form. However, during the long-term operation of photovoltaic modules, the dust accumulation problem has become one of the important factors restricting their power generation efficiency.

[0003] To solve the problem of dust accumulation on the modules, currently, it is mainly to detect the weather first, and then based on the weather detection results, control the intelligent cleaning robot to execute a fixed cleaning strategy to clean the dust on the photovoltaic modules, such as a cleaning strategy once a day at night or in the morning. Since the existing cleaning control is timed cleaning, and the weather changes are not fixed. If a sandstorm occurs at the power station during the day, and it happens to be in the time period when the dust accumulation on the modules is not judged and cleaned, at this time, no effective cleaning suggestions or instructions can be given to the power station operation and maintenance personnel or the cleaning robot. Then, the sand and dust impact on that day will continue until the end of power generation. For a large-scale ground power station, due to the sudden sand and dust impact, a large amount of power generation will be lost. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a weather detection method, a cleaning detection method, a device, equipment, and a storage medium to solve the problem that the weather detection scheme process of the photovoltaic power station is complex and inaccurate.

[0005] In a first aspect, an embodiment of the present invention provides a weather detection method, including:

[0006] Obtain the real-time power generation data of the photovoltaic power station, where the real-time power generation data includes a plurality of power generation data segments with equal and continuous time intervals;

[0007] Perform abstract transformation processing on each of the power generation data segments to obtain corresponding line segment graphs;

[0008] Match each of the line segment graphs with the state graphs recorded in the preset state matching library, and determine whether the area where the photovoltaic power station is located is in a sandstorm weather based on the matching results.

[0009] In a second aspect, the present invention also provides a cleaning method for photovoltaic modules, the method including:

[0010] Obtain the real-time power generation data of the photovoltaic power station, where the real-time power generation data includes a plurality of power generation data segments with equal and continuous time intervals;

[0011] Perform abstract transformation processing on each of the power generation data segments to obtain corresponding line segment graphs;

[0012] Match each of the line segment graphs with the state graphs recorded in the preset state matching library, and determine whether the area where the photovoltaic power station is located is in a sandstorm weather based on the matching results;

[0013] If so, determine whether to clean the photovoltaic modules based on the real-time power generation data and the corresponding reference power generation data.

[0014] In a third aspect, the present invention also provides a weather detection device, including:

[0015] A first acquisition module, configured to acquire real-time power generation data of a photovoltaic power station, where the real-time power generation data includes a plurality of continuously power generation data segments with equal time intervals;

[0016] A first abstraction module, configured to perform abstract transformation processing on each of the power generation data segments to obtain corresponding line segment graphs;

[0017] A first determination module, configured to match each of the line segment graphs with the state graphs recorded in the preset state matching library, and determine whether the area where the photovoltaic power station is located is in a sandstorm weather based on the matching results.

[0018] In a fourth aspect, the present invention also provides a cleaning device for photovoltaic modules, including:

[0019] A second acquisition module, configured to acquire real-time power generation data of a photovoltaic power station, where the real-time power generation data includes a plurality of continuously power generation data segments with equal time intervals;

[0020] A second abstraction module, configured to perform abstract transformation processing on each of the power generation data segments to obtain corresponding line segment graphs;

[0021] A second determination module, configured to match each of the line segment graphs with the state graphs recorded in the preset state matching library, and determine whether the area where the photovoltaic power station is located is in a sandstorm weather based on the matching results;

[0022] A cleaning module, configured to determine whether to clean the photovoltaic modules based on the real-time power generation data and the corresponding reference power generation data when it is determined that the area where the photovoltaic power station is located is in a sandstorm weather.

[0023] Fifth aspect, the present invention further provides an electronic device, including: a memory and at least one processor, instructions are stored in the memory, and the memory and the at least one processor are interconnected through a line; the at least one processor calls the instructions in the memory to enable the electronic device to execute the respective steps of the above-mentioned weather detection method or execute the respective steps of the above-mentioned cleaning method of the photovoltaic module.

[0024] Sixth aspect, the present invention further provides a computer-readable storage medium, instructions are stored in the computer-readable storage medium, when it runs on a computer, it enables the computer to execute the respective steps of the above-mentioned weather detection method or execute the respective steps of the above-mentioned cleaning method of the photovoltaic module.

[0025] The embodiments of the present invention bring the following beneficial effects:

[0026] The above-mentioned weather detection method, cleaning method, device, equipment and storage medium, the method includes obtaining real-time power generation data of a photovoltaic power station, the real-time power generation data includes a plurality of power generation data segments with equal and continuous time intervals; performing abstract transformation processing on each of the power generation data segments to obtain corresponding line segment graphics; matching each of the line segment graphics with each state graphic recorded in a preset state matching library, and determining whether the area where the photovoltaic power station is located is in a sandstorm weather based on the matching result.

[0027] In this way, by converting the real-time power generation data into line segment graphics at fixed time intervals and performing orderly matching based on the graphic shape of the line segment graphics, such a detection method can not only ensure the accuracy of weather detection, but also has a relatively simple matching process, and at the same time can ensure the integrity of weather detection.

[0028] Other features and advantages of the present invention will be described in the following specification, and, in part, will be obvious from the specification, or will be understood by implementing the present invention. The objectives and other advantages of the present invention are achieved and obtained by the structures specifically pointed out in the specification, claims and drawings.

[0029] To make the above-mentioned objectives, features and advantages of the present invention more obvious and understandable, the following specific preferred embodiments are given, and in conjunction with the accompanying drawings, the detailed description is as follows. Description of the Drawings

[0030] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. Obviously, the following drawings are some embodiments of the present invention. For those skilled in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0031] Figure 1 Schematic diagram of an embodiment of the weather detection method provided by an embodiment of the present invention;

[0032] Figure 2 Schematic diagram of generating a line segment graph provided by an embodiment of the present invention;

[0033] Figure 3 Schematic diagram of the reference power curve of a photovoltaic power station provided by an embodiment of the present invention;

[0034] Figure 4 Schematic diagram of the division of the actual power curve provided by an embodiment of the present invention;

[0035] Figure 5 is a schematic diagram of six line segment graphs provided by an embodiment of the present invention;

[0036] Figure 6 Schematic diagram of five states provided by an embodiment of the present invention;

[0037] Figure 7 Schematic diagram of an embodiment of the cleaning method of a photovoltaic module provided by an embodiment of the present invention;

[0038] Figure 8 Schematic diagram of a structure of a weather detection device provided by an embodiment of the present invention;

[0039] Figure 9 Schematic diagram of a structure of a cleaning device for a photovoltaic module provided by an embodiment of the present invention;

[0040] Figure 10 Schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. Detailed implementation manners

[0041] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0042] In the related art, the detection of weather is obtained by analyzing the power generation data of a photovoltaic power station, such as calculating its volatility, displacement difference, etc. However, this analysis method requires setting a large number of thresholds to judge various situations. Due to various uncertain factors such as possible failures of the inverter itself and abnormally changeable weather, it is relatively difficult to implement and has low accuracy to judge various possible situations by setting a set of thresholds.

[0043] See Figure 1, a weather detection method proposed in an embodiment of the present invention, which is mainly applied to the detection of sandstorm weather in a substation. The method includes the following steps:

[0044] Step S101, obtain the real-time power generation data of the photovoltaic power station, where the real-time power generation data includes multiple continuously generated data segments with equal time intervals.

[0045] Specifically, the real-time power generation data of the photovoltaic power station is obtained from the data acquisition and control device in the photovoltaic power station. The real-time power generation data can be power generation power data or working current and voltage data. The data acquisition and control device can be a photovoltaic inverter in the photovoltaic power station or other devices such as a busbar box, and no specific limitation is made here. As an example, taking the actual power curve of each photovoltaic inverter in the photovoltaic power station as an example, the actual power curve includes multiple continuously power curve segments with equal time intervals. Here, the multiple continuously power curve segments with equal time intervals should be understood as a fixed sampling frequency, and the sampled power curve segments have a sequential relationship in time.

[0046] In practical applications, when obtaining the real-time power curve, first determine the analysis time period, perform regular sampling within this time period, start sampling from the beginning of this time period, and then sample at regular intervals. The time corresponding to each sampled data is within this time period until the end of this time period, and finally, all the collected data is concatenated to obtain the actual power curve.

[0047] The actual power curve can be understood as real-time or for a time period. That is to say, when obtaining the actual power curve, it can be collected in real-time or by a timed method. The timed method is specifically to intercept a curve segment for a period of time from the power curve output by the photovoltaic inverter as the actual power curve of the photovoltaic power station, that is, it can be a curve segment containing multiple collection times or a curve segment containing one collection time.

[0048] In practical applications, when obtaining the actual power curve, it can be set that the time length of each obtained actual power curve is defined based on the time length of the complete state of the sandstorm weather. That is to say, the time length of each obtained actual power curve should be the total time length of the line segment graphics including the 5 states for judging the sandstorm weather.

[0049] Of course, the actual power curve can be a complete curve composed of 5 continuously sampled curve segments in the time dimension, and the time span corresponding to each sampled curve segment is equal.

[0050] Specifically, first collect the power generation amounts generated by each photovoltaic inverter in the photovoltaic power station in real time at each moment, calculate the corresponding real-time power generation power based on the power generation amounts, and generate an actual power curve based on the real-time power generation power at each moment. It should be noted that in this embodiment, the period of the obtained actual power curve is taken as an example of a day. Of course, it can also be a period of at least one hour, etc.

[0051] In some embodiments, specifically, when it is determined that the photovoltaic inverters in the photovoltaic power station are in a normal operating state, obtain the power data of each photovoltaic inverter to obtain the actual power curve of the photovoltaic power station. Among them, the photovoltaic inverters being in a normal operating state means that the photovoltaic inverters do not malfunction. That is, when the photovoltaic inverters do not malfunction, obtain the actual power curve of the photovoltaic power station for subsequent detection of sandstorm weather based on the actual power curve of the photovoltaic power station.

[0052] Step S102: Perform abstraction processing on each power generation data segment to obtain the corresponding line segment graph.

[0053] Specifically, draw the corresponding power curve segments based on the changes in the data in each power generation data segment, and then perform abstraction transformation processing on each power curve segment to obtain the corresponding line segment graph.

[0054] In this step, first, divide the obtained actual power curve according to time continuity to obtain multiple power curve segments, and then abstract each power curve segment into a line segment graph.

[0055] Specifically, divide the actual power curve into multiple power curve segments with equal time lengths according to a preset time interval, and then use a graphical abstraction method to convert each power curve segment into a line segment graph. Here, the power curve segment can be understood as a smooth curve, and the line segment graph refers to a graph composed of at least two straight line segments.

[0056] In some embodiments, this abstraction processing is specifically to first abstract each of the power curve segments into N target points, where the target point refers to a representation point of power, and then form the corresponding line segment graph with N - 1 line segments drawn by the N target points, where N is greater than or equal to 3. That is, by extracting the target points in each power curve segment and then connecting the target points in the order of time with straight lines, the line segment graph is obtained.

[0057] Specifically, the abstracting each of the power curve segments into N target points and forming the corresponding line segment graph with N - 1 line segments drawn by the N target points includes:

[0058] Evenly divide each of the power curve segments into N equal parts, and sort the power data in each equal part. Here, the sorting can be in ascending order or in descending order;

[0059] Select the middle part of each equal portion of the power data after sorting, and calculate the power average value to obtain N target points of the corresponding power curve segment;

[0060] Connect the N target points in time sequence to generate a line segment graph of the corresponding power curve segment.

[0061] like Figure 2 As shown, a corresponding power curve segment OP is selected from the actual power curve, and then the target point in the power curve segment OP is extracted. Here, three target points (A, B, C) are taken as an example, and each target point is extracted by calculating the average value of power data, that is, the power curve segment OP is divided into three equal small segments, and then the average value of the power in each small segment is calculated to obtain the target point of the small segment. After the target points of the three small segments are calculated, the three target points are connected in series with a straight line to obtain the line segment graph of the power curve segment OP.

[0062] Furthermore, in the process of calculating the average power value in each small segment to obtain the target point of the small segment, the power values at each moment in the small segment are first sorted according to the size of the values, and then the 1 / 3 minimum values and the 1 / 3 maximum values are eliminated, leaving the 1 / 3 power values in the middle of the sorting, and finally the average of the 1 / 3 power values in the middle is calculated to obtain the target point of the small segment, that is, to obtain the coordinate information of the target point of the small segment. The median data is selected by sorting to remove the numerical jitter and outliers in the actual power curve, thereby improving the accuracy of abstraction, such as Figure 4 As shown, until the target points of the three small segments of the power curve segment are calculated, they are connected in series to obtain the line segment curve. Similarly, the principle of the abstract line segment graph with more than three target points is the same as that of the three target points, which will not be repeated here.

[0063] In some embodiments, the abstracting each of the power curve segments into N target points, and forming a corresponding line segment graph with N-1 line segments drawn by the N target points, further includes:

[0064] Calculate the coordinate information of each target point in the line segment graph based on the power average value of each equal portion;

[0065] The angle of the line segment graph, the slopes of N-1 line segments in the line segment graph, and the average slope of the line segment graph are calculated based on the coordinate information of each target point.

[0066] In practical applications, the calculation of the coordinate information for each target point is based on the calculated average power. For example, the average power of the target point is used as the ordinate, and the corresponding time period is used as the abscissa to obtain the coordinate information of the target point, where the time period is determined by the sampling time when collecting the actual power curve. Then, based on the abscissa and ordinate in the coordinate information, the angle between two adjacent line segments in the line graph and the slope of each line segment are calculated. In addition, it also includes calculating the average slope of the line graph based on the slopes of each line segment.

[0067] For example Figure 2 For the line graph ABC obtained in [reference], based on the abscissa and ordinate values of the three target points (A, B, C), the angle between line segment AB and line segment BC in the line graph and the slopes of line segment AB and line segment BC are calculated. Finally, the average value of the slopes of line segment AB and line segment BC is calculated to obtain the average slope of the line graph.

[0068] In addition, when there are more than three target points in the line graph, when calculating the angle, first select the first line segment corresponding to the two target points with the earliest time sorting in the line graph and the (N - 1)th line segment of the two target points with the latest time sorting, and calculate the angle between these two line segments. The calculation methods of the slope and the average slope are the same as those for three target points.

[0069] Step S103: Match each line graph with each state graph recorded in the preset state matching library, and determine whether the area where the photovoltaic power station is located is in a sandstorm weather based on the matching result.

[0070] Among them, the preset state matching library is the state dictionary of sandstorms, which records the state composition of sandstorm weather and the sorting order of each state. This step determines whether the area where the photovoltaic power station is located is in a sandstorm weather by matching each line graph with each state graph in the state dictionary and based on the matching result.

[0071] Specifically, based on the shape of each line graph, compare it with the shape of each state graph recorded in the preset state matching library to determine whether it is a sandstorm weather.

[0072] In practical applications, the determination of the shape of each line graph can be determined by any one or more parameters such as the angle, slope, and average slope between line segments in the line graph, and can also be further determined by combining information such as the magnitude relationship and sorting order of the coordinate values of the target points. In addition, in order to effectively ensure the accuracy of the sandstorm weather determination result, after matching the state corresponding to each line graph, it is also necessary to determine whether the time sequence relationship corresponding to each state satisfies the order of occurrence of each state corresponding to the sandstorm weather. If it is satisfied, it is determined as a sandstorm weather.

[0073] Of course, the sequential determination of each state can be completed while matching the state, or can be completed after the matching. When implemented while matching the state, during the matching process, first sort each line segment graph in chronological order, and then match the state of each line segment graph in sequence according to the sorting.

[0074] In some embodiments, for matching each of the line segment graphs with each state graph recorded in the preset state matching library in step 103, and determining whether the area where the photovoltaic power station is located is in a sandstorm weather based on the matching result, it includes:

[0075] Determine the shape of each of the line segment graphs based on the included angle and / or average slope of each of the line segment graphs, and compare the shape of the line segment graphs with the shapes of each state graph recorded in the preset state matching library to obtain a comparison result;

[0076] Determine the state of each of the power curve segments based on the comparison result, and determine whether the area where the photovoltaic power station is located is in a sandstorm weather based on the state of each of the power curve segments.

[0077] Wherein, determining the shape of each of the line segment graphs based on the included angle and / or average slope of each of the line segment graphs includes determining the shape of each of the line segment graphs based on the included angle, average slope, and the included angle and average slope of each of the line segment graphs.

[0078] In some embodiments, for determining the shape of each of the line segment graphs based on the included angle of each of the line segment graphs, it includes:

[0079] Judge whether the included angle of each of the line segment graphs is greater than the included angle threshold;

[0080] If the included angle of each of the line segment graphs is greater than the included angle threshold, then determine the shape of each of the line segment graphs based on the average slope;

[0081] If the included angle of each of the line segment graphs is not greater than the included angle threshold, then calculate the height difference between each line in each of the line segment graphs, and determine the shape of the corresponding line segment graph based on the height difference between each line.

[0082] This included angle threshold is an empirical value obtained through the analysis of the power curves of photovoltaic modules under multiple sandstorms. When it is greater than the included angle threshold, it indicates that the combination angle between the lines in the line segment graph is relatively large, and the corresponding shape may be the shape of the first category. On the contrary, it is the shape of the second category. The shapes of the first category include flat, slightly rising, slightly falling, and falling, and the shapes of the second category include upper right convex angle and lower left convex angle.

[0083] When the included angle is greater than the included angle threshold, determine the shape of the line segment graph based on the average slope of the corresponding line segment graph, that is, the shape is one of flat, slightly rising, slightly falling, and falling; when the included angle is less than the included angle threshold, continue to calculate the height difference of each line segment in the corresponding line segment graph based on the coordinate information of each target point in the line segment graph, and determine the shape of the corresponding line segment graph based on the height difference of each line segment, that is, the shape is a right upper convex angle or a left lower convex angle.

[0084] In some embodiments, for determining the shape of each of the line segment graphs based on the average slope of each of the line segment graphs, it includes:

[0085] Judge whether the average slope is less than the average slope threshold;

[0086] If the average slope is less than the average slope threshold, determine that the shape of the corresponding line segment graph is the first shape, that is, flat, slightly rising or slightly falling;

[0087] If the average slope is not less than the average slope threshold, determine that the shape of the corresponding line segment graph is the third shape, that is, falling.

[0088] For determining the shape of the corresponding line segment graph based on the height difference of each line segment, it includes:

[0089] Judge whether there is a slope greater than the preset threshold among the slopes of N - 1 line segments in the line segment graph, where the preset threshold is 1;

[0090] If there is a slope greater than the preset threshold among the slopes of N - 1 line segments in the line segment graph, then judge the magnitude relationship between the heights of the adjacent front and back two line segments among the N - 1 line segments to obtain a judgment result;

[0091] If the judgment result is that the height of the previous line segment is less than the height of the next line segment, determine that the shape of the corresponding line segment graph is the second shape, that is, a right upper convex angle;

[0092] If the judgment result is that the height of the previous line segment is greater than the height of the next line segment, determine that the shape of the corresponding line segment graph is the fourth shape, that is, a left lower convex angle.

[0093] In some embodiments, compare the shape of the line segment graph with the shapes of each state graph recorded in the preset state matching library to obtain a comparison result, including:

[0094] If the shape of the line segment graph is the first shape (that is, a in Figure 5), then compare the first shape with the shapes of each state graph recorded in the preset state matching library to obtain that the state of the corresponding line segment graph is the first state (that is, Figure 6 the state 1 in Figure 6State 5) in it; among which, the judgment of the first state and the fifth state further includes: obtaining the reference power generation curve of the photovoltaic power station, then calculating the distance between the corresponding power curve segment and the corresponding segment in the reference power generation curve, and determining whether it is the first state or the fifth state based on the distance. The distance corresponding to the first state is smaller, and the distance corresponding to the fifth state is larger. It can be seen that after calculating the distance here, when the distance is greater than the preset distance value or within the preset distance range, it is determined as the fifth state.

[0095] If the shape of the line segment graph is the second shape (i.e., b in Figure 5), then compare the second shape with the shapes of the state graphs recorded in the preset state matching library to obtain the state of the corresponding line segment graph as the second state (i.e., Figure 6 State 2) in it;

[0096] If the shape of the line segment graph is the third shape (i.e., c in Figure 5), then compare the third shape with the shapes of the state graphs recorded in the preset state matching library to obtain the state of the corresponding line segment graph as the third state (i.e., Figure 6 State 3) in it;

[0097] If the shape of the line segment graph is the fourth shape (i.e., d in Figure 5), then compare the fourth shape with the shapes of the state graphs recorded in the preset state matching library to obtain the state of the corresponding line segment graph as the fourth state (i.e., Figure 6 State 4) in it.

[0098] In this embodiment, in order to improve the accuracy of detecting sandstorm weather, while matching the states corresponding to each power curve segment, it also includes setting a state flag dictionary to record the matching situation of each state, that is, after determining the state of each power curve segment based on the comparison result, it further includes:

[0099] Determine whether there is a state with a higher ranking among the states of the power curve segment;

[0100] If there is, then judge whether the flag bit of the state with a higher ranking is valid;

[0101] If it is valid, modify the flag bit of the state of the power curve segment to be valid;

[0102] If it is invalid, the state of the power curve segment is marked after waiting for the flag bit of the state with a higher ranking to be modified to be valid.

[0103] The flag bits obtained after the above matching are recorded in the state update flag bit dictionary. The use of this flag bit dictionary ensures the orderliness of the states. Only after the previous state is marked as True can the next state be judged, thus realizing the continuity and controllability of the states.

[0104] In some embodiments, determining whether the area where the photovoltaic power station is located is in a sandstorm weather based on the states of the respective power curve segments includes:

[0105] Comparing the determined states of all the power curve segments with the state combination of the sandstorm weather;

[0106] If all the states determined by the comparison are consistent with the state combination, sorting the corresponding states based on the occurrence time of the respective power curve segments to obtain the sorting order of all the states;

[0107] Judging whether the sorting order is consistent with the sorting order of each state in the state combination;

[0108] Determining whether the area where the photovoltaic power station is located is in a sandstorm weather based on the result of the judgment.

[0109] In some embodiments, determining whether the area where the photovoltaic power station is located is in a sandstorm weather based on the result of the judgment includes:

[0110] If the result of the judgment is that the sorting order is consistent with the sorting order of each state in the state combination, judging whether the states of all the power curve segments are all valid;

[0111] If so, determining that the area where the photovoltaic power station is located is in a sandstorm weather;

[0112] If not, determining that the area where the photovoltaic power station is located is not in a sandstorm weather.

[0113] That is, in the process of determining the sandstorm weather, in addition to determining that the actual power curve needs to include all the states of the sandstorm, it is also necessary to determine the time sequence of the occurrence of each state. When both the sequence and the state meet the specific conditions of the sandstorm weather, it is determined to be a sandstorm weather, otherwise it is determined to be other weather.

[0114] In some embodiments, in addition to identifying the sandstorm weather by matching the states and the state sorting, it can be identified by matching some states and converting the states into quantization values, so as to realize the identification of the sandstorm weather even in the case of incomplete states, and improve the identification accuracy and success probability.

[0115] Specifically, determining whether the area where the photovoltaic power station is located is in a sandstorm weather based on the states of the respective power curve segments further includes:

[0116] If all the determined states are inconsistent with the state combination or it is determined that the sorting order is inconsistent with the sorting order of each state in the state combination, identify whether there is a necessary state in the state combination among all the determined states of the power curve segments;

[0117] If there is, calculate the total numerical identifier of all the states of the power curve segments based on the corresponding relationship between the preset state and the numerical identifier;

[0118] Match the total numerical identifier with the numerical identifier range of sandstorm occurrence;

[0119] If the total numerical identifier is within the first numerical identifier range of sandstorm occurrence, determine that a sandstorm weather has occurred in the area where the photovoltaic power station is located;

[0120] If the total numerical identifier is within the second numerical identifier range of sandstorm occurrence, determine that a sandstorm weather has occurred in the area where the photovoltaic power station is located, where the boundary value of the first numerical identifier range is less than the boundary value of the second numerical identifier range.

[0121] Of course, the solution for identifying sandstorm weather by matching some states and converting the states into quantization values can also be that after matching the states of each power curve segment, directly convert the matched states into numerical identifiers and identify sandstorm weather based on the total of the numerical identifiers.

[0122] In practical applications, due to some situations, the turning points of the graph may not be clearly shown. Therefore, the graph shapes of the above 5 stages are simplified into 3 stages, omitting the corners before and after the rising edge. The simplified 3 stages are: a. Stage 1, including: flat, slightly rising, slightly falling; b. Stage 2, including: rising; c. Stage 3, including, flat, slightly rising, slightly falling. That is Figure 6 the states 1, 3 and 5 in

[0123] For Figure 6 the 5-stage states in, respectively configure corresponding numerical identifiers to uniquely identify this stage, which are:

[0124] a. Stage 1, configure the numerical value 16;

[0125] b. Stage 2, configure the numerical value 8;

[0126] c. Stage 3, configure the numerical value 4;

[0127] d. Stage 4, configure the numerical value 2;

[0128] e. Stage 5, configure the numerical value 1;

[0129] By calculating the sum of the values in 5 stages, the successfully matched stage can be uniquely determined.

[0130] After converting the states in the state combination into numerical identifiers, calculate the sum of the numerical identifiers of all states and compare it with a preset range or numerical threshold.

[0131] If the sum of the numerical identifiers of the matched stages is 21, it means that stages 1, 3, and 5 in the complete 5 stages have been matched, that is, the simplified 3-stage matching method. At this time, by checking the amplitude of the curve, that is, the dust accumulation loss, if it is greater than the threshold, it is determined that a sandstorm has occurred.

[0132] If the sum of the numerical identifiers of the matched stages is 31, it means that the complete 5 stages have all been matched, and there is no need to further judge the threshold to determine that a sandstorm has occurred.

[0133] Other numerical values indicate that the matching is not completed and further waiting and calculation are required.

[0134] The above weather detection method abstracts the real-time power generation data of a photovoltaic power station into three coordinate points for characterization, and graphically represents the shape of the image composed of two line segments drawn by the three coordinate points. This approach is equivalent to simplifying the complex features of the power curve, has stronger characterization ability, and greatly reduces the amount of calculation; then creates a time window, performs rolling calculations on the changes in the shapes of the graphic series, and obtains the occurrence and end times of sand and dust by analyzing the state update flag dictionary in sequence. To solve the problem that the existing weather detection scheme for photovoltaic power stations is complex and inaccurate.

[0135] As Figure 7 shown, it is an embodiment of the cleaning method for a photovoltaic module provided by an embodiment of the present invention. This method is implemented by using a dynamic graphic shape matching algorithm, a state dictionary and a logical judgment mechanism, and a state update flag dictionary, etc. This method first abstracts the inverter power generation curve data into at least two points, and graphically represents the shape of the image composed of two line segments drawn by adjacent points; then creates a time window, performs rolling calculations on the changes in the shapes of the graphic series, and obtains the occurrence and end times of sand and dust by analyzing the state update flag dictionary in sequence. When the sandstorm ends, a dust accumulation warning is given in a timely manner when the dust accumulation loss threshold of the sandstorm is reached.

[0136] Among them, the dynamic graphic shape matching algorithm: converts the inverter reference power data and actual power data into geometric graphic shapes, and realizes the dynamic matching of the graphic shapes through a specific algorithm. This algorithm includes dividing the power data into time windows, calculating the mean value of the data within each time window, and creating corresponding geometric graphics, and then matching these graphic shapes with a preset state dictionary to judge the current weather condition. In practical applications, the algorithm can be a matching model. For example, a neural network model as a matching model can learn the state change rules of sandstorms and the shapes of each state, and then use this matching model to perform real-time recognition on the curve segments in the power curve to output the recognition results, and then sort all the recognition results in chronological order to obtain the sorting of the states, so as to output the detection results of sandstorms.

[0137] State dictionary and logical judgment mechanism: includes defining 6 preset graphic shapes and the corresponding state dictionary. Through this state dictionary, the system can determine the current weather state according to the matched graphic shape. In addition, a logical judgment mechanism is designed to ensure that each state is judged in a specific order, so as to ensure the accuracy and integrity of the state.

[0138] State update flag bit dictionary: used to mark the states that have been judged. The use of this flag bit dictionary ensures the orderliness of the states. Only after the previous state is marked as True can the next state be judged, thus realizing the continuity and controllability of the states.

[0139] Based on the above basis, the cleaning detection method provided in this embodiment specifically includes the following steps:

[0140] Step S701, obtain the real-time power generation data of the photovoltaic power station, and the real-time power generation data includes multiple continuously generated power data segments with equal time intervals.

[0141] The real-time power generation data can be the power generation data of the photovoltaic inverter, such as the power generation power curve, working current curve, working voltage curve... The photovoltaic power station includes photovoltaic modules and a photovoltaic inverter, and the photovoltaic inverter is connected to the photovoltaic modules. For example, the photovoltaic inverter can be a centralized inverter or a string inverter.

[0142] In this embodiment, for the power generation of photovoltaic modules, the actual power generation is greatly related to the weather. Abnormal weather will affect the power generation power of photovoltaic modules, and the power generation power of photovoltaic modules will affect the power generation of the photovoltaic power station. Therefore, in this embodiment, by obtaining the actual power generation power of the photovoltaic power station and analyzing the actual power generation power, the detection of abnormal weather is realized.

[0143] Obtain the actual power generation of each PV inverter in the PV power station in real time, and construct the actual power generation curve of the PV inverter according to the actual power generation of the PV inverter obtained at each moment. It should be noted that the period of weather detection in this embodiment can be in days, so the actual power generation curve of the PV power station can be the curve of the actual power generation of the PV power station changing with time on the day of detection.

[0144] In some embodiments, since the effective power generation time of the PV modules is between sunrise and sunset in a day, the actual power generation of the PV power station can be obtained in real time between sunrise and sunset to construct the actual power generation curve of the PV power station, that is, the actual power generation curve of the PV power station can be the curve of the actual power generation of the PV power station changing with time between sunrise and sunset on the day of detection.

[0145] Step S702: Perform abstract transformation processing on each power generation data segment to obtain the corresponding line segment graph.

[0146] Specifically, generate the corresponding power generation curve based on each of the power generation data segments; abstract each of the power generation curve segments into N target points, and form the corresponding line segment graph with N - 1 line segments drawn by the N target points, where N is greater than or equal to 3.

[0147] As Figure 2 shown, taking the example of setting three target points for the power generation curve segment to elaborate on the abstracted line segment graph.

[0148] First, determine the time length of each graph from the preset state dictionary to construct a time window, and divide the actual power generation curve based on this time window to obtain several equal - sized power generation curve segments. For example, divide the actual power generation curve into 5 power curve segments according to the number of sandstorm states, and then perform the conversion of the line segment graph for each power generation curve segment. As Figure 2 and 4 shown, divide the power curve segment OP into three equal parts, then sort the power generation data in each equal part according to the numerical size, and obtain the mean value of the middle 1 / 3 data through sorting. Then, convert these mean values into geometric figures, specifically including three points A, B, and C, whose horizontal and vertical coordinates are calculated respectively. Then, connect points A, B and B, C to obtain the line segment graph ABC composed of line segments AB and BC.

[0149] Assume that the time window of the power generation curve segment OP is T, and divide the data in it into 3 equal parts on average, denoted as d1, d2, and d3 respectively; sort each part of the data in ascending order, and take the mean value of the middle 1 / 3 data. As Figure 4Data in the t2 period. Three values are obtained, denoted as a, b, and c, and three points A, B, and C are obtained. Here, the purpose of taking the middle 1 / 3 of the data is to remove the numerical jitter and outliers in the power generation curve segment.

[0150] Step S703: Match each line segment graph with each state graph recorded in the preset state matching library, and determine whether the area where the photovoltaic power station is located is in a sandstorm weather based on the matching result.

[0151] In this step, the state matching library can be a state dictionary for sandstorm weather. During the matching process, a state field and a state update flag bit dictionary are used to improve the accuracy of matching various states, where:

[0152] The state dictionary is a data structure used to map different graph shapes to specific weather states. In this proposal, the state dictionary defines six preset graph shapes, as shown in Figure 5. Each shape corresponds to a specific sandstorm weather state. For example, if the shape of the power data within a certain time window matches the preset "slightly rising" shape, the system will map it to the corresponding sandstorm weather state, such as "State 2" according to the state dictionary. This mapping relationship helps the system identify the current weather condition.

[0153] The state update flag bit dictionary is a data structure used to mark whether each state has been matched. In this proposal, the state update flag bit dictionary is used to ensure the continuity and orderliness of the states. Whenever the system matches a specific state, the corresponding state flag bit will be set to True. Then, the system will check these flag bits to ensure that they are set to True in a specific order. Only after the flag bit of the previous state is set to True can the next state be judged. This mechanism ensures the orderliness of the states and enables the system to perform further operations under specific conditions, such as determining whether to perform cleaning.

[0154] In some embodiments, before the matching, it is necessary to determine the shape of each line segment graph, and the determination of the shape needs to be based on data such as the included angle and slope. Specifically:

[0155] Calculate the coordinate information of each target point in the line segment graph based on the average power generation value of each equal part;

[0156] Calculate the included angle between two adjacent line segments in the line segment graph based on the coordinate information of each target point, as well as the slopes of N - 1 line segments in the line segment graph and the average slope of the line segment graph.

[0157] In practical applications, it is calculated through the abscissa and ordinate of each target point in the line segment graph. Then, connect points A, B and B, C to obtain line segments AB and BC, and calculate the included angle and slope of these two line segments.

[0158] That is, first create points A, B, and C, and use a, b, and c as the ordinates of points A, B, and C respectively, denoted as A y , B y , C y ; then calculate the abscissas of points A, B, and C in sequence, denoted as:

[0159] A x = A y

[0160] B x = A x + (a + b + c) / 3

[0161] C y = B x + (a + b + c) / 3

[0162] Further, connect points A and B, and B and C respectively to obtain line segment AB and line segment BC; calculate the included angle between line segment AB and line segment BC, denoted as ∠ABC; calculate the slopes of line segment AB and line segment BC respectively, denoted as k AB , k BC ; calculate the average value of the absolute values of the slopes of the two lines, denoted as: k avg = (|k AB | + |k BC |) / 2.

[0163] In some embodiments, the matching process includes: determining the shapes of the respective line segment graphs based on the included angles and / or average slopes of the respective line segment graphs, and comparing the shapes of the line segment graphs with the shapes of the respective state graphs recorded in the preset state matching library to obtain a comparison result; determining the states of the respective power generation curve segments based on the comparison result.

[0164] Step S704, if so, determine whether to clean the photovoltaic modules based on the real-time power generation data and the corresponding reference power generation data.

[0165] In this step, calculate the ratio of the real-time power generation data to the corresponding reference power generation data; if the ratio is less than the threshold, clean the photovoltaic modules; if the ratio is not less than the threshold, do not clean the photovoltaic modules.

[0166] In practical applications, determine whether to clean the photovoltaic modules based on the actual power curve and the corresponding reference power curve.

[0167] Specifically, calculate the ratio of the actual power curve to the corresponding reference power curve; if the ratio is less than the threshold, clean the photovoltaic modules; if the ratio is not less than the threshold, do not clean the photovoltaic modules.

[0168] In the above method, by converting the actual power curve into a geometric shape and then combining it with a preset state dictionary through a dynamic matching algorithm, the rapid and accurate identification of the actual sandstorm weather conditions is achieved. Compared with traditional meteorological sensors, this method does not rely on complex sensor devices, reduces the hardware cost of the system, and improves the real-time performance and applicability. Secondly, the state dictionary and logical judgment mechanism introduced in the technology ensure the continuity and accuracy of monitoring. Through the orderly matching of the graphic shapes, the system can judge different states in a specific order to ensure the integrity of the states. This systematic judgment mechanism makes

[0169] the monitoring results more reliable, avoiding misjudgment and missed reports, and improving the stability and credibility of the system.

[0170] At the same time, this method also gives accurate reference for the operation and maintenance personnel or the cleaning robot to start the cleaning strategy, solves the problem of short-term large amount of dust accumulation on the photovoltaic modules without adding any hardware and data source input, and improves the efficiency and reliability of the solar power generation system.

[0171] The weather detection method provided by the embodiment of the present invention has been described above. Next, the weather detection device provided by the embodiment of the present invention will be described. Please refer to Figure 8 an embodiment of the weather detection device provided by the embodiment of the present invention. The device includes:

[0172] The first acquisition module 910 is used to acquire the real-time power generation data of the photovoltaic power station, where the real-time power generation data includes a plurality of continuously generated power data segments with equal time intervals;

[0173] The first abstraction module 920 is used to perform abstraction conversion processing on each of the power data segments to obtain corresponding line segment graphics;

[0174] The first determination module 930 is used to match each of the line segment graphics with each of the state graphics recorded in the preset state matching library, and determine whether the area where the photovoltaic power station is located is in a sandstorm weather based on the matching result.

[0175] Optionally, the first abstraction module 920 is specifically used for:

[0176] Generating corresponding power curve segments based on each of the power data segments; abstracting each of the power curve segments into N target points, and forming corresponding line segment graphics with N - 1 line segments drawn by the N target points, where N is greater than or equal to 3.

[0177] Optionally, the first abstraction module 920 is specifically used for:

[0178] Divide each of the power curve segments into N equal parts on average, and sort the power data in each equal part;

[0179] Select the middle part of the power data in each equal part after sorting, and calculate the average power to obtain N target points for the corresponding power curve segment;

[0180] Connect the N target points in chronological order to generate a line graph of the corresponding power curve segment.

[0181] Optionally, the first abstraction module 920 is further specifically configured to:

[0182] Calculate the coordinate information of each target point in the line graph based on the average power of each equal part;

[0183] Calculate the included angle of the line graph, the slopes of N - 1 line segments in the line graph, and the average slope of the line graph based on the coordinate information of each target point.

[0184] Optionally, the first determination module 930 is specifically configured to:

[0185] Determine the shape of each line graph based on the included angle and / or average slope of each line graph, and compare the shape of the line graph with the shapes of each state graph recorded in the preset state matching library to obtain a comparison result;

[0186] Determine the state of each power curve segment based on the comparison result, and determine whether the area where the photovoltaic power station is located is in a sandstorm weather based on the state of each power curve segment.

[0187] Optionally, the first determination module 930 is specifically configured to:

[0188] Judge whether the included angle of each line graph is greater than the included angle threshold;

[0189] If so, determine the shape of each line graph based on the average slope;

[0190] If not, calculate the height difference of each line segment in each line graph, and determine the shape of the corresponding line graph based on the height difference of each line segment.

[0191] Optionally, the first determination module 930 is specifically configured to:

[0192] Judge whether the average slope is less than the average slope threshold;

[0193] If so, determine the shape of the corresponding line graph as the first shape;

[0194] If not, determine the shape of the corresponding line graph as the third shape.

[0195] Optionally, the first determination module 930 is specifically configured to:

[0196] Determine whether there is a slope greater than a preset threshold among the slopes of N-1 line segments in the line graph;

[0197] If so, determine the magnitude relationship between the heights of two adjacent front and back line segments among the N-1 line segments to obtain a determination result;

[0198] If the determination result is that the height of the previous line segment is less than the height of the next line segment, determine that the shape of the corresponding line graph is the second shape;

[0199] If the determination result is that the height of the previous line segment is greater than the height of the next line segment, determine that the shape of the corresponding line graph is the fourth shape.

[0200] Optionally, the first determination module 930 is specifically configured to:

[0201] If the shape of the line graph is the first shape, compare the first shape with the shapes of the state graphs recorded in the preset state matching library to obtain that the state of the corresponding line graph is the first state or the fifth state;

[0202] If the shape of the line graph is the second shape, compare the second shape with the shapes of the state graphs recorded in the preset state matching library to obtain that the state of the corresponding line graph is the second state;

[0203] If the shape of the line graph is the third shape, compare the third shape with the shapes of the state graphs recorded in the preset state matching library to obtain that the state of the corresponding line graph is the third state;

[0204] If the shape of the line graph is the fourth shape, compare the fourth shape with the shapes of the state graphs recorded in the preset state matching library to obtain that the state of the corresponding line graph is the fourth state.

[0205] Optionally, the weather detection device further includes a marking module 940, which is specifically configured to:

[0206] Determine whether there is a state with a higher ranking among the states of the power curve segment;

[0207] If there is, determine whether the flag bit of the state with a higher ranking is valid;

[0208] If it is valid, modify the flag bit of the state of the power curve segment to be valid;

[0209] If it is invalid, the state of the power curve segment is marked after waiting for the flag bit of the state with a higher ranking to be modified to be valid.

[0210] Optionally, the first determination module 930 is specifically configured to:

[0211] Compare the states of all the determined power curve segments with the state combination of the sandstorm weather;

[0212] If all the states determined by the comparison are consistent with the state combination, sort the corresponding states based on the occurrence time of each power curve segment to obtain the sorting order of all the states;

[0213] Determine whether the sorting order is consistent with the sorting order of each state in the state combination;

[0214] Based on the result of the determination, determine whether the area where the photovoltaic power station is located is in a sandstorm weather.

[0215] Optionally, the first determination module 930 is specifically configured to:

[0216] If the result of the determination is that the sorting order is consistent with the sorting order of each state in the state combination, determine whether the states of all the power curve segments are all valid;

[0217] If so, determine that the area where the photovoltaic power station is located is in a sandstorm weather;

[0218] If not, determine that the area where the photovoltaic power station is located is not in a sandstorm weather.

[0219] Optionally, the first determination module 930 is further specifically configured to:

[0220] If all the states determined by the comparison are inconsistent with the state combination or the sorting order is inconsistent with the sorting order of each state in the state combination, identify whether there is a necessary state in the state combination among the states of all the determined power curve segments;

[0221] If there is, calculate the total numerical identifier of the states of all the power curve segments based on the corresponding relationship between the preset state and the numerical identifier;

[0222] Match the total numerical identifier with the numerical identifier range of the occurrence of the sandstorm;

[0223] If the total numerical identifier is within the first numerical identifier range of the occurrence of the sandstorm, determine that a sandstorm weather has occurred in the area where the photovoltaic power station is located;

[0224] If the total numerical identifier is within the second numerical identifier range of the occurrence of the sandstorm, determine that a sandstorm weather has already occurred in the area where the photovoltaic power station is located, where the boundary value of the first numerical identifier range is less than the boundary value of the second numerical identifier range.

[0225] In summary, by converting real-time power generation data into line graphs at fixed time intervals and performing ordered matching based on the shapes of these line graphs, such a detection method can not only ensure the accuracy of weather detection, but also has a relatively simple matching process, and at the same time can ensure the integrity of weather detection.

[0226] As Figure 9 shown, an embodiment of a cleaning device for a photovoltaic module provided by an embodiment of the present invention, the device includes:

[0227] A second acquisition module 1010, configured to acquire real-time power generation data of a photovoltaic power station, where the real-time power generation data includes a plurality of continuously generated power data segments with equal time intervals;

[0228] A second abstraction module 1020, configured to perform abstraction conversion processing on each of the generated power data segments to obtain corresponding line graphs;

[0229] A second determination module 1030, configured to match each of the line graphs with each state graph recorded in a preset state matching library, and determine whether the area where the photovoltaic power station is located is in a sandstorm weather based on the matching result;

[0230] A cleaning module 1040, configured to determine whether to clean the photovoltaic module based on the real-time power generation data and corresponding reference power generation data when it is determined that the area where the photovoltaic power station is located is in a sandstorm weather.

[0231] In this embodiment, the cleaning module 1040 is specifically configured to:

[0232] Calculate the ratio of the real-time power generation data to the corresponding reference power generation data;

[0233] If the ratio is less than a threshold value, clean the photovoltaic module;

[0234] If the ratio is not less than the threshold value, do not clean the photovoltaic module.

[0235] This embodiment acquires real-time power generation data of a photovoltaic power station, where the real-time power generation data includes a plurality of continuously generated power data segments with equal time intervals; performs abstraction conversion processing on each of the generated power data segments to obtain corresponding line graphs; matches each of the line graphs with each state graph recorded in a preset state matching library, and determines whether the area where the photovoltaic power station is located is in a sandstorm weather based on the matching result; if so, determines whether to clean the photovoltaic module based on the real-time power generation data and corresponding reference power generation data. This solves the problem that due to the complex and inaccurate weather detection scheme of the photovoltaic power station, the credibility of dust accumulation detection of the photovoltaic module is low and the power generation efficiency is low.

[0236] Figure 10 It is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. The electronic device 1100 may vary greatly due to different configurations or performances, and may include one or more processors (central processing units, CPUs) 1110 and a memory 1120, and one or more storage media 1130 (such as one or more mass storage devices) for storing application programs 1133 or data 1132. Among them, the memory 1120 and the storage medium 1130 may be transient storage or persistent storage. The program stored in the storage medium 1130 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations on the electronic device 1100. Further, the processor 1110 may be configured to communicate with the storage medium 1130 and execute a series of instruction operations in the storage medium 1130 on the electronic device 1100 to implement the method provided in the above embodiment.

[0237] The electronic device 1100 may further include one or more power supplies 1140, one or more wired or wireless network interfaces 1150, one or more input / output interfaces 1160, and / or one or more operating systems 1131, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art can understand that Figure 10 the shown structure of the electronic device does not limit the electronic device, and it may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0238] The present invention also provides a computer-readable storage medium. The computer-readable storage medium may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions run on a computer, the computer is caused to execute the steps of the weather detection method provided in the above embodiment, or execute the steps of the cleaning method of the photovoltaic module provided in the above embodiment.

[0239] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of power generation of the photovoltaic power generation system and maximum power point tracking control in the system described above can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0240] In addition, in the description of the embodiments of the present invention, unless otherwise clearly specified and limited, the terms "installation", "connection", and "coupling" shall be understood in a broad sense. For example, it may be a fixed connection, a detachable connection, or an integral connection; it may be a mechanical connection or an electrical connection; it may be a direct connection or an indirect connection through an intermediate medium, and it may be the communication inside two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0241] If the above-mentioned functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing an electronic device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs.

[0242] In the description of the present invention, it should be noted that the orientation or positional relationship indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying 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. In addition, the terms "first", "second", and "third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0243] Finally, it should be noted that the above embodiments are only specific embodiments of the present invention, used to illustrate the technical solutions of the present invention, rather than limiting them. The protection scope of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: any person skilled in the art within the technical scope disclosed by the present invention can still modify the technical solutions recorded in the foregoing embodiments, or can easily think of changes, or perform equivalent replacements on some of the technical features; and these modifications, changes, or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A weather detection method, characterized in that, The method includes: Obtaining real-time power generation data of a photovoltaic power station, where the real-time power generation data includes multiple power generation data segments with equal and continuous time intervals; Performing abstract transformation processing on each of the power generation data segments to obtain corresponding line graphs; Matching each of the line graphs with the state graphs recorded in a preset state matching library, and determining whether the area where the photovoltaic power station is located is in a sandstorm weather based on the matching result.

2. The weather detection method according to claim 1, characterized in that, The performing abstract transformation processing on each of the power generation data segments to obtain corresponding line graphs includes: Generating corresponding power curve segments based on each of the power generation data segments; Abstracting each of the power curve segments into N target points, and forming a corresponding line graph with N - 1 line segments drawn by the N target points, where N is greater than or equal to 3.

3. The weather detection method according to claim 2, wherein The abstracting each of the power curve segments into N target points, and forming a corresponding line graph with N - 1 line segments drawn by the N target points includes: Evenly dividing each of the power curve segments into N equal parts, and sorting the power data in each part; Selecting the middle part of the power data in each part after sorting, and calculating the average power value to obtain N target points of the corresponding power curve segment; Connecting the N target points in chronological order to generate a line graph of the corresponding power curve segment.

4. The weather detection method according to claim 3, wherein The abstracting each of the power curve segments into N target points, and forming a corresponding line graph with N - 1 line segments drawn by the N target points further includes: Calculating the coordinate information of each target point in the line graph based on the average power value of each part; Calculating the angle between adjacent two line segments in the line graph based on the coordinate information of each target point, as well as the slopes of the N - 1 line segments in the line graph and the average slope of the line graph.

5. The weather detection method according to claim 4, wherein The matching each of the line graphs with the state graphs recorded in a preset state matching library, and determining whether the area where the photovoltaic power station is located is in a sandstorm weather based on the matching result includes: Determining the shape of each of the line graphs based on the angle and / or average slope of each of the line graphs, and comparing the shape of the line graph with the shapes of the state graphs recorded in the preset state matching library to obtain a comparison result; Determining the state of each of the power curve segments based on the comparison result, and determining whether the area where the photovoltaic power station is located is in a sandstorm weather based on the state of each of the power curve segments.

6. The weather detection method according to claim 5, characterized in that, The determining the shape of each of the line graphs based on the angle and / or average slope of each of the line graphs includes: Judging whether the angle of each of the line graphs is greater than an angle threshold; If so, determining the shape of each of the line graphs based on the average slope; If not, calculating the height difference between each line segment in each of the line graphs, and determining the shape of the corresponding line graph based on the height difference between each line segment.

7. The weather detection method according to claim 6, characterized in that The determining the shape of each of the line graphs based on the average slope includes: Judging whether the average slope is less than an average slope threshold; If so, determining the shape of the corresponding line graph as the first shape; If not, determining the shape of the corresponding line graph as the third shape.

8. The weather detection method according to claim 6, characterized in that, The determining the shape of the corresponding line graph based on the height difference between each line segment includes: Determine whether there is a slope greater than a preset threshold among the slopes of N - 1 line segments in the line graph; If so, then determine the magnitude relationship between the heights of two adjacent front and rear line segments among the N - 1 line segments to obtain a judgment result; If the judgment result is that the height of the previous line segment is less than the height of the next line segment, then determine that the shape of the corresponding line graph is the second shape; If the judgment result is that the height of the previous line segment is greater than the height of the next line segment, then determine that the shape of the corresponding line graph is the fourth shape.

9. The weather detection method according to claim 7 or 8, characterized in that The comparing the shape of the line graph with the shapes of each state graph recorded in the preset state matching library to obtain a comparison result includes: When the shape of the line graph is the first shape, then compare the first shape with the shapes of each state graph recorded in the preset state matching library to obtain that the state of the corresponding line graph is the first state or the fifth state; When the shape of the line graph is the second shape, then compare the second shape with the shapes of each state graph recorded in the preset state matching library to obtain that the state of the corresponding line graph is the second state; When the shape of the line graph is the third shape, then compare the third shape with the shapes of each state graph recorded in the preset state matching library to obtain that the state of the corresponding line graph is the third state; When the shape of the line graph is the fourth shape, then compare the fourth shape with the shapes of each state graph recorded in the preset state matching library to obtain that the state of the corresponding line graph is the fourth state.

10. The weather detection method according to claim 9, wherein After determining the states of each of the power curve segments based on the comparison result, it further includes: Determine whether there is a state with a higher ranking among the states of the power curve segments; If there is, then determine whether the flag bit of the state with a higher ranking is valid; If it is valid, then modify the flag bit of the state of the power curve segment to be valid; If it is invalid, then the state of the power curve segment is marked after waiting for the flag bit of the state with a higher ranking to be modified to be valid.

11. The weather detection method according to claim 10, characterized in that, The determining whether the area where the photovoltaic power station is located is in a sandstorm weather based on the states of each of the power curve segments includes: Compare all the determined states of the power curve segments with the state combination of the sandstorm weather; If all the states determined by the comparison are consistent with the state combination, sort the corresponding states based on the occurrence time of each of the power curve segments to obtain the sorting order of all the states; Determine whether the sorting order is consistent with the sorting order of each state in the state combination; Determine whether the area where the photovoltaic power station is located is in a sandstorm weather based on the judgment result.

12. The weather detection method according to claim 11, wherein The determining whether the area where the photovoltaic power station is located is in a sandstorm weather based on the judgment result includes: If the judgment result is that the sorting order is consistent with the sorting order of each state in the state combination, then determine whether all the states of the power curve segments are valid; If so, then determine that the area where the photovoltaic power station is located is in a sandstorm weather; If not, then determine that the area where the photovoltaic power station is located is not in a sandstorm weather.

13. The weather detection method according to claim 11, wherein, The determining whether the area where the photovoltaic power station is located is in a sandstorm weather based on the states of each of the power curve segments further includes: If all the determined states are inconsistent with the state combination or it is determined that the sorting order is inconsistent with the sorting order of each state in the state combination, identify whether there is a necessary state in the state combination among all the determined states of the power curve segments; If there is, calculate the sum of the numerical identifiers of all the states of the power curve segments based on the correspondence between the preset states and the numerical identifiers; Match the sum of the numerical identifiers with the numerical identifier range of the occurrence of sandstorms; If the sum of the numerical identifiers is within the first numerical identifier range of the occurrence of sandstorms, determine that a sandstorm weather has occurred in the area where the photovoltaic power station is located; If the sum of the numerical identifiers is within the second numerical identifier range of the occurrence of sandstorms, determine that a sandstorm weather has already occurred in the area where the photovoltaic power station is located, where the boundary value of the first numerical identifier range is less than the boundary value of the second numerical identifier range.

14. A cleaning method for a photovoltaic module, characterized in that, The method includes: Obtain the real-time power generation data of the photovoltaic power station, where the real-time power generation data includes multiple power generation data segments with equal and continuous time intervals; Perform an abstract transformation process on each of the power generation data segments to obtain corresponding line segment graphics; Match each of the line segment graphics with the state graphics recorded in the preset state matching library, and determine whether the area where the photovoltaic power station is located is in a sandstorm weather based on the matching result; If so, determine whether to clean the photovoltaic modules based on the real-time power generation data and the corresponding reference power generation data.

15. The cleaning method of the photovoltaic module according to claim 14, wherein The determination of whether to clean the photovoltaic modules based on the real-time power generation data and the corresponding reference power generation data includes: Calculate the ratio of the real-time power generation data to the corresponding reference power generation data; If the ratio is less than the threshold, clean the photovoltaic modules; If the ratio is not less than the threshold, do not clean the photovoltaic modules.

16. A weather detection device, characterized in that, The device includes: A first acquisition module, configured to acquire the real-time power generation data of the photovoltaic substation, where the real-time power generation data includes multiple power generation data segments with equal and continuous time intervals; A first abstraction module, configured to perform an abstract transformation process on each of the power generation data segments to obtain corresponding line segment graphics; A first determination module, configured to match each of the line segment graphics with the state graphics recorded in the preset state matching library, and determine whether the area where the photovoltaic power station is located is in a sandstorm weather based on the matching result.

17. A cleaning device for a photovoltaic module, characterized in that, The device includes: A second acquisition module, configured to acquire the real-time power generation data of the photovoltaic power station, where the real-time power generation data includes multiple power generation data segments with equal and continuous time intervals; A second abstraction module, configured to perform an abstract transformation process on each of the power generation data segments to obtain corresponding line segment graphics; A second determination module, configured to match each of the line segment graphics with the state graphics recorded in the preset state matching library, and determine whether the area where the photovoltaic power station is located is in a sandstorm weather based on the matching result; A cleaning module, configured to determine whether to clean the photovoltaic modules based on the real-time power generation data and the corresponding reference power generation data when it is determined that the area where the photovoltaic power station is located is in a sandstorm weather.

18. An electronic device, characterized in that, The electronic device includes: a memory and at least one processor, and instructions are stored in the memory; The at least one processor invokes the instructions in the memory, so that the electronic device executes the weather detection method according to any one of claims 1-13, or the cleaning method of the photovoltaic module according to claim 14 or 15.

19. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the weather detection method according to any one of claims 1-13, or the cleaning method of the photovoltaic module according to claim 14 or 15.

Images