Photovoltaic cleaning robot path planning method fusing group string level power generation analysis

Abstract

The application discloses a photovoltaic cleaning robot path planning method fusing group string level power generation analysis, and belongs to the technical field of data processing, and specifically comprises the following steps: constructing an entropy value model by dividing historical group string power generation data according to the same interval, collecting data in real time to calculate interval entropy values and comparing the interval entropy values with the model to identify suspected low power generation intervals; combining spatial neighborhood information to filter key cleaning group strings, calculating posterior probability through a Bayesian probability model to determine cleaning priorities; combining power station layout constraints and robot motion characteristics to generate an optimal cleaning path by using a path planning algorithm; and through multi-dimensional data fusion and intelligent algorithms, the application realizes accurate identification of photovoltaic group string cleaning requirements and path optimization, effectively improves cleaning efficiency and reduces operation and maintenance costs.

Description

A path planning method for photovoltaic cleaning robots that integrates string-level power generation analysis Technical Field

[0001] This invention belongs to the field of data processing technology, specifically a path planning method for photovoltaic cleaning robots that integrates string-level power generation analysis. Background Technology

[0002] With the rapid development of photovoltaic (PV) power generation technology, the scale of PV power plants is constantly expanding. During the operation of PV power plants, dust and dirt on the surface of PV panels can reduce their power generation efficiency, thus requiring regular cleaning. Currently, using PV cleaning robots to clean PV panels has become a common practice. However, existing PV cleaning robot path planning methods lack precise analysis of the power generation of PV strings, leading to unreasonable cleaning path planning, low cleaning efficiency, and high cleaning costs. For example, some methods do not fully consider the differences in power generation among different PV strings, cleaning all PV strings uniformly, resulting in a waste of cleaning resources; other methods, while considering power generation, lack detailed analysis and cannot accurately identify the PV strings that truly require focused cleaning, thus affecting the cleaning effect and the overall power generation efficiency of the power plant. Therefore, there is an urgent need for a method that can accurately analyze the power generation of PV strings and rationally plan the path of the cleaning robot accordingly. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention proposes a path planning method for photovoltaic cleaning robots that integrates string-level power generation analysis. It constructs an entropy model by dividing historical string power generation data into identical intervals, collects data in real time to calculate interval entropy values, and compares these values ​​with the model to identify potentially low-power intervals. It then uses spatial neighborhood information to screen key cleaning strings and calculates posterior probabilities using a Bayesian probability model to determine cleaning priorities. Finally, it combines power plant layout constraints with robot motion characteristics to generate the optimal cleaning path using a path planning algorithm. This invention, through multi-dimensional data fusion and intelligent algorithms, achieves accurate identification and path optimization of photovoltaic string cleaning needs, effectively improving cleaning efficiency and reducing operation and maintenance costs.

[0004] To achieve the above objectives, the present invention provides the following technical solution:

[0005] A path planning method for photovoltaic cleaning robots that integrates string-level power generation analysis includes:

[0006] Step S1: Divide the pre-acquired photovoltaic string power generation key value data set into intervals, calculate the interval entropy value, and construct a string-level power generation entropy value model based on the interval entropy value statistics.

[0007] Step S2: In real time, acquire the set of key-value pairs of photovoltaic string power generation data at each moment within a preset time period in the photovoltaic power station, and perform interval division and real-time entropy value calculation for the interval;

[0008] Step S3: Compare the real-time entropy value of each interval with the string-level power generation entropy value model to obtain the suspected low power generation intervals;

[0009] Step S4: For each suspected low-power generation interval, analyze its spatial neighborhood information, and use the entropy change trend of the neighborhood interval to determine the key cleaning string and mark the low-power generation interval.

[0010] Step S5: Construct a Bayesian probability model, take the entropy value of the key cleaning string in real time as input, calculate the posterior probability, and combine it with the probability threshold to obtain the priority score of string cleaning.

[0011] Step S6: Based on the results of entropy analysis and Bayesian probability model, and combined with the layout structure of the photovoltaic power station and the motion characteristics of the cleaning robot, a path planning algorithm is used to plan the cleaning path of the cleaning robot.

[0012] Specifically, step S1 includes the following steps:

[0013] S1.1: Obtain the high-order data set of string power generation at each moment within a preset time period from the photovoltaic strings that have been generating power normally in the past, and perform normalization processing to obtain the photovoltaic string power generation key-value pair data set; each data in the high-order data set of string power generation includes string number, timestamp and corresponding power generation value;

[0014] S1.2: Divide the key-value pair dataset of photovoltaic string power generation into N 8×8 intervals;

[0015] S1.3: Calculate the histogram of power generation distribution for each interval, P = {p k,1 ,...,p k,a} and calculate the entropy value of each interval based on the power generation distribution histogram. Where, p k,i Let w represent the probability that the power generation value i appears in the k-th interval, where i∈[1,a], k∈[1,N], a represents the power generation level, and w represents the probability that the power generation value i appears in the k-th interval. k,i δ represents the weighting coefficient of the power generation value i in the k-th interval, and δ represents the smoothing parameter;

[0016] S1.4: Set the entropy value abnormal thresholds h1 and h2, where h1 represents the minimum abnormal entropy value and h2 represents the maximum abnormal entropy value;

[0017] If E k k If the interval is greater than h2, then the interval is removed.​

[0018] If h1≤E k If ≤h2, then based on the entropy value E of each interval k Calculate the mean μ and standard deviation σ of the entropy values ​​for all intervals;

[0019] S1.5: Construct a string-level power generation entropy model based on the mean μ and standard deviation σ of all interval entropy values, satisfying Y=[μ-3σ,μ+3σ], where Y represents the range of the string-level power generation entropy model.

[0020] Specifically, step S3 includes the following steps:

[0021] S3.1: Obtain the real-time entropy value E of each interval in the dataset of photovoltaic string power generation key-value pairs within the preset time period calculated in step S2. k ′;

[0022] S3.2: Obtain the range Y of the string-level power generation entropy model;

[0023] S3.3: For each interval, determine its real-time entropy value E. k Is it within the range Y of the string-level power generation entropy value model?

[0024] If E k If ′∈[μ-3σ,μ+3σ], then the photovoltaic string power generation in the corresponding interval is normal;

[0025] like The interval is then determined to be a suspected low power generation interval, meaning that the power generation of the photovoltaic string corresponding to the interval is affected by dust and dirt.

[0026] S3.4: Based on the interval index value (z) row ,z col ), calculate the position coordinates (x, y) of the photovoltaic string corresponding to the suspected low power generation range. row ×8,z col ×8), where z row and z col These represent the x-coordinate and y-coordinate index values ​​of the interval, respectively;

[0027] S3.5: Record the location coordinates (x, y) of the suspected low power generation zone and its corresponding photovoltaic string.

[0028] Specifically, step S4 includes the following steps:

[0029] S4.1: Obtain suspected low-power interval patches m Given the position coordinates (x, y), extract the neighborhood interval patch′ obtained by expanding the neighborhood of the position coordinates (x, y) by 3×3.m And calculate the patch for suspected low power generation areas. m entropy value and neighborhood interval patch' m entropy value Where m represents the number of suspected low-power zones;

[0030] S4.2: Based on neighborhood interval patch′ m entropy value The mean entropy value μ of the neighborhood interval is obtained by summing and taking the mean. lin The entropy standard deviation σ is obtained using the standard deviation calculation formula. lin ;

[0031] S4.3: Through calculation and The absolute value of the difference is used to obtain the entropy difference between the suspected low-power generation interval and the neighboring interval.

[0032] S4.4: Set the entropy difference threshold h3;

[0033] like The photovoltaic strings corresponding to the suspected low power generation range are identified as key cleaning strings and marked, and their location information (x′, y′) is recorded.

[0034] like The photovoltaic strings corresponding to the suspected low power generation range are then determined to be affected by non-cleaning factors and excluded from the cleaning scope.

[0035] Specifically, step S5 includes the following steps:

[0036] S5.1: Obtain the real-time entropy value of the key cleaning string. And r ≤ m, where r represents the number of key cleaning strings in real time;

[0037] S5.2: Define a Bayesian probability model. Based on the historical photovoltaic (PV) string power generation key-value pair data set, obtain the prior probability P(X) of power generation reduction in the historical PV string power generation key-value pair data set, and the prior probability P(X′) of power generation change caused by non-cleaning factors in the historical PV string power generation key-value pair data set. Fit the probability distribution of entropy value under the condition of power generation reduction in the historical PV string power generation key-value pair data set. The probability distribution of entropy values ​​under conditions where power generation changes due to non-cleaning factors. Where X represents the data in the historical photovoltaic string power generation key-value pair dataset that shows a decrease in power generation, X′ represents the data in the historical photovoltaic string power generation key-value pair dataset that shows a change in power generation caused by non-cleaning factors, and r′ represents the number of data in the historical photovoltaic string power generation key-value pair dataset that shows a decrease in power generation. This indicates that the historical photovoltaic string power generation key value reduces the entropy value of the power generation data in the dataset. This represents the entropy value of the key-value pair data on the historical photovoltaic string power generation, which is affected by changes in power generation due to non-cleaning factors.

[0038] Specifically, step S5 further includes the following steps:

[0039] S5.3: Probability distribution of entropy values ​​under conditions of reduced power generation in the dataset based on fitted historical photovoltaic string power generation key pairs The probability distribution of entropy values ​​under conditions where power generation changes due to non-cleaning factors. The entropy value of the key cleaning string will be measured in real time. Substitute as input and The calculation formula calculates the probability distribution of entropy values ​​for real-time photovoltaic string power generation key values ​​in the dataset under the condition of reduced power generation. The probability distribution of entropy values ​​under conditions where power generation changes due to non-cleaning factors.

[0040] S5.4: Through P(X) and The product of, plus P(X′) and The product of these factors yields the marginal probability of the entropy value in the dataset for the real-time photovoltaic string power generation key-value pair.

[0041] Specifically, step S5 further includes the following steps:

[0042] S5.5: Through P(X) and The marginal probability of the product divided by the entropy value Obtain the posterior probability

[0043] S5.6: Set the probability threshold h4;

[0044] like The key cleaning string is determined to be affected by changes in power generation caused by factors other than cleaning.

[0045] like Then, the key cleaning string is determined to be affected by the change in power generation caused by the cleaning factor, and the posterior probability is set. As a priority score for string cleaning.

[0046] Specifically, in S5.2, the probability distribution of entropy values ​​under the condition of reduced power generation in the historical photovoltaic string power generation key-value pair data set is fitted. The probability distribution of entropy values ​​under conditions where power generation changes due to non-cleaning factors. The specific process includes:

[0047] S5.21: Reduce the entropy of the historical photovoltaic string power generation key-value pairs in the dataset. The mean μ of the data entropy value of the decrease in power generation in the historical photovoltaic string power generation key-value pair dataset. X The standard deviation σ of the historical photovoltaic string power generation key value to the data entropy value of the power generation reduction in the dataset. X Substituting into the formula for the Gaussian distribution function, we fit the probability distribution of entropy values ​​under the condition of reduced power generation in the historical photovoltaic string power generation key-value pair dataset.

[0048] S5.22: Extract the entropy value of the power generation change data caused by non-cleaning factors from the historical photovoltaic string power generation key-value pair dataset. The mean μ of the entropy values ​​of the historical photovoltaic string power generation key-value pair dataset, which are affected by non-cleaning factors in power generation changes. X The standard deviation σ of the entropy values ​​of historical photovoltaic string power generation key-value pairs in the data set, which are affected by changes in power generation due to non-cleaning factors. X′ Substituting into the formula for the Gaussian distribution function, we fit the probability distribution of entropy values ​​in the historical photovoltaic string power generation key-value pair dataset under the condition that power generation changes are caused by non-cleaning factors.

[0049] Specifically, step S6 includes the following steps:

[0050] S6.1: Obtain the set of all strings with abnormal entropy values, and obtain the set of all strings with posterior probabilities exceeding the probability threshold from the Bayesian probability model. Take the intersection of the two sets to generate a preliminary cleaned candidate string set.

[0051] S6.2: Based on the string number, map its physical coordinates in the photovoltaic array and extract obstacle information from the photovoltaic power plant layout database;

[0052] S6.3: Bayesian posterior probabilities The entropy result and the attenuation term based on the previous cleaning time are weighted and summed to obtain the cleaning priority; the entropy result is the difference between 1 and the real-time entropy; the real-time entropy is the ratio of the first variable to the second variable; the first variable is the difference between the real-time entropy of string b and the abnormal minimum entropy value; the second variable is the difference between the abnormal maximum entropy value and the abnormal minimum entropy value.

[0053] S6.4: Normalize the cleaning priority to the [0,1] interval and divide it into three levels; set the threshold nodes in the division process to 0.4 and 0.7;

[0054] S6.5: Combining the photovoltaic array layout constraints, a path planning algorithm is used to plan the cleaning path of the cleaning robot, and the robot is controlled to perform the cleaning task according to the cleaning path.

[0055] Specifically, the photovoltaic string power generation key-value pair data set in step S2 includes multiple photovoltaic string power generation key-value pairs at different times, and each power generation key-value pair is a data pair of string number and power generation value.

[0056] Compared with the prior art, the beneficial effects of the present invention are:

[0057] 1. This invention proposes a path planning method for photovoltaic cleaning robots that integrates string-level power generation analysis. It quantifies the degree of power generation anomalies through a string-level power generation entropy model and eliminates environmental interference by combining spatial neighborhood information, thus avoiding misjudgments of dust and equipment failures in traditional methods. The Bayesian probability model further distinguishes between cleanable and non-cleanable factors, generating a list of cleaning strings with priorities, making the cleaning targets more targeted, reducing ineffective operations, and improving resource utilization.

[0058] 2. This invention proposes a path planning method for photovoltaic cleaning robots that integrates string-level power generation analysis. Based on photovoltaic array layout constraints, such as row and column structure and obstacles, a distance matrix is ​​constructed. Combined with priority design heuristic functions, a path planning algorithm is used to generate cleaning paths that balance efficiency and energy consumption. A dynamic feedback mechanism supports real-time replanning during the cleaning process, adapting to environmental changes and ensuring that the cleaning robot completes the task with the shortest path and lowest energy consumption, thereby improving the intelligence level of photovoltaic power plant operation and maintenance. Attached Figure Description

[0059] Figure 1 is a schematic diagram of the photovoltaic cleaning robot path planning method that integrates string-level power generation analysis according to the present invention.

[0060] Figure 2 is a flowchart illustrating the principle of the photovoltaic cleaning robot path planning method based on the cascade power generation analysis of the present invention.

[0061] Figure 3 is a flowchart of the string-level power generation entropy model construction in the photovoltaic cleaning robot path planning method of the present invention, which integrates string-level power generation analysis. Detailed Implementation

[0062] Example 1:

[0063] Please refer to Figures 1-3. One embodiment of the present invention provides a path planning method for a photovoltaic cleaning robot that integrates string-level power generation analysis, comprising the following steps:

[0064] Step S1: Divide the pre-acquired photovoltaic string power generation key value data set into intervals, calculate the interval entropy value, and construct a string-level power generation entropy value model based on the interval entropy value statistics.

[0065] It should be noted that the photovoltaic string power generation key-value pair data set refers to the time-series data set of photovoltaic string power generation. It is time-series data, composed of the power generation values ​​of each photovoltaic string in the photovoltaic power plant, along with their corresponding timestamps and string identifiers. It belongs to structured numerical data. In the string power generation key-value pair, the key identifies the string's power generation data at a specific moment, typically composed of the string number and timestamp. The value is the power generation value of that string at the corresponding moment, collected in real-time through the photovoltaic power plant's smart combiner box, string-level inverter, or monitoring system.

[0066] Step S2: In real time, acquire the set of key-value pairs of photovoltaic string power generation at each moment within a preset time period in the photovoltaic power station, and perform interval division and real-time entropy value calculation for the interval; the set of key-value pairs of photovoltaic string power generation includes key-value pairs of power generation of multiple photovoltaic strings at different times, and each key-value pair of power generation is a data pair of string number and power generation value;

[0067] It is important to understand that in step S2, when calculating the real-time entropy value for each interval, the specific process and formula are the same as in step S1.

[0068] Step S3: Compare the real-time entropy value of each interval with the string-level power generation entropy value model, and obtain the suspected low power generation intervals based on the comparison results;

[0069] Step S4: For each suspected low-power generation interval, analyze its spatial neighborhood information, and use the entropy change trend of the neighborhood interval to determine the key cleaning string and mark the low-power generation interval.

[0070] Step S5: Construct a Bayesian probability model, take the entropy value of the key cleaning string in real time as input, calculate the posterior probability, and combine it with the probability threshold to obtain the priority score of string cleaning.

[0071] Step S6: Based on the results of entropy analysis and Bayesian probability model, and combined with the layout structure of the photovoltaic power station and the motion characteristics of the cleaning robot, a path planning algorithm is used to plan the cleaning path of the cleaning robot.

[0072] In summary, this invention achieves power generation reduction detection by analyzing the entropy of the photovoltaic string power generation key-value pair data set, combining spatial neighborhood information, and using a Bayesian probability model method. Moreover, all entropy calculation formulas in this application refer to the entropy calculation formula in S1.3.

[0073] The specific steps of step S1 include:

[0074] S1.1: Obtain a high-order data set of the power generation of the photovoltaic strings at each moment within a preset time period from the photovoltaic strings that have been generating power normally in the past, and perform normalization processing to obtain a key-value pair data set of photovoltaic string power generation; each data in the key-value pair data set of photovoltaic string power generation includes the string number, timestamp and corresponding power generation value;

[0075] S1.2: Divide the key-value pair dataset of photovoltaic string power generation into N 8×8 intervals, and Among them, A row A represents the number of string power generation data sets along the string number dimension. col This represents the quantity of the string power generation data set in the time dimension. Indicates rounding down;

[0076] S1.3: Calculate the histogram of power generation distribution for each interval, P = {p k,1 ,...,p k,a} and calculate the entropy value of each interval based on the power generation distribution histogram. Where, p k,i Let w represent the probability that the power generation value i appears in the k-th interval, where i∈[1,a], k∈[1,N], a represents the power generation level, and w represents the probability that the power generation value i appears in the k-th interval. k,i δ represents the weighting coefficient of the power generation value i in the k-th interval, and δ represents the smoothing parameter;

[0077] in, f k,i This represents the number of times the power generation value i appears in the k-th interval.

[0078] max(f k,1 ,...,f k,a ) represents the frequency of the highest power generation value in the interval.

[0079] It should be noted that in actual calculations, there may be a situation where the probability of the power generation value i appearing in the k-th interval is 0, i.e., p k,i =0, resulting in log(p) calculation k,i The case of p reaching infinity was addressed by adding a smoothing parameter δ. k,i The problem of abnormal logarithmic calculation when = 0 has been addressed, improving the stability of the calculation. At the same time, the weighting coefficient can enhance the contribution of high-frequency power generation values ​​and suppress the noise influence of low-frequency power generation values, making the entropy calculation more consistent with the distribution characteristics of power generation reduction in actual scenarios. In summary, the modified formula can better adapt to complex environments.

[0080] S1.4: Set the entropy value abnormal thresholds h1 and h2, where h1 represents the minimum abnormal entropy value and h2 represents the maximum abnormal entropy value;

[0081] If E k k If the value is greater than h2, then the interval is removed.

[0082] If h1≤E k If ≤h2, then based on the entropy value E of each interval k Calculate the mean μ and standard deviation σ of the entropy values ​​for all intervals;

[0083] S1.5: Construct a string-level power generation entropy model based on the mean μ and standard deviation σ of all interval entropy values, satisfying Y=[μ-3σ,μ+3σ], where Y represents the range of the string-level power generation entropy model.

[0084] It should be noted that the string-level power generation entropy model is a statistical model based on entropy values, used to describe the power distribution characteristics of string-level power generation in a key-value pair dataset. In the string-level power generation detection method based on entropy values, the range of the string-level power generation entropy model is Y = [μ-3σ, μ+3σ]. This is because the mean and standard deviation of the entropy values ​​are statistical results based on a large number of string-level power generation datasets, which can reflect the typical entropy distribution of string-level power generation. Through the mean and standard deviation, the concentration range and dispersion of the string-level power generation entropy values ​​can be determined. Moreover, the entropy values ​​of string-level power generation usually fluctuate within a certain range, while the entropy values ​​of power generation reduction often exceed this range. By defining a range Y, it is possible to simply and efficiently determine whether the entropy values ​​of a certain interval belong to the typical range of string-level power generation. Therefore, in actual detection, it is only necessary to compare the entropy values ​​of the real-time interval with the range Y of the string-level power generation entropy model to quickly determine whether it belongs to the string-level power generation value.

[0085] For example, assuming the size of the key-value pair data set of photovoltaic string power generation is 256×256, that is, the string number dimension and the time dimension are both 256, and the interval size is 8×8, then the number of intervals N=1024, and the number of power generation levels k=256;

[0086] (1) Calculate the entropy value:

[0087] 1) Calculate the power generation distribution histogram P for each interval;

[0088] 2) Calculate the entropy value E for each interval according to the entropy formula. k ;

[0089] (2) Calculate the mean μ and standard deviation σ of the entropy values ​​for all intervals;

[0090] (3) Constructing a string-level power generation entropy model:​

[0091] Assuming μ = 5.2 and σ = 0.8, the range Y of the string-level power generation entropy model is: [5.2 - 3 × 0.8, 5.2 + 3 × 0.8] = [2.8, 7.6].

[0092] The specific steps of step S3 include:

[0093] S3.1: Obtain the real-time entropy value E of each interval in the dataset of photovoltaic string power generation key-value pairs within the preset time period calculated in step S2. k ′;

[0094] S3.2: Obtain the range Y of the string-level power generation entropy model;

[0095] S3.3: For each interval, determine its real-time entropy value E. k Is it within the range Y of the string-level power generation entropy value model?

[0096] If E k If ′∈[μ-3σ,μ+3σ], then the photovoltaic string power generation in the corresponding interval is determined to be normal;

[0097] like The interval is then determined to be a suspected low power generation interval, meaning that the power generation of the photovoltaic string corresponding to this interval is affected by dust and dirt.

[0098] S3.4: Based on the interval index value (z) row ,z col ), calculate the position coordinates (x, y) of the photovoltaic string corresponding to the suspected low power generation range. row ×8,z col ×8), where z row and z col These represent the x-coordinate and y-coordinate index values ​​of the interval, respectively;

[0099] S3.5: Record the location coordinates (x, y) of the suspected low power generation zone and its corresponding photovoltaic string.

[0100] The specific steps of step S4 include:

[0101] S4.1: Obtain suspected low-power interval patches m Given the position coordinates (x, y), extract the neighborhood interval patch′ obtained by expanding the neighborhood of the position coordinates (x, y) by 3×3. m And calculate the patch for suspected low power generation areas. m entropy value and neighborhood interval patch' m entropy value Where m represents the number of suspected low-power zones;

[0102] S4.2: Based on neighborhood interval patch′ m entropy value The mean entropy value μ of the neighborhood interval is obtained by summing and taking the mean. lin The entropy standard deviation σ is obtained using the standard deviation calculation formula. lin The formula for calculating the standard deviation is existing technology in this field and is not an inventive solution of this application, so it will not be described in detail here.

[0103] S4.3: Through calculation and The absolute value of the difference is used to obtain the entropy difference between the suspected low-power generation interval and the neighboring interval.

[0104] S4.4: Set the entropy difference threshold h3;

[0105] like The photovoltaic string corresponding to the suspected low power generation range is then identified as a key string for cleaning, marked, and its location information (x′, y′) is recorded.

[0106] like The photovoltaic string corresponding to the suspected low power generation range is then determined to be affected by non-cleaning factors and excluded from the cleaning scope.

[0107] The specific steps of step S5 include:

[0108] S5.1: Obtain the real-time entropy value of the key cleaning string. And r ≤ m, where r represents the number of key cleaning strings in real time;

[0109] S5.2: Define a Bayesian probability model, based on a set of historical photovoltaic string power generation key-value pairs, to obtain the prior probability of a decrease in power generation in the set of historical photovoltaic string power generation key-value pairs. The prior probability P(X′) = 1 - P(X) of the change in power generation caused by non-cleaning factors in the historical photovoltaic string power generation key-value pair dataset is determined, and the probability distribution of entropy values ​​under the condition of reduced power generation in the historical photovoltaic string power generation key-value pair dataset is fitted. The probability distribution of entropy values ​​under conditions where power generation changes due to non-cleaning factors. Where X represents the data in the historical photovoltaic string power generation key-value pair dataset that shows a decrease in power generation, X′ represents the data in the historical photovoltaic string power generation key-value pair dataset that shows a change in power generation caused by non-cleaning factors, and r′ represents the number of data in the historical photovoltaic string power generation key-value pair dataset that shows a decrease in power generation. This indicates that the historical photovoltaic string power generation key value reduces the entropy value of the power generation data in the dataset. This represents the entropy value of the historical photovoltaic string power generation key-value pair dataset, which is affected by changes in power generation due to non-cleaning factors.

[0110] S5.3: Probability distribution of entropy values ​​under conditions of reduced power generation in the dataset based on fitted historical photovoltaic string power generation key pairs The probability distribution of entropy values ​​under conditions where power generation changes due to non-cleaning factors. The entropy value of the key cleaning string will be measured in real time. Substitute as input and The calculation formula calculates the probability distribution of entropy values ​​for real-time photovoltaic string power generation key values ​​in the dataset under the condition of reduced power generation. The probability distribution of entropy values ​​under conditions where power generation changes due to non-cleaning factors.

[0111] S5.4: Through P(X) and The product of, plus P(X′) and The product of these factors yields the marginal probability of the entropy value in the dataset for the real-time photovoltaic string power generation key-value pair.

[0112] S5.5: Through P(X) and The marginal probability of the product divided by the entropy value Obtain the posterior probability

[0113] S5.6: Set the probability threshold h4;

[0114] like Then it is determined that the change in power generation of the key cleaning string is caused by non-cleaning factors;

[0115] like Then, the key cleaning string is determined to be affected by the change in power generation caused by the cleaning factor, and the posterior probability is set. This serves as the priority score for cleaning this string.

[0116] In S5.2, the probability distribution of entropy values ​​under the condition of reduced power generation in the dataset of historical photovoltaic string power generation key pairs is fitted. The probability distribution of entropy values ​​under conditions where power generation changes due to non-cleaning factors. The specific process includes:

[0117] S5.21: Reduce the entropy of the historical photovoltaic string power generation key-value pairs in the dataset. The mean μ of the data entropy value of the decrease in power generation in the historical photovoltaic string power generation key-value pair dataset. XThe standard deviation σ of the historical photovoltaic string power generation key value to the data entropy value of the power generation reduction in the dataset. X Substituting into the formula for the Gaussian distribution function, we fit the probability distribution of entropy values ​​under the condition of reduced power generation in the historical photovoltaic string power generation key-value pair dataset.

[0118] S5.22: Extract the entropy value of the power generation change data caused by non-cleaning factors from the historical photovoltaic string power generation key-value pair dataset. The mean μ of the entropy values ​​of the historical photovoltaic string power generation key-value pair dataset, which are affected by non-cleaning factors in power generation changes. X′ The standard deviation σ of the entropy values ​​of historical photovoltaic string power generation key-value pairs in the data set, which are affected by non-cleaning factors in power generation changes. X′ Substituting into the formula for the Gaussian distribution function, we fit the probability distribution of entropy values ​​in the historical photovoltaic string power generation key-value pair dataset under the condition that power generation changes are caused by non-cleaning factors.

[0119] Example 2:

[0120] In this embodiment, the specific steps of step S6 include:

[0121] S6.1: Obtain the set of all strings with abnormal entropy values, and obtain the set of all strings with posterior probabilities exceeding the probability threshold from the Bayesian probability model. Take the intersection of the two sets to generate a preliminary cleaned candidate string set.

[0122] S6.2: Based on the string number, map its physical coordinates in the photovoltaic array and extract obstacle information from the photovoltaic power plant layout database;

[0123] S6.3: Bayesian posterior probabilities The entropy result and the attenuation term based on the previous cleaning time are weighted and summed to obtain the cleaning priority; the entropy result is the difference between 1 and the real-time entropy; the real-time entropy is the ratio of the first variable to the second variable; the first variable is the difference between the real-time entropy of string b and the abnormal minimum entropy value; the second variable is the difference between the abnormal maximum entropy value and the abnormal minimum entropy value.

[0124] S6.4: Normalize the cleaning priority to the [0,1] interval and divide it into three levels; the threshold nodes in the division process are set to 0.4 and 0.7. The normalization method is the prior art in this field and is not an inventive solution of this application, so it will not be described in detail here.

[0125] S6.5: Combining the photovoltaic array layout constraints, a path planning algorithm is used to plan the cleaning path of the cleaning robot, and the robot is controlled to perform the cleaning task according to the cleaning path. The path planning algorithm is the prior art in this field and is not an inventive solution of this application, so it will not be described in detail here.

[0126] Furthermore, the photovoltaic array layout constraints are as follows:

[0127]

[0128] in, This represents the constraint distance from string b to string c. Let θ represent the Euclidean distance from string b to c, and let θ represent the weight coefficient of the layout constraint penalty. b,c represents the layout penalty term, and b and c represent the group string index.

[0129] Among them, for the layout penalty item LayoutPenalty b,c ,satisfy:

[0130] If the route from string b to c requires crossing obstacles, then LayoutPenalty b,c =C1, if the transition from string b to c does not require crossing any obstacles, then LayoutPenalty b,c =0, where C1 represents the obstacle constraint penalty value;

[0131] If the path from string b to c passes through an obstacle region, then LayoutPenalty b,c =C2, if the path from string b to c does not pass through the obstacle region, then LayoutPenalty b,c =0, where C2 represents the path constraint penalty value;

[0132] If the robot's turning cost needs to be considered when moving from string b to c, then LayoutPenalty b,c =η×Δψ, if the robot turning cost does not need to be considered when moving from string b to c, then LayoutPenalty b,c =0, where η represents the direction constraint penalty value and Δψ represents the steering angle.

[0133] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments under the guidance of the present invention without departing from the spirit and scope of the present invention. All of these variations are within the protection scope of the present invention.

Claims

1. A path planning method for a photovoltaic cleaning robot that integrates string-level power generation analysis, characterized in that, include: Step S1: Divide the pre-acquired photovoltaic string power generation key-value pair data set into intervals, calculate the interval entropy value, and construct a string-level power generation entropy value model based on the interval entropy value statistics; Step S2: Acquire the photovoltaic string power generation key-value pair data set at each moment within a preset time period in the photovoltaic power station in real time, and perform interval division and real-time interval entropy value calculation; Step S3: Compare the real-time entropy value of each interval with the string-level power generation entropy value model to obtain suspected low power generation intervals; Step S4: For each suspected low power generation interval, analyze its spatial neighborhood information, and use the entropy value change trend of the neighborhood interval to determine the key cleaning strings and mark the low power generation intervals; Step S5: Construct a Bayesian probability model to determine the key cleaning strings in real time. The entropy value of the string is used as input to calculate the posterior probability. Combined with the probability threshold, the priority score of string cleaning is obtained. Step S6: Based on the results of entropy analysis and Bayesian probability model, combined with the layout structure of the photovoltaic power station and the motion characteristics of the cleaning robot, a path planning algorithm is used to plan the cleaning path of the cleaning robot. The specific steps of step S1 include: S1.1: Obtain the high-order data set of string power generation at each moment within a preset time period from the photovoltaic strings that have been generating normally in the past, and perform normalization processing to obtain the photovoltaic string power generation key-value pair data set; each data in the high-order data set of string power generation contains the string number, timestamp and corresponding power generation value; S1.2: Divide the photovoltaic string power generation key-value pair data set into N groups. S1.3: Calculate the power generation distribution histogram for each interval; And calculate the entropy value of each interval based on the power generation distribution histogram. ,in, Let represent the probability that the power generation value i occurs in the k-th interval, and ， 'a' represents the power generation level. This represents the weighting coefficient of the power generation value i in the k-th interval. S1.4: Sets the entropy value anomaly threshold, representing the smoothing parameter. and ,in, This indicates an abnormally low entropy value. This indicates an abnormally high entropy value; if ,or If, then the interval is removed; if Then, based on the entropy value of each interval Calculate the mean of the entropy values ​​for all intervals. and standard deviation S1.5: Based on the mean of all interval entropy values and standard deviation Construct a string-level power generation entropy model, and satisfy... , where Y represents the range of the string-level power generation entropy model.

2. The photovoltaic cleaning robot path planning method based on integrated string-level power generation analysis as described in claim 1, characterized in that, The specific steps of step S3 include: S3.1: Obtaining the real-time entropy value of each interval in the dataset of photovoltaic string power generation key-value pairs at each moment within the preset time period calculated in step S2. S3.2: Obtain the range Y of the string-level power generation entropy model; S3.3: For each interval, determine its real-time entropy value. Is it within the range Y of the string-level power generation entropy model? If If so, the photovoltaic string power generation in the corresponding interval is considered normal; if If the interval is determined to be a suspected low-power generation interval, it means that the photovoltaic string power generation corresponding to the interval is affected by dust and dirt; S3.4: Based on the interval index value Calculate the location coordinates of the photovoltaic strings corresponding to the suspected low power generation range. ,in, and S3.5: Record the location coordinates of suspected low-power generation intervals and their corresponding photovoltaic strings. 。 3. The photovoltaic cleaning robot path planning method based on integrated string-level power generation analysis as described in claim 2, characterized in that, The specific steps of step S4 include: S4.1: Obtaining suspected low-power generation zones. Position coordinates Based on location coordinates Extract it Neighborhood intervals obtained after neighborhood expansion And calculate the suspected low power generation range. entropy value and neighborhood interval entropy value Where m represents the number of suspected low-power intervals; S4.2: Based on neighborhood intervals entropy value The mean entropy of the neighborhood interval is obtained by summing and averaging. The standard deviation of entropy is obtained using the standard deviation calculation formula. S4.3: Through calculation and The absolute value of the difference is used to obtain the entropy difference between the suspected low-power generation interval and the neighboring interval. S4.4: Set the entropy difference threshold ;like The photovoltaic strings corresponding to suspected low power generation areas are identified as key strings for cleaning, marked, and their location information is recorded. ;like If the photovoltaic strings corresponding to the suspected low power generation range are identified as being affected by non-cleaning factors, they will be excluded from the cleaning scope.

4. The photovoltaic cleaning robot path planning method based on integrated string-level power generation analysis as described in claim 3, characterized in that, The specific steps of step S5 include: S5.1: Obtain the real-time entropy value of the key cleaning string. ,and Where r represents the number of key cleaning strings in real time; S5.2: Define a Bayesian probability model, based on the historical photovoltaic string power generation key-value pair data set, to obtain the prior probability of power generation reduction in the historical photovoltaic string power generation key-value pair data set. Prior probability of changes in power generation caused by non-cleaning factors in the historical photovoltaic string power generation key-value pair dataset. And fit the probability distribution of entropy values ​​under the condition of reduced power generation in the historical photovoltaic string power generation key-value pair dataset. The probability distribution of entropy values ​​under conditions where power generation changes due to non-cleaning factors. Where X represents the historical photovoltaic string power generation key-value pair data on power generation reduction in the dataset. This represents the key-value pair data of historical photovoltaic (PV) string power generation, which shows changes in power generation caused by non-cleaning factors within the dataset. This indicates the number of data points in the dataset showing a decrease in power generation, representing the historical photovoltaic string power generation key-value pairs. This indicates that the historical photovoltaic string power generation key value reduces the entropy value of the power generation data in the dataset. This represents the entropy value of the key-value pair data on the historical photovoltaic string power generation, which is affected by changes in power generation due to non-cleaning factors.

5. The photovoltaic cleaning robot path planning method based on integrated string-level power generation analysis as described in claim 4, characterized in that, The specific steps of step S5 further include: S5.3: Based on the fitted historical photovoltaic string power generation key value pair, the probability distribution of entropy values ​​under the condition of reduced power generation in the data set. The probability distribution of entropy values ​​under conditions where power generation changes due to non-cleaning factors. The entropy value of the key cleaning string will be measured in real time. Substitute as input and The calculation formula calculates the probability distribution of entropy values ​​for real-time photovoltaic string power generation key values ​​in the dataset under the condition of reduced power generation. The probability distribution of entropy values ​​under conditions where power generation changes due to non-cleaning factors. S5.4: Through and The product, plus and The product of these factors yields the marginal probability of the entropy value in the dataset for the real-time photovoltaic string power generation key-value pair. 。 6. The photovoltaic cleaning robot path planning method based on integrated string-level power generation analysis as described in claim 5, characterized in that, The specific steps of step S5 further include: S5.5: through and The marginal probability of the product divided by the entropy value , obtain the posterior probability S5.6: Set probability threshold ;like If so, the key cleaning unit is determined to be affected by changes in power generation caused by factors other than cleaning; if If the key cleaning string is determined to be affected by the change in power generation caused by the cleaning factor, then the posterior probability will be... As a priority score for string cleaning.

7. The photovoltaic cleaning robot path planning method based on integrated string-level power generation analysis as described in claim 6, characterized in that, In step S5.2, the probability distribution of entropy values ​​under the condition of reduced power generation in the dataset of historical photovoltaic string power generation key pairs is fitted. The probability distribution of entropy values ​​under conditions where power generation changes due to non-cleaning factors. The specific process includes: S5.21: reducing the entropy of the power generation data in the historical photovoltaic string power generation key-value pair dataset. The mean of the decrease in data entropy values ​​of historical photovoltaic string power generation key-value pairs in the data set. The standard deviation of the historical photovoltaic string power generation key value relative to the data entropy value of the power generation reduction in the dataset. Substituting into the formula for the Gaussian distribution function, we fit the probability distribution of entropy values ​​under the condition of reduced power generation in the historical photovoltaic string power generation key-value pair dataset. S5.22: Extract the entropy value of the power generation change data caused by non-cleaning factors from the historical photovoltaic string power generation key-value pair dataset. The mean of the entropy values ​​of the historical photovoltaic string power generation key-value pairs in the dataset, which are affected by changes in power generation due to non-cleaning factors. The standard deviation of the entropy values ​​of historical photovoltaic string power generation key-value pairs in the dataset, which are affected by non-cleaning factors in power generation changes. Substituting into the formula for the Gaussian distribution function, we fit the probability distribution of entropy values ​​in the historical photovoltaic string power generation key-value pair dataset under the condition that power generation changes are caused by non-cleaning factors. 。 8. The photovoltaic cleaning robot path planning method based on integrated string-level power generation analysis as described in claim 7, characterized in that, The specific steps of step S6 include: S6.1: Obtaining a set of all strings with abnormal entropy values, and obtaining a set of all strings with posterior probabilities exceeding the probability threshold from the Bayesian probability model, taking the intersection of the two to generate a preliminary cleaned candidate string set; S6.2: Mapping the physical coordinates of the string in the photovoltaic array according to its string number, and extracting obstacle information from the photovoltaic power station layout database; S6.3: Calculating the Bayesian posterior probability... The cleaning priority is obtained by weighted summation of the entropy result and the attenuation term based on the previous cleaning time; the entropy result is the difference between 1 and the real-time entropy; the real-time entropy is the ratio of the first variable to the second variable; the first variable is the difference between the real-time entropy of string b and the abnormal minimum entropy value; the second variable is the difference between the abnormal maximum entropy and the abnormal minimum entropy; S6.4: The cleaning priority is normalized to the [0,1] interval and divided into three levels; S6.5: Combined with the photovoltaic array layout constraints, the cleaning path of the cleaning robot is planned using a path planning algorithm, and the robot is controlled to perform the cleaning task according to the cleaning path.

9. The photovoltaic cleaning robot path planning method based on integrated string-level power generation analysis as described in claim 8, characterized in that, The photovoltaic string power generation key-value pair data set in step S2 includes multiple photovoltaic strings power generation key-value pairs at different times, and each power generation key-value pair is a data pair of string number and power generation value.

Images