An automatic tracking photovoltaic system for a track panel

Abstract

The application discloses an automatic tracking photovoltaic system for track plates and relates to the technical field of photovoltaic technology.The system comprises a tracking support set obtaining module, a reliability evaluation module, a driven shaft determining module and a tracking regulation scheme obtaining module.The tracking support set obtaining module is used to lay photovoltaic plates on a target ballastless track, and K shafts and K tracking support sets connected by the K shafts are obtained after the laying is completed.The reliability evaluation module is used to obtain a K tracking reliability evaluation result set.The driven shaft determining module is used to determine a main shaft and K-1 driven shafts from the K shafts.The tracking regulation scheme obtaining module is used to obtain a main shaft tracking regulation scheme.The automatic tracking regulation scheme obtaining module is used to obtain K-1 driven shaft tracking regulation schemes for regulating the K-1 driven shafts, and an automatic tracking regulation scheme is obtained after the schemes are summarized.The application solves the technical problem that the photovoltaic automatic tracking cannot achieve the expectation and the tracking regulation accuracy is low in the prior art, and achieves the technical effect of improving the tracking regulation quality.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic technology, and more specifically to an automatic tracking photovoltaic system for track slabs. Background Technology

[0002] The extensive use of ballastless track in the field not only results in low land utilization but also poses a significant challenge to the track slabs, as prolonged exposure to sunlight can lead to buckling deformation and interlayer separation due to temperature variations. This compromises the safety and stability of the track system during operation. Photovoltaic systems, which convert solar energy into electricity using solar panels, can improve land and space utilization by utilizing the space provided by ballastless track. Furthermore, the absorption of solar energy by the photovoltaic system reduces the amount of sunlight reaching the track slabs, thus mitigating buckling deformation. However, the power generation efficiency of photovoltaic systems is affected by factors such as solar radiation intensity, duration of sunlight exposure, and the installation angle of the photovoltaic modules. Currently, traditional fixed-support photovoltaic systems cannot dynamically adjust to changes in the sun's position, limiting their power generation efficiency.

[0003] Existing technologies suffer from technical problems such as the inability to achieve expected results during automatic photovoltaic tracking and low tracking and control accuracy. Summary of the Invention

[0004] This application provides an automatic tracking photovoltaic system for track slabs, which addresses the technical problems of low tracking and control accuracy in existing automatic photovoltaic tracking systems, which fail to achieve the expected results.

[0005] In view of the above problems, this application provides an automatic tracking photovoltaic system for track slabs, the system comprising:

[0006] The tracking support set acquisition module is used to lay photovoltaic panels on the target ballastless track, and to acquire the K completed rotating shafts and the K tracking support sets connected by the K rotating shafts respectively, where K is a positive integer;

[0007] The reliability evaluation module is used to obtain historical synchronous tracking data and corresponding timestamps of K tracking bracket sets to evaluate the tracking reliability of each tracking bracket and obtain K tracking reliability evaluation result sets.

[0008] The driven shaft determination module is used to determine the master shaft and K-1 driven shafts among the K shafts based on the K sets of tracking reliability evaluation results;

[0009] The tracking and control scheme acquisition module is used to collect the current solar position information, combine it with the information captured by the sensor components deployed on each tracking support connected to the main shaft, perform tracking adjustment analysis, and obtain the main shaft tracking and control scheme.

[0010] The automatic tracking regulation scheme obtaining module is configured to perform same-axis collaborative analysis on the K tracking reliability evaluation result sets and the main shaft tracking regulation scheme, obtain K-1 slave shaft tracking regulation schemes for regulating the K-1 slave shafts, and perform summarization on the K-1 slave shaft tracking regulation schemes and the main shaft tracking regulation scheme to obtain an automatic tracking regulation scheme.

[0011] One or more technical solutions provided in the present application have at least the following technical effects or advantages:

[0012] The present application provides one or more technical solutions provided in the present application have at least the following technical effects or advantages: BRIEF DESCRIPTION OF DRAWINGS

[0013] Figure 1 A structure diagram of an automatic tracking photovoltaic system for a track plate is provided for the embodiments of the present application;

[0014] Figure 2 A flowchart for obtaining K tracking reliability evaluation results in an automatic tracking photovoltaic system for a track plate is provided for the embodiments of the present application;

[0015] The reference signs are explained as follows: a tracking support set obtaining module 11, a reliability evaluation module 12, a slave shaft determining module 13, a tracking regulation scheme obtaining module 14, and an automatic tracking regulation scheme obtaining module 15. DETAILED DESCRIPTION

[0016] The present application provides an automatic tracking photovoltaic system for a track plate, which is used to solve the technical problem that the photovoltaic automatic tracking cannot achieve the expected tracking regulation accuracy in the prior art.

[0017] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0018] It should be noted that the terms "comprising" and "having" are intended to cover the inclusion of not exclusive inclusion, for example, a process, system, system, product or server including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or modules that are not clearly listed or inherent to these processes, systems, products or devices.

[0019] As shown in the embodiments, the present application provides an automatic tracking photovoltaic system for track slab, wherein the system comprises: Figure 1

[0020] The tracking bracket set obtaining module 11 is used for photovoltaic panel laying on the target ballastless track, and K rotating shafts and K tracking bracket sets connected by the K rotating shafts respectively after laying are obtained, wherein K is a positive integer.

[0021] In one possible embodiment, the target ballastless track is any track on which photovoltaic panel laying and photovoltaic tracking are needed. The specific process of laying the tracking bracket and the rotating shaft is as follows: it is assumed that the target ballastless track is long in the track direction (i.e. the longitudinal direction) and wide in the transverse direction, so that the following parameters are set: the length of the track slab is a, the width of the track slab is b, the length of the photovoltaic panel is c, the width of the photovoltaic panel is d, the distance between the inner sides of the left and right rails is e, the height from the rail top to the upper surface of the track slab is f, the longitudinal distance between the photovoltaic panels is g, the safety distance from the top surface of the photovoltaic panel to the top surface of the rail is h, the safety distance between the photovoltaic panel and the left and right rails is i, the length of the sleeper is j, the transverse distance between the sleepers is k, and the installation height of the photovoltaic panel is L. Wherein a, b, c, d, e, f, g, h, i, j, k and L are positive integers set by a person skilled in the art according to the actual situation.

[0022] Wherein the length c of the photovoltaic panel on the outer side of the rail should be less than or equal to 2k+2j-g, and the width d of the photovoltaic panel on the outer side of the rail can be arbitrarily selected, but should maintain a safety distance i with the rail. Each photovoltaic tracking bracket is installed on the symmetry axis of the transversely adjacent two sleepers in a "one every two sleepers" manner, and the hinge base installation position of the photovoltaic tracking bracket is at a distance A from the outer edge of the track slab, wherein A is a positive integer. Wherein the "one every two sleepers" means that one photovoltaic tracking bracket is installed every two sleepers, so that the center distance between the adjacent two photovoltaic tracking brackets is 2j+2k, and the number of photovoltaic tracking brackets required is the number of sleepers / 2.

[0023] ​The selection of the width d of the photovoltaic panel between the left and right rails should be less than or equal to e-2i, and the width of the hinge base of the photovoltaic tracking support should not exceed the distance between the left and right sleepers. Each photovoltaic tracking support is installed on the symmetry axis between the left and right sleepers in the longitudinal direction, and the center distance between adjacent two photovoltaic supports is c+g. The photovoltaic tracking supports are uniformly arranged at this distance, and the number of photovoltaic tracking supports required is a / (c+g), thereby realizing the laying of the photovoltaic tracking supports on the target ballastless track. Further, the photovoltaic panel is installed on the top of the tracking support, and the sensor assembly is installed on the tracking support, thereby completing the laying of the photovoltaic panel.

[0024] The rotating shaft is a mechanical assembly for driving a plurality of tracking supports to rotate, which is usually laid at a certain interval along the track direction. Each rotating shaft is connected to a group of photovoltaic supports through a connecting structure, forming K tracking support sets. After the photovoltaic assembly is laid on the target ballastless track, the laying structure is identified and grouped, K independent rotating shafts and K tracking support sets connected thereto are determined, and each rotating shaft and a group of tracking supports form a relatively independent control unit, which lays the foundation for subsequent regulation and control process.

[0025] The reliability evaluation module 12 is used for obtaining K tracking support set historical synchronous tracking data and corresponding time stamp, performing tracking reliability evaluation on each tracking support, and obtaining K tracking reliability evaluation result set.

[0026] Further, as shown in Figure 2 The reliability evaluation module 12 is used for performing the following steps:

[0027] The K historical synchronous tracking data group sets of the K tracking support sets are compared with the corresponding historical automatic tracking regulation and control scheme, and K historical synchronous tracking deviation data group sets are obtained. The K historical synchronous tracking deviation data group sets are heuristically screened to determine K first historical synchronous tracking deviation coefficient sets. The K historical synchronous tracking deviation data group sets are analyzed based on the time stamp to determine K second historical synchronous tracking deviation coefficient sets. The K first historical synchronous tracking deviation coefficient sets and the K second historical synchronous tracking deviation coefficient sets are subjected to tracking reliability evaluation to obtain the K tracking reliability evaluation results.

[0028] In one embodiment, the K set of historical synchronous tracking data groups refers to the tracking execution data recorded by the K set of tracking supports in the same time period, including parameters such as rotation angle, execution time, response delay, energy consumption, error value, etc. The historical automatic tracking regulation scheme is data describing the situation that the tracking support needs to be adjusted in the historical time. The time stamp is used to identify the specific time point of each data collection, for time series data modeling and trend analysis. The tracking reliability evaluation refers to the quantitative evaluation of the ability of each support to stably track the sun's trajectory in the past operation, thereby outputting a set of reliability scores. The K set of tracking reliability evaluation results, i.e., the historical performance score results corresponding to each set of tracking supports, will be used as the basis for subsequent master and slave shaft determination.

[0029] In one possible embodiment, by analyzing the historical tracking situation of the K set of tracking supports, the tracking reliability of each tracking support is determined, which is mainly to identify the tracking support regulation response reliability due to factors such as dust accumulation, rain and snow. The tracking angle in the K set of historical synchronous tracking data groups and the regulation angle in the corresponding historical automatic tracking regulation scheme are compared, and the difference between the two is used as the historical synchronous tracking deviation data, thereby obtaining the K set of historical synchronous tracking deviation data groups. Further, from the two dimensions of determining the generally representative data deviation degree and deviation data development trend, the K set of historical synchronous tracking deviation data groups is heuristically filtered and iteratively analyzed in time sequence, thereby obtaining the corresponding K set of first historical synchronous tracking deviation coefficients and K set of second historical synchronous tracking deviation coefficients. Each first historical synchronous tracking deviation coefficient reflects the general deviation of the corresponding historical synchronous tracking deviation data group. Each second historical synchronous tracking deviation coefficient set reflects the deviation trend of the corresponding historical synchronous tracking deviation data group.

[0030] The pre-constructed tracking reliability evaluation network layer is called to evaluate the tracking reliability of the K set of first historical synchronous tracking deviation coefficient sets and the K set of second historical synchronous tracking deviation coefficient sets, and the K set of tracking reliability evaluation results is obtained. Preferably, a plurality of sample first historical synchronous tracking deviation coefficient sets, a plurality of sample second historical synchronous tracking deviation coefficient sets, and a plurality of sample tracking reliability evaluation results corresponding thereto are obtained as training data. The framework based on the feedforward neural network is supervised trained using the training data, and the network parameters of the framework are adjusted and updated during the training until the training converges, and the trained tracking reliability evaluation network layer is obtained. By obtaining the K set of tracking reliability evaluation results, the technical effect of providing a basis for subsequent master shaft and slave shaft differentiation is achieved.

[0031] Further, the reliability evaluation module 12 is configured to perform the following steps:

[0032] randomly selecting one data from each of the K sets of historical synchronous tracking deviation data as a K set of heuristic data; constructing a heuristic neighborhood in the K sets of historical synchronous tracking deviation data according to a preset radius, taking the K set of heuristic data as a screening starting point, to obtain a K set of heuristic neighborhoods; determining a heuristic direction based on the K set of heuristic neighborhoods, iterating the K set of heuristic data according to the K heuristic direction, determining a K set of iterated heuristic data, and iterating for multiple times until a difference between neighborhood densities of adjacent two iterations is less than or equal to a preset neighborhood density difference threshold, to obtain a K set of target heuristic data; and comparing the K set of target heuristic data with a preset deviation tolerance threshold, to obtain the K set of first historical synchronous tracking deviation coefficients.

[0033] Further, the reliability evaluation module 12 is configured to perform the following steps:

[0034] searching for a neighborhood dense center in the K set of heuristic neighborhoods based on a preset neighborhood center search function, to determine a K set of neighborhood dense centers; and taking a direction from the K set of heuristic data to the K set of neighborhood dense centers as a K set of heuristic directions.

[0035] In an embodiment of the present application, the set of historical synchronous tracking deviation data refers to a set of deviation data records between an actual angle and a set target angle of each tracking support in a historical control process. The K set of heuristic data is an initial sample randomly extracted from the K set of historical synchronous tracking deviation data, used to guide a subsequent data clustering and optimization search process. The heuristic neighborhood refers to a data subset constructed within a preset radius with the heuristic sample as a center, representing a similar group of the initial sample in a local space. The preset radius is a distance that can be divided into the same neighborhood by a person skilled in the art in advance, wherein the distance can be an Euclidean distance between the historical synchronous tracking deviation data and the heuristic data corresponding to the screening starting point.

[0036] The neighborhood density is obtained by respectively counting the amount of data contained in the K set of heuristic neighborhoods, and comparing the statistical result with 2 times of the preset radius. The preset neighborhood density difference threshold is a maximum difference between heuristic neighborhood densities after adjacent two iterations, set by a person skilled in the art when stopping iteration, which is a condition for controlling iteration convergence. The heuristic direction is a vector direction from the current heuristic data set to the neighborhood dense center, guiding iteration update. The K set of target heuristic data is a representative deviation data set obtained after iteration optimization. The first set of historical synchronous tracking deviation coefficients is a reliability performance index of the tracking support in a spatial deviation dimension.

[0037] Preferably, the preset neighborhood center search function is used to search the neighborhood dense center in the K heuristic neighborhood set, to determine the more representative and more densely distributed historical synchronous tracking deviation data in each heuristic neighborhood, and to obtain the K neighborhood dense center set. Then, the direction of the K heuristic data set to the K neighborhood dense center set is obtained as the K heuristic direction set.

[0038] Preferably, the preset neighborhood center search function is: wherein, is the neighborhood dense center, is the Gaussian kernel function, is the i-th historical synchronous tracking deviation data in the heuristic field, is the heuristic data corresponding to the screening starting point.

[0039] In one embodiment, after determining the K heuristic directions, the K heuristic data set is moved by the preset radius according to the K heuristic directions to obtain K iteration heuristic data sets, and K iteration heuristics are constructed based on the same construction principle of the K heuristic neighborhood set. It is judged whether the neighborhood density difference between the K iteration heuristic neighborhoods and the K heuristic neighborhoods is less than or equal to the preset neighborhood density difference threshold. If not, the K iteration heuristic directions are determined based on the same principle as determining the K heuristic directions, and the K iteration heuristic data sets are iterated until the neighborhood density difference between the adjacent two iterations is less than or equal to the preset neighborhood density difference threshold. The data obtained by the last iteration is taken as the K target heuristic data set.

[0040] Preferably, the preset deviation tolerance threshold is a tolerance interval of data deviation in the corresponding historical automatic tracking control scheme preset by a person skilled in the art. The K target heuristic data set is compared with the preset deviation tolerance threshold to obtain the K first historical synchronous tracking deviation coefficient set. In this way, by analyzing the density of the neighborhood, calling the neighborhood center search function, determining the dense center in each neighborhood, and constructing the heuristic direction set, the initial heuristic point is moved along the direction to realize the next iteration. After each iteration, it is judged whether the neighborhood density changes before and after the two times. If the density is less than the set threshold, it is considered to be converged, and the final target heuristic data set is formed. The technical effect of improving the data processing quality and providing high-quality data for determining the reliability of the tracking support is achieved.

[0041] Further, the reliability evaluation module 12 is configured to perform the following steps:

[0042] activating a plurality of adaptive encoders to respectively extract adaptive data trend features from each of the K sets of historical synchronization tracking deviation data groups in chronological order from early to late, to obtain K sets of deviation trend feature groups; integrating the K sets of deviation trend feature groups respectively to determine K sets of fusion deviation trend features; scoring the K sets of fusion deviation trend features to obtain the K sets of second historical synchronization tracking deviation coefficients.

[0043] Further, the reliability evaluation module 12 is configured to perform the following steps:

[0044] extracting two deviation trend feature groups from the K sets of deviation trend feature groups randomly to perform similarity comparison, and determining a comparison similarity; when the comparison similarity is greater than or equal to a preset similarity threshold, associating the two deviation trend features to obtain K sets of associated deviation trend feature groups and K sets of isolated deviation trend feature groups; calculating feature similarity of any two associated deviation trend features with an association identifier in the K sets of associated deviation trend feature groups, and performing normalization processing on the calculation result to obtain an association matrix; using a convolutional network to calculate convolution analysis results of the two associated deviation trend features and the association matrix respectively to obtain corresponding two associated fusion deviation trend features, and constructing K sets of associated fusion deviation trend feature groups; calculating the mean of the K sets of associated fusion deviation trend feature groups and the K sets of isolated deviation trend feature groups to obtain the K sets of fusion deviation trend features.

[0045] In one possible embodiment, the adaptive encoder is a neural network structure capable of dynamically extracting features according to the change law of the data itself, including different data processing scales, and is commonly used to process trend changes and periodic fluctuations in time series data. Using the plurality of adaptive encoders to perform convolution analysis of the K sets of historical synchronization tracking deviation data groups in different scales can extract deviation trends of the tracking stent's regulatory response reliability over time. The K sets of deviation trend feature groups are used to reflect the deviation trend features of the K tracking stents in the time dimension. The fusion deviation trend feature set is a comprehensive representation of the deviation trend feature groups belonging to the same tracking stent after fusion, which facilitates subsequent unified scoring. By quantitatively evaluating the fused trend features, the K sets of second historical synchronization tracking deviation coefficients are generated, which are used to measure the stability of the long-term dynamic behavior of the stent.

[0046] Preferably, a plurality of sample synchronous tracking deviation data sets are obtained, and after analysis according to a convolution scale of an adaptive encoder, a plurality of sample deviation trend feature sets are obtained. The plurality of sample synchronous tracking deviation data sets and the plurality of sample deviation trend feature sets are used to supervise training of a framework constructed based on a feedforward neural network until training converges, and a trained adaptive encoder is obtained. Based on the same construction principle, a plurality of trained adaptive encoders are obtained.

[0047] In an embodiment, each of the K sets of historical synchronous tracking deviation data is sorted in chronological order from front to back to obtain a set of K sequences of historical synchronous tracking deviation data. Convolution feature extraction is performed on the set of K sequences of historical synchronous tracking deviation data using the plurality of adaptive encoders to obtain a set of K deviation trend feature groups. Then, feature integration is performed on each of the deviation trend feature groups to obtain a set of K fused deviation trend features. A second set of K historical synchronous tracking deviation coefficients corresponding to the set of K fused deviation trend features is obtained by performing deviation scoring based on the set of K fused deviation trend features and a preset feature-deviation scoring table constructed by a person skilled in the art according to a mapping relationship between historical trend features and corresponding historical deviation scores.

[0048] In an embodiment, the process of integrating the set of K deviation trend feature groups is as follows: two deviation trend feature groups are randomly extracted from the set of K deviation trend feature groups for similarity comparison, the similarity between the two deviation trend feature groups is analyzed using a cosine similarity calculation formula to obtain a comparison similarity, and when the comparison similarity is greater than or equal to a preset similarity threshold (the minimum similarity when features are associated according to a preset of a person skilled in the art), the two deviation trend features are associated and labeled to obtain a set of K associated deviation trend feature groups. Features in the set of K deviation trend feature groups that are not labeled with an association label are taken as a set of K isolated deviation trend feature groups.

[0049] Preferably, for any two associated deviation trend features in the set of K associated deviation trend feature groups with an association label, a cosine similarity calculation formula is used to calculate the feature similarity, and the calculation result is normalized, and optionally, the normalized result is added to an initially empty matrix to obtain an association matrix. Then, convolution analysis results of the two associated deviation trend features and the association matrix are calculated using a convolutional network to enhance the features of the two associated deviation trend features, and two associated fused deviation trend features corresponding to the two associated deviation trend features are obtained to construct a set of K associated fused deviation trend feature groups.

[0050] Preferably, a plurality of correlation deviation trend features and corresponding correlation matrices, and a plurality of correlation fusion deviation trend features after enhanced correlation analysis are obtained as training data, and the convolution network is supervised training until the training converges. Further, an associated correlation deviation trend feature and a correlation matrix are input into the convolution network, and the corresponding correlation fusion deviation trend feature is output. Further, the K correlation fusion deviation trend feature group sets and the corresponding K isolated deviation trend feature group sets are calculated by mean value, and the K fusion deviation trend feature sets are obtained.

[0051] By obtaining a second historical synchronous tracking deviation coefficient set, that is, a coefficient for representing the deviation stability of each group of tracking supports in the time dimension, and summarizing it with the spatial deviation coefficient, the technical effect of providing double-dimensional support for subsequent master-slave rotating shaft selection is achieved.

[0052] The driven rotating shaft determination module 13 is used to determine the master rotating shaft and K-1 driven rotating shafts in the K rotating shafts based on the K tracking reliability evaluation result sets.

[0053] Further, the driven rotating shaft determination module 13 is used to perform the following steps:

[0054] The difference between the maximum value and the minimum value in the K tracking reliability evaluation result sets is calculated to determine K first evaluation factors; the fluctuation variance of the K tracking reliability evaluation result sets is calculated to obtain K second evaluation factors; the K first evaluation factors and the K second evaluation factors are analyzed by weighting to obtain K rotating shaft evaluation factors; the rotating shaft corresponding to the maximum value in the K rotating shaft evaluation factors is taken as the master rotating shaft, and the remaining rotating shafts are taken as the K-1 driven rotating shafts.

[0055] In one possible embodiment, the master rotating shaft and K-1 driven rotating shafts are determined by screening the K rotating shafts according to the obtained K tracking reliability evaluation result sets, that is, the operation stability scores of the tracking supports corresponding to each rotating shaft. The first evaluation factor refers to the operation performance distinction degree obtained from the range (maximum value-minimum value) of the rotating shaft evaluation value, and the second evaluation factor refers to the statistical fluctuation degree (variance) calculated based on the reliability score, reflecting the consistency of the operation stability of the rotating shaft. The weighted analysis is to fuse the two factors into a comprehensive rotating shaft evaluation factor, which is used for the final master-slave division. Finally, the rotating shaft with the highest score is selected as the master rotating shaft, and the remaining rotating shafts are taken as the driven rotating shafts to participate in the collaborative tracking execution.

[0056] The master rotating shaft is the rotating shaft with the optimal comprehensive stability and the highest reliability, which is used to undertake the core tracking control responsibility. The remaining K-1 rotating shafts are taken as the driven rotating shafts, which will be adjusted collaboratively under the guidance of the master rotating shaft driving scheme. Thus, the technical effect of laying a foundation for improving the reliability of photovoltaic tracking is achieved.

[0057] The tracking regulation scheme obtaining module 14 is configured to collect the sun position information at the current time, combine the information captured by the sensor assembly arranged on each tracking support connected to the main rotating shaft, perform tracking adjustment analysis, and obtain the main rotating shaft tracking regulation scheme.

[0058] The automatic tracking regulation scheme obtaining module 15 is configured to combine the K tracking reliability evaluation result sets and the main rotating shaft tracking regulation scheme to perform the same rotating shaft collaborative analysis, obtain the K-1 driven rotating shaft tracking regulation schemes for regulating the K-1 driven rotating shafts, and combine the K-1 driven rotating shaft tracking regulation schemes and the main rotating shaft tracking regulation scheme to obtain the automatic tracking regulation scheme.

[0059] Further, the automatic tracking regulation scheme obtaining module 15 is configured to perform the following steps:

[0060] The minimum values of the K-1 tracking reliability evaluation result sets except the tracking reliability evaluation result set corresponding to the main rotating shaft are extracted from the K tracking reliability evaluation result sets to obtain K-1 minimum tracking reliability evaluation results. The K-1 minimum tracking reliability evaluation results and the main rotating shaft tracking regulation scheme are identified by the regulation scheme collaborator to obtain the K-1 driven rotating shaft tracking regulation schemes.

[0061] In one possible embodiment, the main rotating shaft tracking regulation scheme is combined with the driven rotating shaft regulation logic to realize the synchronous and efficient adjustment of the K tracking support sets. The regulation scheme collaborator is a decision unit responsible for formulating the corresponding following strategy for each driven rotating shaft according to the action plan of the main rotating shaft and the operating capacity difference of the driven rotating shaft. The minimum tracking reliability evaluation result refers to the selection of the evaluation result with the lowest value from the tracking support set corresponding to the driven rotating shaft, which represents the potential “weak response” unit and is used to set the most conservative adjustment scheme boundary. The finally output automatic tracking regulation scheme is the tracking adjustment strategy after the control instructions of the main rotating shaft and the K-1 driven rotating shafts are integrated.

[0062] Preferably, the main rotating shaft is excluded from the K tracking reliability evaluation results to obtain the K-1 tracking reliability evaluation result sets for the K-1 driven rotating shafts. The minimum values in the K-1 tracking reliability evaluation result sets are extracted to form the K-1 minimum tracking reliability evaluation results. This step ensures that each set of regulation strategies can cover the tracking support under the weakest performance.

[0063] The control scheme coordinator is called to match and analyze the master shaft tracking control scheme and each minimum tracking reliability evaluation result, and to identify the control tolerance range and synchronization mode of each driven shaft. For example, the master shaft plans to rotate 12°, but a driven shaft responds slowly and can only tolerate an adjustment range of 8°. The driven shaft is set to adjust in small steps or introduce a compensation mechanism (such as starting early or delaying response) through the control scheme coordinator. The coordinator forms K-1 driven shaft tracking control schemes through this personalized control strategy. Preferably, a plurality of sample master shaft tracking control schemes, a plurality of sample minimum tracking reliability evaluation results, and a plurality of sample driven shaft tracking control schemes corresponding thereto are obtained as training data for the coordinator, the training data is divided into a training set and a validation set according to a predetermined ratio (usually 3:2), the framework based on a convolutional neural network is supervised trained using the training set, the plurality of sample master shaft tracking control schemes and the plurality of sample minimum tracking reliability evaluation results in the validation set are input into the framework, and a plurality of validation driven shaft tracking control schemes are obtained as output. The number of identical quantities between the plurality of validation driven shaft tracking control schemes and the plurality of sample driven shaft tracking control schemes in the validation set is counted. When the statistical result is less than a predetermined required number, the network parameters of the framework are updated until the validation is passed, and the trained control scheme coordinator is obtained.

[0064] By uniformly collecting all control schemes, a complete and directly executable automatic tracking control scheme is output, ensuring that multiple shafts work efficiently and do not interfere with each other when working together, thereby improving the tracking stability and response accuracy of the overall system.

[0065] In summary, the embodiments of the present application have at least the following technical effects:

[0066] 1. The present application integrates the control strategy of the master shaft and the performance level of the driven shaft through the control scheme coordinator, generates K-1 differentiated driven control schemes, and combines them with the master scheme to generate a complete automatic tracking control instruction, effectively realizing multi-axis synchronous control of the system and improving tracking accuracy and response efficiency.

[0067] 2. The present application comprehensively extracts the spatial offset features and time response trends of the tracking support through the combination of heuristic clustering, neighborhood density determination, adaptive encoder, and trend similarity convolution modeling, realizes deep and multi-angle evaluation of the running state, and provides high-reliability data support for subsequent control strategies.

[0068] It should be noted that the above-mentioned embodiment sequences of the present application are merely for description only, but not for representing the advantages and disadvantages of the embodiments. And the above-mentioned embodiments of the present specification have been described. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or can be advantageous.

[0069] The above only describes the preferred embodiments of the present application, and does not limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

[0070] The specification and drawings are merely exemplary of the present application, and any and all modifications, variations, combinations or equivalents that are within the scope of the present application should be included. Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. An automatically tracking photovoltaic system for a track panel, characterized by, The system comprises: The tracking support set obtaining module is used for laying photovoltaic panels on the target ballastless track, and obtaining K rotating shafts laid and K tracking support sets connected through the K rotating shafts respectively, wherein K is a positive integer; The reliability evaluation module is used for obtaining historical synchronous tracking data and corresponding time stamps of the K tracking support sets, performing tracking reliability evaluation on each tracking support, and obtaining a K tracking reliability evaluation result set; The driven rotating shaft determining module is used for determining a master rotating shaft and K-1 driven rotating shafts from the K rotating shafts based on the K tracking reliability evaluation result set; The tracking regulation scheme obtaining module is used for collecting solar position information at a current time, combining information captured by sensor assemblies arranged on each tracking support connected with the master rotating shaft, performing tracking adjustment analysis, and obtaining a master rotating shaft tracking regulation scheme; The automatic tracking regulation scheme obtaining module is used for performing same-rotating-shaft collaborative analysis on the K tracking reliability evaluation result set and the master rotating shaft tracking regulation scheme, obtaining K-1 driven rotating shaft tracking regulation schemes for regulating the K-1 driven rotating shafts, and combining the K-1 driven rotating shaft tracking regulation schemes and the master rotating shaft tracking regulation scheme to obtain an automatic tracking regulation scheme; The reliability evaluation module is used for performing the following steps: The K historical synchronous tracking data group sets of the K tracking support sets are compared with corresponding historical automatic tracking regulation schemes to obtain K historical synchronous tracking deviation data group sets; The K historical synchronous tracking deviation data group sets are heuristically screened to determine K first historical synchronous tracking deviation coefficient sets; The K historical synchronous tracking deviation data group sets are analyzed in terms of forward and backward time sequence iteration trends based on time stamps to determine K second historical synchronous tracking deviation coefficient sets; The K first historical synchronous tracking deviation coefficient sets and the K second historical synchronous tracking deviation coefficient sets are subjected to tracking reliability evaluation to obtain the K tracking reliability evaluation results; The driven rotating shaft determining module is used for performing the following steps: The difference between the maximum value and the minimum value in the K tracking reliability evaluation result set is calculated to determine K first evaluation factors; The fluctuation variance of the K tracking reliability evaluation result set is calculated to obtain K second evaluation factors; The K first evaluation factors and the K second evaluation factors are subjected to weighted analysis to obtain K rotating shaft evaluation factors; The rotating shaft corresponding to the maximum value in the K rotating shaft evaluation factors is taken as the master rotating shaft, and the remaining rotating shafts are taken as the K-1 driven rotating shafts.

2. An automatically tracking photovoltaic system for a track panel as defined in claim 1, characterized in that The reliability evaluation module is used for performing the following steps: One data is randomly selected from each of the K historical synchronous tracking deviation data group sets as a K heuristic data set; The K heuristic data sets are taken as screening starting points, and heuristic neighborhoods are constructed in the K historical synchronous tracking deviation data group sets according to a preset radius to obtain a K heuristic neighborhood set; Based on the K heuristic neighborhood set, heuristic direction determination is performed, and the K heuristic data sets are iterated according to K heuristic directions to determine K iterative heuristic data sets. After multiple iterations, the difference between the neighborhood densities of two adjacent iterations is less than or equal to a preset neighborhood density difference threshold, and K target heuristic data sets are obtained. The K target heuristic data sets are compared with a preset deviation tolerance threshold to obtain the K first historical synchronization tracking deviation coefficient set.

3. An automatically tracking photovoltaic system for a track panel as defined in claim 2, wherein, The reliability evaluation module is configured to perform the following steps: Based on a preset neighborhood center search function, neighborhood dense center search is performed in the K heuristic neighborhood set to determine a K neighborhood dense center set. The K heuristic directions are respectively set as K heuristic direction sets.

4. An automatically tracking photovoltaic system for a track panel as defined in claim 1, wherein, The reliability evaluation module is configured to perform the following steps: A plurality of adaptive encoders are activated to respectively perform adaptive data trend feature extraction on each historical synchronization tracking deviation data group in the K historical synchronization tracking deviation data group set in chronological order from the front to the back to obtain a K deviation trend feature group set. The K deviation trend feature group sets are respectively integrated to determine a K fusion deviation trend feature set. The K fusion deviation trend feature sets are scored for deviation to obtain the K second historical synchronization tracking deviation coefficient set.

5. An automatically tracking photovoltaic system for a track panel as defined in claim 4, wherein, The reliability evaluation module is configured to perform the following steps: Two deviation trend feature groups are randomly extracted from the K deviation trend feature group sets for similarity comparison to determine a comparison similarity. When the comparison similarity is greater than or equal to a preset similarity threshold, the two deviation trend features are associated and identified to obtain a K associated deviation trend feature group set and a K isolated deviation trend feature group set. Feature similarity calculation is performed on any two associated deviation trend features with an associated identifier in the K associated deviation trend feature group set, and the calculation result is normalized to obtain an association matrix. Convolutional network is used to calculate the convolutional analysis results of the two associated deviation trend features and the association matrix respectively to obtain corresponding two associated fusion deviation trend features, and a K associated fusion deviation trend feature group set is constructed. The mean values of the K associated fusion deviation trend feature group set and the K isolated deviation trend feature group set are calculated to obtain the K fusion deviation trend feature set.

6. An automatically tracking photovoltaic system for a track panel as defined in claim 1, wherein, The automatic tracking control scheme obtaining module is configured to perform the following steps: From the K tracking reliability evaluation result sets, the minimum value of the tracking reliability evaluation result set corresponding to the main rotating shaft is extracted to obtain K-1 minimum tracking reliability evaluation results. The K-1 minimum tracking reliability evaluation results and the main rotating shaft tracking control scheme are identified by a control scheme coordinator to obtain K-1 driven rotating shaft tracking control schemes.

Images