Self-adaptive tracking control method and system for photovoltaic support in complex terrain

Abstract

The invention relates to the technical field of photovoltaic support tracking control, and discloses a photovoltaic support adaptive tracking control method and system for a complex terrain, and the method comprises the steps: setting a plurality of terrain types, and constructing photovoltaic support tracking control models of all terrain types; obtaining real-time characteristic parameters of a plurality of photovoltaic supports, and obtaining a real-time tracking control strategy of each photovoltaic support according to the real-time characteristic parameters and the photovoltaic support tracking control model; according to the method, a feedback data packet of a real-time tracking control strategy of each photovoltaic support is generated, a real-time control evaluation value is calculated according to the feedback data packet, whether the real-time tracking control strategy is corrected or not is judged, the defects of the tracking control strategy are found and adjusted in time, efficient self-adaptive tracking is achieved, and the tracking efficiency is improved. And the power generation efficiency and reliability of the photovoltaic power generation system under complex terrain conditions are improved.

Description

Technical Field

[0001] This application relates to the field of photovoltaic support tracking control technology, and in particular to an adaptive tracking control method and system for photovoltaic supports in complex terrain. Background Technology

[0002] Traditional photovoltaic (PV) mounting systems are mostly fixed supports, and their angle cannot be adjusted after installation. As a result, PV modules cannot always maintain the optimal angle for receiving sunlight, which limits power generation efficiency. In order to improve power generation efficiency, tracking brackets have been proposed. Tracking brackets can automatically adjust the angle of PV modules according to changes in the position of the sun, so that the PV modules are always facing the sun as directly as possible, thereby significantly increasing PV power generation.

[0003] In existing technologies, photovoltaic support tracking control technology mainly adopts open-loop control or simple closed-loop control strategies. This approach does not consider or only considers the influence of a single actual environmental factor, resulting in low tracking accuracy in complex terrain and difficulty in achieving efficient adaptive tracking. This reduces the power generation efficiency and reliability of photovoltaic power generation systems under complex terrain conditions. Summary of the Invention

[0004] To address the aforementioned technical issues, this application provides an adaptive tracking control method and system for photovoltaic (PV) supports in complex terrain. By setting multiple terrain categories and constructing a PV support tracking control model, a real-time tracking control strategy is generated. By combining feedback data packets to calculate real-time control evaluation values, deficiencies in the tracking control strategy can be promptly identified and adjusted, achieving efficient adaptive tracking and improving the power generation efficiency and reliability of PV power generation systems under complex terrain conditions.

[0005] In some embodiments of this application, an adaptive tracking control method for photovoltaic supports in complex terrain is provided, including:

[0006] Several terrain categories are defined, and a photovoltaic support tracking control model is constructed for all terrain categories;

[0007] Obtain real-time characteristic parameters of several photovoltaic supports, and obtain the real-time tracking control strategy for each photovoltaic support based on the real-time characteristic parameters and the photovoltaic support tracking control model;

[0008] Feedback data packets for the real-time tracking control strategy of each photovoltaic support are generated. The real-time control evaluation value is calculated based on the feedback data packets, and it is determined whether the real-time tracking control strategy should be corrected.

[0009] In some embodiments of this application, several terrain categories are defined, including:

[0010] Multiple terrain feature indicators are set according to preset terrain parameters;

[0011] Multiple initial terrain categories are generated based on all terrain feature indicators;

[0012] Obtain historical control logs for photovoltaic supports with the same initial terrain category, and determine several environmental impact parameters for the same initial terrain category based on the historical control logs;

[0013] Multiple environmental characteristic indicators are set based on several environmental impact parameters of the same initial terrain category, and multiple environmental categories corresponding to the initial terrain category are generated based on all environmental characteristic indicators.

[0014] Randomly select an environment category for each initial terrain category and combine it with the corresponding initial behavior category to generate a terrain category;

[0015] Several terrain categories are generated sequentially.

[0016] In some embodiments of this application, a photovoltaic support tracking control model is constructed for all terrain types, including:

[0017] Obtain historical control logs for photovoltaic supports of each terrain type, and extract historical illumination data, historical environmental impact data, historical control parameters, and historical parameters to be evaluated from each historical control log;

[0018] The historical control parameters include historical adjustment angle, historical adjustment speed, and historical control duration.

[0019] Calculate the historical control evaluation value of the corresponding historical control parameter based on the historical parameters to be evaluated.

[0020] Pre-set control evaluation thresholds;

[0021] Training data sets corresponding to different terrain categories are constructed based on historical control parameters whose historical control evaluation values ​​are greater than the control evaluation value threshold. The training data sets include several parameter control groups.

[0022] Establish a tracking control sub-model corresponding to the terrain category based on the training data package;

[0023] Based on historical data packets, predictive risk values ​​for each terrain category are generated. Based on the predicted risk values, monitoring time intervals for the corresponding terrain categories are set. Based on the monitoring time intervals, several monitoring time nodes are generated.

[0024] A photovoltaic support tracking and control model is constructed based on the tracking and control sub-models for all terrain types and the corresponding monitoring time nodes.

[0025] In some embodiments of this application, the historical control evaluation value of the corresponding historical control parameter is calculated based on the historical parameter to be evaluated, including:

[0026] Several safety evaluation indicators and several efficiency evaluation indicators are pre-defined;

[0027] Based on several safety evaluation indicators and several efficiency evaluation indicators, the historical parameters to be evaluated in the same historical control log are evaluated and analyzed to obtain several historical evaluation values. Combined with the weight coefficients of the corresponding indicators, the historical control evaluation value is calculated.

[0028] The formula for calculating the historical control evaluation value is as follows:

[0029] ;

[0030] Where P is the historical control evaluation value, u1 is the first weighting coefficient, u2 is the second weighting coefficient, and n1 is the number of safety evaluation indicators. Let be the historical evaluation value of the i-th safety evaluation indicator. Let be the weight coefficient of the i-th safety evaluation indicator, and n2 be the number of efficiency evaluation indicators. Let s be the historical evaluation value of the s-th efficiency evaluation indicator. is the weighting coefficient of the s-th efficiency evaluation index.

[0031] In some embodiments of this application, a predicted risk value for each terrain category is generated based on historical data packets, and a monitoring time interval for the corresponding terrain category is set based on the predicted risk value, including:

[0032] Obtain the degree of fluctuation and rate of change of each environmental feature index in the environmental category of the terrain category;

[0033] Based on the degree of fluctuation and rate of change of each environmental characteristic indicator, a predicted sub-risk value for the corresponding environmental characteristic indicator is generated, and a predicted risk value for the corresponding terrain category is generated by combining the weight coefficient of the corresponding environmental characteristic indicator.

[0034] Pre-set a first preset risk value range, a second preset risk value range, a third preset risk value range, and a fourth preset risk value range;

[0035] When the predicted risk value is within the first preset risk value range, the fourth preset monitoring time interval is selected as the monitoring time interval for the corresponding terrain category;

[0036] When the predicted risk value is within the second preset risk value range, the third preset monitoring time interval is selected as the monitoring time interval for the corresponding terrain category;

[0037] When the predicted risk value is within the third preset risk value range, the second preset monitoring time interval is selected as the monitoring time interval for the corresponding terrain category;

[0038] When the predicted risk value is within the fourth preset risk value range, the first preset monitoring time interval is selected as the monitoring time interval for the corresponding terrain category.

[0039] In some embodiments of this application, real-time characteristic parameters of several photovoltaic supports are obtained, and a real-time tracking control strategy for each photovoltaic support is obtained based on the real-time characteristic parameters and the photovoltaic support tracking control model, including:

[0040] Based on the real-time characteristic parameters, several real-time terrain feature indicators and several real-time environmental feature indicators are determined for the corresponding photovoltaic support.

[0041] A similarity analysis is performed between several real-time terrain feature indicators and several real-time environmental feature indicators and the terrain feature indicators and environmental feature indicators of each terrain category, and several similarity scores are generated based on the analysis results.

[0042] All terrain categories are sorted according to similarity, and the monitoring time point of the top-ranked terrain category is set as the monitoring strategy for the corresponding photovoltaic support.

[0043] Real-time monitoring data of the corresponding photovoltaic support is collected according to the monitoring strategy and input into the sub-tracking control model of the top-ranked terrain category to obtain the real-time tracking control strategy for the current monitoring time node.

[0044] The real-time tracking control strategy includes real-time angle adjustment, real-time speed adjustment, and real-time control duration.

[0045] In some embodiments of this application, the similarity includes:

[0046] Generate the number of first corresponding indicators between all real-time terrain feature indicators and terrain feature indicators of the same terrain category, and calculate the correlation and overlap between each first corresponding indicator;

[0047] The first similarity is obtained by quantitative calculation based on the correlation and overlap of all the first corresponding indicators;

[0048] Generate the number of second corresponding indicators between all real-time environmental feature indicators and environmental feature indicators of the same terrain category, and calculate the correlation and overlap between each second corresponding indicator;

[0049] The second similarity is obtained by quantitative calculation based on the correlation and overlap of all the second corresponding indicators;

[0050] The similarity score is obtained by weighting and averaging the first and second similarity scores.

[0051] In some embodiments of this application, the generation of feedback data packets for the real-time tracking control strategy of each photovoltaic support includes:

[0052] The feedback time node is set according to the monitoring time nodes in the monitoring strategy between various photovoltaics;

[0053] Real-time parameters to be evaluated for the real-time tracking and control strategy are collected according to the feedback time node and the corresponding monitoring time node.

[0054] Feedback data packets for the real-time tracking and control strategy at the current monitoring time point are generated based on the real-time parameters to be evaluated.

[0055] In some embodiments of this application, a real-time control evaluation value is calculated based on the feedback data packet, and a determination is made as to whether to modify the real-time tracking control strategy, including:

[0056] The real-time control evaluation value at the feedback time node is generated based on the feedback data packet, several security evaluation indicators and several efficiency evaluation indicators.

[0057] The first evaluation value difference is generated based on the real-time control evaluation value and the predicted control evaluation value of the real-time tracking control strategy generated by the photovoltaic bracket tracking control model.

[0058] A second evaluation value difference is generated based on the real-time control evaluation value and the control evaluation value threshold.

[0059] Pre-set a first difference threshold and a second difference threshold;

[0060] When the difference between the first evaluation values ​​is not greater than the first difference threshold, it is determined that the real-time tracking control strategy will not be modified.

[0061] When the difference between the first evaluation values ​​is greater than the first difference threshold and the difference between the second evaluation values ​​is not greater than the preset second difference threshold, it is determined that the real-time tracking control strategy should be corrected and a first-level correction instruction should be generated.

[0062] When the difference between the first evaluation value and the difference between the second evaluation value are greater than the first difference threshold and the difference between the second evaluation value and the preset second difference threshold, it is determined that the real-time tracking control strategy should be corrected and a second-level correction instruction should be generated.

[0063] In some embodiments of this application, an adaptive tracking control system for photovoltaic supports in complex terrain is also included. The system includes: a construction module for setting several terrain categories and constructing a photovoltaic support tracking control model for all terrain categories;

[0064] The control module is used to acquire real-time characteristic parameters of several photovoltaic brackets, and to obtain the real-time tracking control strategy for each photovoltaic bracket based on the real-time characteristic parameters and the photovoltaic bracket tracking control model.

[0065] The correction module is used to generate feedback data packets for the real-time tracking control strategy of each photovoltaic support, calculate the real-time control evaluation value based on the feedback data packets, and determine whether to correct the real-time tracking control strategy.

[0066] The adaptive tracking control method and system for photovoltaic supports in complex terrain, as described in this application, have the following advantages compared to existing technologies:

[0067] By setting multiple terrain categories and constructing a photovoltaic support tracking control model, a real-time tracking control strategy is generated. By combining feedback data packets to calculate real-time control evaluation values, shortcomings of the tracking control strategy can be identified and adjusted in a timely manner, achieving efficient adaptive tracking and improving the power generation efficiency and reliability of the photovoltaic power generation system under complex terrain conditions. Attached Figure Description

[0068] Figure 1 This is a flowchart illustrating an adaptive tracking control method for photovoltaic supports in complex terrain, as described in an embodiment of this application. Detailed Implementation

[0069] The specific embodiments of this application will be described in further detail below with reference to the accompanying drawings and examples. The following examples are used to illustrate this application, but are not intended to limit the scope of this application.

[0070] In the description of this application, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0071] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0072] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0073] like Figure 1 As shown in the figure, an adaptive tracking control method for photovoltaic supports in complex terrain according to an embodiment of this application includes:

[0074] Step S101: Define several terrain categories and construct a photovoltaic support tracking control model for all terrain categories;

[0075] Step S102: Obtain real-time characteristic parameters of several photovoltaic supports, and obtain the real-time tracking control strategy for each photovoltaic support based on the real-time characteristic parameters and the photovoltaic support tracking control model;

[0076] Step S103: Generate feedback data packets for the real-time tracking control strategy of each photovoltaic support, calculate the real-time control evaluation value based on the feedback data packets, and determine whether to correct the real-time tracking control strategy.

[0077] In some embodiments of this application, several terrain categories are defined, including:

[0078] Multiple terrain feature indicators are set according to preset terrain parameters;

[0079] Multiple initial terrain categories are generated based on all terrain feature indicators;

[0080] Obtain historical control logs for photovoltaic supports with the same initial terrain category, and determine several environmental impact parameters for the same initial terrain category based on the historical control logs;

[0081] Multiple environmental characteristic indicators are set based on several environmental impact parameters of the same initial terrain category, and multiple environmental categories corresponding to the initial terrain category are generated based on all environmental characteristic indicators.

[0082] Randomly select an environment category for each initial terrain category and combine it with the corresponding initial behavior category to generate a terrain category;

[0083] Several terrain categories are generated sequentially.

[0084] In this embodiment, the preset terrain parameters include, but are not limited to, topography, mountains, hills, orientation, and surface roughness, which are determined based on the terrain parameters of the multiple photovoltaic supports. The terrain feature index is a parameter used to characterize the terrain features after statistically analyzing the preset terrain parameters belonging to the same data type.

[0085] In this embodiment, the environmental impact parameters include at least one of wind speed, precipitation, snowfall, and sunshine duration. The environmental characteristic index refers to setting different data intervals for environmental impact parameters belonging to the same data type. The environmental characteristic index is randomly combined to obtain the environmental category. The environmental category refers to a specific environmental scenario constructed by combining several environmental impact parameters and their specific data intervals.

[0086] In this embodiment, several terrain categories are constructed based on all initial terrain categories and several environmental categories corresponding to each initial terrain category, thereby obtaining a set of terrain categories with different terrain characteristics and different environmental scenarios, providing a foundation for the precise tracking and control of the photovoltaic support in the future.

[0087] In some embodiments of this application, a photovoltaic support tracking control model is constructed for all terrain types, including:

[0088] Obtain historical control logs for photovoltaic supports of each terrain type, and extract historical illumination data, historical environmental impact data, historical control parameters, and historical parameters to be evaluated from each historical control log;

[0089] The historical control parameters include historical adjustment angle, historical adjustment speed, and historical control duration.

[0090] Calculate the historical control evaluation value of the corresponding historical control parameter based on the historical parameters to be evaluated.

[0091] Pre-set control evaluation thresholds;

[0092] Training data sets corresponding to different terrain categories are constructed based on historical control parameters whose historical control evaluation values ​​are greater than the control evaluation value threshold. The training data sets include several parameter control groups.

[0093] Establish a tracking control sub-model corresponding to the terrain category based on the training data package;

[0094] Based on historical data packets, predictive risk values ​​for each terrain category are generated. Based on the predicted risk values, monitoring time intervals for the corresponding terrain categories are set. Based on the monitoring time intervals, several monitoring time nodes are generated.

[0095] A photovoltaic support tracking and control model is constructed based on the tracking and control sub-models for all terrain types and the corresponding monitoring time nodes.

[0096] In this embodiment, historical environmental impact data refers to several environmental impact parameters in the environmental category of the terrain category and the specific data values ​​in the specific data range. Historical illumination data includes information such as historical illumination intensity and historical illumination direction. These data together constitute a comprehensive description of the illumination conditions of the photovoltaic support.

[0097] In this embodiment, the historical control parameters record in detail the adjustment actions of the photovoltaic support under different environmental conditions, including the angle before and after adjustment, the size of the adjustment angle, the speed of adjustment, and the duration of control. The historical parameters to be evaluated refer to the curves of the change in the illuminated area of ​​the photovoltaic module, the change in the power generation efficiency, and the change in the stability of the photovoltaic support after the historical control parameters. These parameters can intuitively reflect the impact of the control actions on the safety and efficiency performance of the photovoltaic support.

[0098] In this embodiment, the parameter control group refers to historical illumination data, historical environmental impact data, and corresponding historical control parameters. The historical control evaluation values ​​of the historical control parameters of the parameter control group are all greater than the control evaluation value threshold. The control evaluation value threshold refers to the minimum control evaluation value that simultaneously meets the safety target and the performance target.

[0099] In this embodiment, by constructing a training data set, the control parameters with better performance are selected as training samples to improve the accuracy and reliability of the model. Combined with the monitoring time nodes determined according to the predicted risk values, the control efficiency under different terrain and environmental conditions is improved, and potential control risks are identified in advance, thereby realizing dynamic monitoring and management of photovoltaic supports.

[0100] In some embodiments of this application, the historical control evaluation value of the corresponding historical control parameter is calculated based on the historical parameter to be evaluated, including:

[0101] Several safety evaluation indicators and several efficiency evaluation indicators are pre-defined;

[0102] Based on several safety evaluation indicators and several efficiency evaluation indicators, the historical parameters to be evaluated in the same historical control log are evaluated and analyzed to obtain several historical evaluation values. Combined with the weight coefficients of the corresponding indicators, the historical control evaluation value is calculated.

[0103] The formula for calculating the historical control evaluation value is as follows:

[0104] ;

[0105] Where P is the historical control evaluation value, u1 is the first weighting coefficient, u2 is the second weighting coefficient, and n1 is the number of safety evaluation indicators. Let be the historical evaluation value of the i-th safety evaluation indicator. Let be the weight coefficient of the i-th safety evaluation indicator, and n2 be the number of efficiency evaluation indicators. Let s be the historical evaluation value of the s-th efficiency evaluation indicator. is the weighting coefficient of the s-th efficiency evaluation index.

[0106] In this embodiment, the safety evaluation index refers to the structural stability index, wind pressure resistance index, and weather resistance index used to evaluate the photovoltaic support under historical control parameters and corresponding terrain conditions and environmental conditions. Specifically, the structural stability index refers to the structural stability, the wind pressure resistance index refers to the resistance to strong winds, and the weather resistance index refers to the durability.

[0107] In this embodiment, efficiency evaluation indicators refer to energy conversion efficiency, tracking accuracy, and tracking efficiency indicators used to assess photovoltaic (PV) mounting systems under historical control parameters and corresponding terrain and environmental conditions. Specifically, energy conversion efficiency refers to the efficiency with which the PV mounting system converts solar energy into electrical energy; tracking accuracy refers to the deviation between the actual tracking angle and the theoretical optimal tracking angle; and tracking efficiency refers to the time it takes for the actual tracking angle of the PV mounting system to reach the theoretical optimal tracking angle.

[0108] In this embodiment, the historical evaluation value refers to the comparison between the historical parameter to be evaluated and the standard parameter range mapped to each indicator. When the historical evaluation value is within the standard parameter range, the larger the historical evaluation value corresponding to the indicator is, and vice versa. The range of the historical evaluation value is 0-1.

[0109] In this embodiment, u1 is 0.6 and u2 is 0.4.

[0110] In this embodiment, by calculating the historical control evaluation value of each historical control parameter, the foundation for constructing a tracking control sub-model for each terrain category is laid. That is, historical control parameters that meet the safety and efficiency objectives under the corresponding terrain and environmental conditions are selected, thereby improving the control efficiency of the subsequent tracking control strategy for each photovoltaic support, as well as the power generation efficiency and safety of the photovoltaic support.

[0111] In some embodiments of this application, a predicted risk value for each terrain category is generated based on historical data packets, and a monitoring time interval for the corresponding terrain category is set based on the predicted risk value, including:

[0112] Obtain the degree of fluctuation and rate of change of each environmental feature index in the environmental category of the terrain category;

[0113] Based on the degree of fluctuation and rate of change of each environmental characteristic indicator, a predicted sub-risk value for the corresponding environmental characteristic indicator is generated, and a predicted risk value for the corresponding terrain category is generated by combining the weight coefficient of the corresponding environmental characteristic indicator.

[0114] Pre-set a first preset risk value range, a second preset risk value range, a third preset risk value range, and a fourth preset risk value range;

[0115] When the predicted risk value is within the first preset risk value range, the fourth preset monitoring time interval is selected as the monitoring time interval for the corresponding terrain category;

[0116] When the predicted risk value is within the second preset risk value range, the third preset monitoring time interval is selected as the monitoring time interval for the corresponding terrain category;

[0117] When the predicted risk value is within the third preset risk value range, the second preset monitoring time interval is selected as the monitoring time interval for the corresponding terrain category;

[0118] When the predicted risk value is within the fourth preset risk value range, the first preset monitoring time interval is selected as the monitoring time interval for the corresponding terrain category.

[0119] In this embodiment, the degree of fluctuation of each environmental characteristic indicator refers to the magnitude of change of the specific data value of the environmental impact parameter of the same data type corresponding to the environmental characteristic indicator within a preset time interval, while the rate of change refers to the amount of change of these data values ​​per unit time.

[0120] In this embodiment, the smaller the change amplitude, the smaller the change amount, and the data value after the change is still within the specific data range of the corresponding environmental characteristic indicator, the smaller the predicted sub-risk value, and vice versa. By analyzing the fluctuation degree and change rate of environmental characteristic indicators of the same environmental category, the trend of environmental conditions in the future can be predicted, and the risks that photovoltaic brackets may face can be assessed.

[0121] In this embodiment, the predicted risk value is obtained by weighted summation of the predicted sub-risk values ​​of all environmental characteristic indicators. The weight coefficients of the environmental characteristic indicators are set according to the degree of influence on the current terrain category, which comprehensively reflects the overall risk level that the photovoltaic support may face under the terrain category.

[0122] In this embodiment, the preset risk value range and preset monitoring time interval are set based on historical data, and the first preset risk value range < the second preset risk value range < the third preset risk value range < the fourth preset risk value range, and the first preset monitoring time interval < the second preset monitoring time interval < the third preset monitoring time interval < the fourth preset monitoring time interval. Different preset risk value ranges correspond to different preset monitoring time intervals. The higher the risk, the shorter the monitoring time interval, so as to ensure that potential safety hazards can be detected and dealt with in a timely manner.

[0123] In this embodiment, by setting the monitoring time interval corresponding to the terrain category based on the predicted risk value, dynamic monitoring and management of the photovoltaic support structure is realized. This dynamic monitoring mechanism can flexibly adjust the monitoring frequency according to the risk level under different terrain and environmental conditions, which not only ensures the safe operation of the photovoltaic support structure, but also improves monitoring efficiency and reduces operation and maintenance costs.

[0124] In some embodiments of this application, real-time characteristic parameters of several photovoltaic supports are obtained, and a real-time tracking control strategy for each photovoltaic support is obtained based on the real-time characteristic parameters and the photovoltaic support tracking control model, including:

[0125] Based on the real-time characteristic parameters, several real-time terrain feature indicators and several real-time environmental feature indicators are determined for the corresponding photovoltaic support.

[0126] A similarity analysis is performed between several real-time terrain feature indicators and several real-time environmental feature indicators and the terrain feature indicators and environmental feature indicators of each terrain category, and several similarity scores are generated based on the analysis results.

[0127] All terrain categories are sorted according to similarity, and the monitoring time point of the top-ranked terrain category is set as the monitoring strategy for the corresponding photovoltaic support.

[0128] Real-time monitoring data of the corresponding photovoltaic support is collected according to the monitoring strategy and input into the sub-tracking control model of the top-ranked terrain category to obtain the real-time tracking control strategy for the current monitoring time node.

[0129] The real-time tracking control strategy includes real-time angle adjustment, real-time speed adjustment, and real-time control duration.

[0130] In this embodiment, the real-time feature parameters include the real-time terrain parameters and real-time environmental impact data at each photovoltaic support. A real-time terrain feature index is determined for real-time terrain parameters of the same data type. The same applies to the real-time environmental feature index, which will not be elaborated here.

[0131] In this embodiment, sorting all terrain categories according to similarity can quickly identify the terrain category that best matches the current environment of the photovoltaic support, thereby determining the corresponding monitoring time node and tracking control sub-model.

[0132] In this embodiment, the monitoring strategy fully considers the risk level under different terrain categories. By setting reasonable monitoring time nodes, it ensures that the photovoltaic support is monitored in real time at critical moments, and potential problems are detected and dealt with in a timely manner. The real-time tracking control strategy is based on the sub-tracking control model of the top-ranked terrain category. This model combines excellent control parameters from historical data and can provide the optimal adjustment angle, adjustment speed and control duration according to the current environmental conditions, thereby achieving efficient and safe operation of the photovoltaic support.

[0133] In some embodiments of this application, the similarity includes:

[0134] Generate the number of first corresponding indicators between all real-time terrain feature indicators and terrain feature indicators of the same terrain category, and calculate the correlation and overlap between each first corresponding indicator;

[0135] The first similarity is obtained by quantitative calculation based on the correlation and overlap of all the first corresponding indicators;

[0136] Generate the number of second corresponding indicators between all real-time environmental feature indicators and environmental feature indicators of the same terrain category, and calculate the correlation and overlap between each second corresponding indicator;

[0137] The second similarity is obtained by quantitative calculation based on the correlation and overlap of all the second corresponding indicators;

[0138] The similarity score is obtained by weighting and averaging the first and second similarity scores.

[0139] In this embodiment, the formula for calculating the first similarity is: Where X1 represents the first similarity score, and m1 represents the number of first corresponding indicators. Let c be the correlation degree of the first corresponding indicator. Let z1 be the overlap of the c-th first corresponding indicator, kc be the weight coefficient of the c-th first corresponding indicator, z1 be the first conversion coefficient, z2 be the second conversion coefficient, and the calculation logic of the second similarity is the same.

[0140] In this embodiment, the first corresponding indicator refers to the indicator that there is a correlation between the real-time terrain feature indicator and the feature indicator of the terrain category. Similarly, the second corresponding indicator refers to the indicator that there is a correlation between the real-time environmental feature indicator and the environmental feature indicator of the terrain category.

[0141] In this embodiment, correlation and overlap refer to the degree of consistency between the parameters corresponding to the first or second corresponding indicator and the specific data range in which the parameters are located. For example, if both indicators reflect terrain slope information and the degree of overlap in their numerical ranges is higher, then the correlation and overlap are higher.

[0142] In this embodiment, the correlation and overlap are calculated and converted into values ​​with the same dimension as the similarity by combining the first and second conversion coefficients. Different indicators are assigned corresponding weight coefficients, which can comprehensively consider the influence of multiple factors on the similarity, thereby obtaining a more accurate first and second similarity. On this basis, the weight coefficients are set according to the importance of each indicator to the operation of the photovoltaic support. The first and second similarities are fused by weighted averaging to obtain the final similarity. The higher the correlation and overlap, the higher the first similarity, and vice versa. This similarity can comprehensively reflect the matching degree between real-time terrain feature indicators, real-time environmental feature indicators and terrain category feature indicators, providing a strong basis for subsequent monitoring strategy formulation and real-time tracking control strategy generation.

[0143] In some embodiments of this application, the generation of feedback data packets for the real-time tracking control strategy of each photovoltaic support includes:

[0144] The feedback time node is set according to the monitoring time nodes in the monitoring strategy between various photovoltaics;

[0145] Real-time parameters to be evaluated for the real-time tracking and control strategy are collected according to the feedback time node and the corresponding monitoring time node.

[0146] Feedback data packets for the real-time tracking and control strategy at the current monitoring time point are generated based on the real-time parameters to be evaluated.

[0147] In this embodiment, the feedback time node is determined by combining the real-time adjustment duration in the real-time tracking control strategy with the preset feedback delay time, so as to ensure that data is collected at an appropriate time after the photovoltaic support completes the adjustment action, so as to accurately evaluate the control effect.

[0148] In this embodiment, according to the feedback time node, the system will automatically collect the real-time evaluation parameters of the real-time tracking control strategy at the corresponding monitoring time node. These parameters include the irradiated area of ​​the photovoltaic module, power generation efficiency, and stability indicators of the photovoltaic support, etc., which correspond to the historical evaluation parameters and are used to reflect the actual effect of the control action in real time, providing a basis for subsequent control strategy optimization.

[0149] In some embodiments of this application, a real-time control evaluation value is calculated based on the feedback data packet, and a determination is made as to whether to modify the real-time tracking control strategy, including:

[0150] The real-time control evaluation value at the feedback time node is generated based on the feedback data packet, several security evaluation indicators and several efficiency evaluation indicators.

[0151] The first evaluation value difference is generated based on the real-time control evaluation value and the predicted control evaluation value of the real-time tracking control strategy generated by the photovoltaic bracket tracking control model.

[0152] A second evaluation value difference is generated based on the real-time control evaluation value and the control evaluation value threshold.

[0153] Pre-set a first difference threshold and a second difference threshold;

[0154] When the difference between the first evaluation values ​​is not greater than the first difference threshold, it is determined that the real-time tracking control strategy will not be modified.

[0155] When the difference between the first evaluation values ​​is greater than the first difference threshold and the difference between the second evaluation values ​​is not greater than the preset second difference threshold, it is determined that the real-time tracking control strategy should be corrected and a first-level correction instruction should be generated.

[0156] When the difference between the first evaluation value and the difference between the second evaluation value are greater than the first difference threshold and the difference between the second evaluation value and the preset second difference threshold, it is determined that the real-time tracking control strategy should be corrected and a second-level correction instruction should be generated.

[0157] In this embodiment, the correction intensity and correction range of the first-level correction command and the second-level correction command are different. Compared with the first-level correction command, the second-level correction command has a greater correction intensity and a wider correction range. The correction range includes correcting one or more control parameters and the photovoltaic support tracking control model to deal with more serious control deviations.

[0158] In this embodiment, the calculation method of the real-time control evaluation value is similar to that of the historical control evaluation value. It is also calculated based on several safety evaluation indicators and several efficiency evaluation indicators to ensure the comprehensiveness and accuracy of the evaluation results.

[0159] In this embodiment, by calculating the real-time control evaluation value and comparing it with the predicted control evaluation value and the control evaluation value threshold, problems in the real-time tracking control strategy can be identified in a timely manner, and corresponding corrective measures can be taken.

[0160] In this embodiment, the preset first difference threshold and the preset second difference threshold are set based on historical data and actual operating experience. They reflect the system's tolerance for control deviations. When the difference between the first evaluation value and the difference between the second evaluation value exceed the corresponding threshold, the system will automatically trigger a correction mechanism to adjust the real-time tracking control strategy to ensure the safe and efficient operation of the photovoltaic support.

[0161] In this embodiment, the correction mechanism includes adjustments to the real-time adjustment angle, real-time adjustment speed, real-time control duration, and a tracking control model. By optimizing these parameters, the photovoltaic support can better adapt to current terrain and environmental conditions, improving power generation efficiency and stability. Simultaneously, the system records the status and effects of each correction, providing data support for subsequent control strategy optimization.

[0162] In some embodiments of this application, an adaptive tracking control system for photovoltaic supports in complex terrain is also included. The system includes: a construction module for setting several terrain categories and constructing a photovoltaic support tracking control model for all terrain categories;

[0163] The control module is used to acquire real-time characteristic parameters of several photovoltaic brackets, and to obtain the real-time tracking control strategy for each photovoltaic bracket based on the real-time characteristic parameters and the photovoltaic bracket tracking control model.

[0164] The correction module is used to generate feedback data packets for the real-time tracking control strategy of each photovoltaic support, calculate the real-time control evaluation value based on the feedback data packets, and determine whether to correct the real-time tracking control strategy.

[0165] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and substitutions can be made without departing from the technical principles of this application, and these improvements and substitutions should also be considered within the scope of protection of this application.

Claims

1. An adaptive tracking control method for photovoltaic supports in complex terrain, characterized in that, include: Several terrain categories are defined, and a photovoltaic support tracking control model is constructed for all terrain categories; Obtain real-time characteristic parameters of several photovoltaic supports, and obtain the real-time tracking control strategy for each photovoltaic support based on the real-time characteristic parameters and the photovoltaic support tracking control model; Feedback data packets for the real-time tracking control strategy of each photovoltaic support are generated. The real-time control evaluation value is calculated based on the feedback data packets, and it is determined whether the real-time tracking control strategy should be corrected.

2. The adaptive tracking control method for photovoltaic supports in complex terrain as described in claim 1, characterized in that, Define several terrain categories, including: Multiple terrain feature indicators are set according to preset terrain parameters; Multiple initial terrain categories are generated based on all terrain feature indicators; Obtain historical control logs for photovoltaic supports with the same initial terrain category, and determine several environmental impact parameters for the same initial terrain category based on the historical control logs; Multiple environmental characteristic indicators are set based on several environmental impact parameters of the same initial terrain category, and multiple environmental categories corresponding to the initial terrain category are generated based on all environmental characteristic indicators. Randomly select an environment category for each initial terrain category and combine it with the corresponding initial behavior category to generate a terrain category; Several terrain categories are generated sequentially.

3. The adaptive tracking control method for photovoltaic supports in complex terrain as described in claim 2, characterized in that, Construct photovoltaic support tracking control models for all terrain types, including: Obtain historical control logs for photovoltaic supports of each terrain type, and extract historical illumination data, historical environmental impact data, historical control parameters, and historical parameters to be evaluated from each historical control log; The historical control parameters include historical adjustment angle, historical adjustment speed, and historical control duration. Calculate the historical control evaluation value of the corresponding historical control parameter based on the historical parameters to be evaluated. Pre-set control evaluation thresholds; Training data sets corresponding to different terrain categories are constructed based on historical control parameters whose historical control evaluation values ​​are greater than the control evaluation value threshold. The training data sets include several parameter control groups. Establish a tracking control sub-model corresponding to the terrain category based on the training data package; Based on historical data packets, predictive risk values ​​for each terrain category are generated. Based on the predicted risk values, monitoring time intervals for the corresponding terrain categories are set. Based on the monitoring time intervals, several monitoring time nodes are generated. A photovoltaic support tracking and control model is constructed based on the tracking and control sub-models for all terrain types and the corresponding monitoring time nodes.

4. The adaptive tracking control method for photovoltaic supports in complex terrain as described in claim 3, characterized in that, Calculate the historical control evaluation values ​​for the corresponding historical control parameters based on the historical parameters to be evaluated, including: Several safety evaluation indicators and several efficiency evaluation indicators are pre-defined; Based on several safety evaluation indicators and several efficiency evaluation indicators, the historical parameters to be evaluated in the same historical control log are evaluated and analyzed to obtain several historical evaluation values. Combined with the weight coefficients of the corresponding indicators, the historical control evaluation value is calculated. The formula for calculating the historical control evaluation value is as follows: ； Where P is the historical control evaluation value, u1 is the first weighting coefficient, u2 is the second weighting coefficient, and n1 is the number of safety evaluation indicators. Let be the historical evaluation value of the i-th safety evaluation indicator. Let be the weight coefficient of the i-th safety evaluation indicator, and n2 be the number of efficiency evaluation indicators. Let s be the historical evaluation value of the s-th efficiency evaluation indicator. is the weighting coefficient of the s-th efficiency evaluation index.

5. The adaptive tracking control method for photovoltaic supports in complex terrain as described in claim 4, characterized in that, Based on historical data packets, predicted risk values ​​are generated for each terrain category. Based on these predicted risk values, monitoring time intervals are set for the corresponding terrain categories, including: Obtain the degree of fluctuation and rate of change of each environmental feature index in the environmental category of the terrain category; Based on the degree of fluctuation and rate of change of each environmental characteristic indicator, a predicted sub-risk value for the corresponding environmental characteristic indicator is generated, and a predicted risk value for the corresponding terrain category is generated by combining the weight coefficient of the corresponding environmental characteristic indicator. Pre-set a first preset risk value range, a second preset risk value range, a third preset risk value range, and a fourth preset risk value range; When the predicted risk value is within the first preset risk value range, the fourth preset monitoring time interval is selected as the monitoring time interval for the corresponding terrain category; When the predicted risk value is within the second preset risk value range, the third preset monitoring time interval is selected as the monitoring time interval for the corresponding terrain category; When the predicted risk value is within the third preset risk value range, the second preset monitoring time interval is selected as the monitoring time interval for the corresponding terrain category; When the predicted risk value is within the fourth preset risk value range, the first preset monitoring time interval is selected as the monitoring time interval for the corresponding terrain category.

6. The adaptive tracking control method for photovoltaic supports in complex terrain as described in claim 5, characterized in that, Obtain real-time characteristic parameters of several photovoltaic (PV) brackets, and derive a real-time tracking control strategy for each PV bracket based on these parameters and the PV bracket tracking control model, including: Based on the real-time characteristic parameters, several real-time terrain feature indicators and several real-time environmental feature indicators are determined for the corresponding photovoltaic support. A similarity analysis is performed between several real-time terrain feature indicators and several real-time environmental feature indicators and the terrain feature indicators and environmental feature indicators of each terrain category, and several similarity scores are generated based on the analysis results. All terrain categories are sorted according to similarity, and the monitoring time point of the top-ranked terrain category is set as the monitoring strategy for the corresponding photovoltaic support. Real-time monitoring data of the corresponding photovoltaic support is collected according to the monitoring strategy and input into the sub-tracking control model of the top-ranked terrain category to obtain the real-time tracking control strategy for the current monitoring time node. The real-time tracking control strategy includes real-time angle adjustment, real-time speed adjustment, and real-time control duration.

7. The adaptive tracking control method for photovoltaic supports in complex terrain as described in claim 6, characterized in that, The similarity includes: Generate the number of first corresponding indicators between all real-time terrain feature indicators and terrain feature indicators of the same terrain category, and calculate the correlation and overlap between each first corresponding indicator; The first similarity is obtained by quantitative calculation based on the correlation and overlap of all the first corresponding indicators; Generate the number of second corresponding indicators between all real-time environmental feature indicators and environmental feature indicators of the same terrain category, and calculate the correlation and overlap between each second corresponding indicator; The second similarity is obtained by quantitative calculation based on the correlation and overlap of all the second corresponding indicators; The similarity score is obtained by weighting and averaging the first and second similarity scores.

8. The adaptive tracking control method for photovoltaic supports in complex terrain as described in claim 7, characterized in that, Generate feedback data packets for the real-time tracking control strategy of each photovoltaic support, including: The feedback time node is set according to the monitoring time nodes in the monitoring strategy between various photovoltaics; Real-time parameters to be evaluated for the real-time tracking and control strategy are collected according to the feedback time node and the corresponding monitoring time node. Feedback data packets for the real-time tracking and control strategy at the current monitoring time point are generated based on the real-time parameters to be evaluated.

9. The adaptive tracking control method for photovoltaic supports in complex terrain as described in claim 8, characterized in that, The real-time control evaluation value is calculated based on the feedback data packet, and it is determined whether the real-time tracking control strategy needs to be modified, including: The real-time control evaluation value at the feedback time node is generated based on the feedback data packet, several security evaluation indicators and several efficiency evaluation indicators. The first evaluation value difference is generated based on the real-time control evaluation value and the predicted control evaluation value of the real-time tracking control strategy generated by the photovoltaic bracket tracking control model. A second evaluation value difference is generated based on the real-time control evaluation value and the control evaluation value threshold. Pre-set a first difference threshold and a second difference threshold; When the difference between the first evaluation values ​​is not greater than the first difference threshold, it is determined that the real-time tracking control strategy will not be modified. When the difference between the first evaluation values ​​is greater than the first difference threshold and the difference between the second evaluation values ​​is not greater than the preset second difference threshold, it is determined that the real-time tracking control strategy should be corrected and a first-level correction instruction should be generated. When the difference between the first evaluation value and the difference between the second evaluation value are greater than the first difference threshold and the difference between the second evaluation value and the preset second difference threshold, it is determined that the real-time tracking control strategy should be corrected and a second-level correction instruction should be generated.

10. An adaptive tracking control system for photovoltaic supports in complex terrain, characterized in that, include: The module is used to define several terrain categories and build a photovoltaic support tracking control model for all terrain categories. The control module is used to acquire real-time characteristic parameters of several photovoltaic brackets, and to obtain the real-time tracking control strategy for each photovoltaic bracket based on the real-time characteristic parameters and the photovoltaic bracket tracking control model. The correction module is used to generate feedback data packets for the real-time tracking control strategy of each photovoltaic support, calculate the real-time control evaluation value based on the feedback data packets, and determine whether to correct the real-time tracking control strategy.

Images