An artificial intelligence-based photovoltaic support adaptive regulation system and method, and a readable storage medium

Abstract

This invention relates to an artificial intelligence-based adaptive control system, method, and readable storage medium for photovoltaic (PV) mounting systems, relating to the field of PV panel control. It includes a control unit, an analysis unit, an environmental monitoring unit, an evaluation unit, and an execution unit. The environmental monitoring unit detects surrounding environmental data and feeds the data back to the analysis unit, evaluation unit, and control unit. The analysis unit receives data from the environmental monitoring unit and generates decision suggestions. The control unit receives data from the environmental monitoring unit, extracts environmental features, and obtains adjustment data based on the decision suggestions from the analysis unit. This invention achieves predictive feedback. Through the analysis and control units, the system analyzes and predicts environmental data, enabling precise decision-making and thus predictive feedback. This allows for advance adjustment of the PV panel angle to adapt to future weather changes, improving system stability and power generation efficiency.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic panel regulation, and in particular to an artificial intelligence-based adaptive regulation system, method, and readable storage medium for photovoltaic supports. Background Technology

[0002] With the deterioration of the environment and the deepening of human environmental awareness, photovoltaic power generation, as one of the main green energy sources, is being used more and more. However, due to the different angles of sunlight in different latitude regions, the installation angle of photovoltaic panels needs to be adjusted. At the same time, the sun rises in the east and sets in the west every day, and the angle of traditional photovoltaic panels is fixed and difficult to change. In this case, the time period during which photovoltaic panels can face the sunlight is actually not long, so the power generation effect of photovoltaic panels is not very efficient.

[0003] In existing technologies, automatic tracking of photovoltaic brackets is usually based on simple sensor data or preset motion trajectories, which often cannot fully adapt to complex and ever-changing environmental conditions. Summary of the Invention

[0004] In view of this, the present invention aims to propose an artificial intelligence-based adaptive control system for photovoltaic supports to solve the problems in the prior art.

[0005] To achieve the above objectives, the technical solution of the present invention is implemented as follows:

[0006] This invention proposes an adaptive control system for photovoltaic brackets based on artificial intelligence:

[0007] It includes a control unit, an analysis unit, an environmental monitoring unit, an evaluation unit, and an execution unit;

[0008] The environmental detection unit is used to detect surrounding environmental data and feed the data back to the analysis unit, evaluation unit, and control unit.

[0009] The analysis unit is used to receive data from the environmental monitoring unit and generate decision recommendations;

[0010] The control unit is used to receive data from the environmental detection unit, extract environmental features, and obtain adjustment data based on the decision suggestions of the analysis unit.

[0011] The execution unit is used to receive data from the control unit and adjust the position and angle of the photovoltaic panel;

[0012] The evaluation unit is able to evaluate overall performance based on the adjustment data of the control unit and the decision recommendations of the analysis unit.

[0013] Furthermore, the control unit includes a collection module, a control module, and a prediction module;

[0014] The collection module is used to receive information from the environmental detection unit and extract data features according to the convolutional neural network feature extraction algorithm;

[0015] The prediction module is used to input data features into the feature calculation prediction model to obtain prediction results;

[0016] The control module is used to send adjustment data to the execution unit to adjust the angle of the photovoltaic panel.

[0017] Furthermore, the feature extraction formula of the convolutional neural network algorithm is as follows:

[0018] conv(xw)=σ(x * w+bconv)

[0019] In the formula: conv is the convolution operation; σ is the ReLU function; * is the convolution operation; x is the input data; w and b are the model weights and biases, which are learned by training the network.

[0020] Furthermore, the feature calculation prediction model is calculated using the following formula:

[0021] y = β1x1 + β2x2 + ... + βnxn + b

[0022] In the formula: y is the prediction result, which is the optimal tilt angle of the photovoltaic panel; x1, x2, ..., xn are the extracted feature values; β1, β2, ..., βn and b are the weights and biases of the model, which are learned by training the model.

[0023] Furthermore, the analysis unit includes a decision-making module and a storage module for storing historical data;

[0024] The analysis unit is used to obtain decision recommendations using the following formula:

[0025] D = F(H,C,P(C),S)

[0026] In the formula: D is the decision structure; H is historical data and environmental data; C is implementation environment data; P(C) is the prediction of the future environment; F(.) is the decision function. The deep neural network learns the relationship between H, C, P(C) and S, and predicts the optimal D based on the current input.

[0027] Furthermore, the analysis unit also includes an optimization module, which obtains the final decision by combining neural networks and support vector machines, and through voting or weighted averaging.

[0028] Furthermore, the evaluation unit obtains the evaluation data using the following formula:

[0029] Eout=η·G·A·cos(θ)·f(C,D,S,M)

[0030] In the formula: Eout represents the output power of the photovoltaic system; η is the energy conversion efficiency of the photovoltaic panel; G is the incident solar radiation intensity; A is the effective area of ​​the photovoltaic panel; θ is the angle between the incident ray and the normal to the photovoltaic panel;

[0031] f(.) is the synthesis function; C is the environmental data collected by the environmental sensor; D represents the decision suggestions of the analysis unit; S is the self-learning and optimization of the control unit; M represents the adjustment of the photovoltaic panel position by the actuator;

[0032] The actuator adjusts the position M of the photovoltaic panel based on y generated by the prediction model.

[0033] This invention also proposes a photovoltaic support control method, which is applied to the aforementioned artificial intelligence-based adaptive control system for photovoltaic supports.

[0034] The control method includes the following steps:

[0035] S1, the environmental monitoring unit collects real-time environmental data and transmits it to the analysis unit, control unit and evaluation unit;

[0036] S2, the analysis unit generates decision recommendations based on historical data, real-time data, future predictions, and control unit status;

[0037] S3, the control unit receives the decision from the analysis unit and, combined with its own learning and optimization results, calculates the optimal angle of the photovoltaic panel;

[0038] S4, The evaluation unit evaluates the adjusted performance based on the information from the analysis unit and the control unit;

[0039] S5, the control unit sends the final adjustment information to the execution unit, and the execution unit controls the photovoltaic panel to adjust its angle.

[0040] The present invention also proposes a computer-readable storage medium having a computer program stored thereon, characterized in that the program, when executed by a processor, implements the steps of the above-described method.

[0041] Compared with the prior art, the present invention has the following advantages:

[0042] In this invention, adaptive control of the photovoltaic support structure is achieved: the system can calculate the optimal angle of the photovoltaic panel based on environmental data to achieve adaptive control and maximize the efficiency of solar energy utilization.

[0043] It achieves continuous learning and self-optimization: The analysis unit can utilize the decision given by the cloud analysis server using the big data model of artificial intelligence semantics, and continuously optimize itself through continuous learning of environmental data, constantly improving decision-making and control strategies to adapt to constantly changing environmental conditions and improve the system's adaptability and efficiency.

[0044] Predictive feedback is achieved: By analyzing and predicting environmental data through the analysis unit and control unit, the system can make accurate decisions, thereby achieving predictive feedback, adjusting the angle of the photovoltaic panels in advance to adapt to future weather changes, and improving the system's stability and power generation efficiency. Attached Figure Description

[0045] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0046] Figure 1 This is a schematic diagram of the overall structure of the present invention;

[0047] Figure 2 This is a structural block diagram of the control system of the present invention;

[0048] Figure 3 This is a structural block diagram of the control unit of the present invention;

[0049] Figure 4 This is a structural block diagram of the analysis unit of the present invention;

[0050] Figure 5 This is a flowchart of the control method of the present invention. Detailed Implementation

[0051] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0052] In the description of this invention, it should be noted that the terms "upper," "lower," "inner," and "back," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for 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 the invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0053] Furthermore, in the description of this invention, unless otherwise explicitly defined, the terms "installation," "connection," "linking," and "connector" 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 of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention in light of the specific circumstances.

[0054] The following will refer to the appendix. Figures 1 to 5 The present invention will be described in detail with reference to the embodiments.

[0055] Example 1

[0056] Overall, this embodiment relates to an artificial intelligence-based adaptive control system for photovoltaic brackets, such as... Figure 1 and Figure 2 As shown, the control system includes a control unit, an analysis unit, an environmental detection unit, an evaluation unit, and an execution unit. The environmental detection unit is used to detect surrounding environmental data and feed the data back to the analysis unit, the evaluation unit, and the control unit.

[0057] The analysis unit receives data from the environmental monitoring unit and generates decision recommendations.

[0058] The control unit is used to receive data from the environmental monitoring unit, extract environmental features, and obtain adjustment data based on the decision suggestions of the analysis unit.

[0059] The execution unit is used to receive data from the control unit and adjust the position and angle of the photovoltaic panel;

[0060] The evaluation unit is able to assess overall performance based on the adjustment data from the control unit and the decision recommendations from the analysis unit.

[0061] In this embodiment, the control unit receives sensor data, analyzes the given decisions, and calculates the optimal angle for the photovoltaic panel. The actuator adjusts the position of the photovoltaic panel according to the instructions of the control unit to maximize solar energy utilization efficiency. The evaluation unit can calculate the adjusted performance, facilitating subsequent optimization adjustments by the control unit and the analysis unit.

[0062] The environmental monitoring unit detects surrounding environmental data and feeds it back to the analysis, evaluation, and control units. In practice, this unit is an environmental sensor. The working principle of environmental sensors is based on various physical, chemical, or biological effects, such as thermoelectric, piezoelectric, photoelectric, and chemical reactions. These effects cause changes in the sensor's electrical parameters, such as resistance, capacitance, and inductance, thereby enabling the measurement of environmental quantities. The sensor converts the collected information into signals that the equipment can process for subsequent analysis and processing.

[0063] The analysis unit's decision-making combines multiple factors, taking into account historical and real-time data, as well as predictions of the future environment, to provide optimal decision recommendations (such as the optimal tilt angle for photovoltaic panels).

[0064] like Figure 4 As shown, the analysis unit includes a decision-making module and a storage module for storing historical data;

[0065] The analysis unit is used to obtain decision recommendations through the following formula:

[0066] D = F(H,C,P(C),S)

[0067] In the formula: D is the decision structure; H is historical data and environmental data; C is implementation environment data; P(C) is the prediction of the future environment; F(.) is the decision function. The deep neural network learns the relationship between H, C, P(C) and S, and predicts the optimal D based on the current input.

[0068] The analysis unit also includes an optimization module, which combines neural networks and support vector machines to obtain the final decision through voting or weighted averaging.

[0069] In detail, the analysis unit continuously learns from industry best practices and a large amount of historical meteorological data, including various meteorological variables such as temperature, humidity, wind speed, air pressure, and precipitation. By extracting features from the meteorological data, it identifies and learns the complex relationships and patterns between meteorological variables to achieve accurate predictions of future weather. It then uses a deep neural network to learn the relationship between H, C, P(C), and S using the formula D = F(H, C, P(C), S). Through reinforcement learning and ensemble learning, it interacts with the environment and optimizes decision-making strategies based on reward signals. Finally, it combines neural networks and support vector machines to obtain the final decision through voting or weighted averaging.

[0070] It should be noted that the preferred analysis unit is an AI semantic large-scale model cloud analysis server. The AI ​​semantic large-scale model cloud analysis server uses large-scale semantic models and deep learning algorithms to analyze and predict environmental data, and can achieve accurate prediction of future weather through continuous learning, thereby making accurate decisions.

[0071] In addition, such as Figure 3 As shown, the control unit described above includes a collection module, a control module, and a prediction module;

[0072] The collection module is used to receive information from the environmental detection unit and extract data features according to the convolutional neural network feature extraction algorithm;

[0073] The prediction module is used to input data features into the feature calculation prediction model to obtain prediction results;

[0074] The control module is used to send adjustment data to the execution unit to adjust the angle of the photovoltaic panel.

[0075] The feature extraction formula of the convolutional neural network algorithm is as follows:

[0076] conv(xw)=σ(x * w+bconv)

[0077] In the formula: conv is the convolution operation; σ is the ReLU function; * is the convolution operation; x is the input data; w and b are the model weights and biases, which are learned by training the network.

[0078] The feature-based prediction model is calculated using the following formula:

[0079] y = β1x1 + β2x2 + ... + βnxn + b

[0080] In the formula: y is the prediction result, which is the optimal tilt angle of the photovoltaic panel; x1, x2, ..., xn are the extracted feature values; β1, β2, ..., βn and b are the weights and biases of the model, which are learned by training the model.

[0081] After obtaining the prediction results, they are compared and analyzed with the decision recommendations obtained by the analysis unit to obtain the best adjustment results.

[0082] In detail, firstly, environmental data is input into a convolutional neural network for feature extraction. The convolutional neural network uses the formula conv(xw) = σ(xw) * The operation is performed using w+bconv). x represents the input environmental data, W and b are the weights and biases of the convolutional layer, σ is the ReLU function, and * represents the convolution operation. The convolutional layer performs convolution operations on the input data x at multiple locations, extracting various features, and then uses the ReLU function σ to perform a non-linear transformation on the output, resulting in a new set of feature values ​​that reflect the characteristics of the input data.

[0083] The artificial intelligence control unit calculates using an internal prediction model: the features extracted by the convolutional neural network in the previous step are input into the prediction model for calculation. The prediction model uses the formula y = β1x1 + β2x2 + ... + βnxn + b. x1, x2, ..., xn are the feature values ​​extracted by the convolutional neural network, β1, β2, ..., βn and b are the weights and biases of the prediction model, and y is the predicted output, which is the optimal tilt angle of the photovoltaic panel. The prediction model uses the feature values ​​extracted from the convolutional neural network, combines these features through a weighted average, and adds a bias term to form the final prediction result y.

[0084] The connection between Convolutional Neural Networks (CNNs) and predictive model computation lies in the fact that the output of the CNN, i.e., the result of feature extraction, is used as the input to the predictive model for computation. This makes the predictions more accurate and robust, enabling better adjustment of the tilt angle of photovoltaic panels and improving the utilization rate of solar energy.

[0085] By combining control and analysis units, the photovoltaic panels can dynamically adjust their angle based on real-time environmental conditions and future forecasts, maximizing solar energy utilization efficiency. Furthermore, it can predict environmental changes in advance and adjust the panel angle accordingly, improving system stability and power generation efficiency.

[0086] In addition, the evaluation unit obtains evaluation data using the following formula:

[0087] Eout=η·G·A·cos(θ)·f(C,D,S,M)

[0088] In the formula: Eout represents the output power of the photovoltaic system; η is the energy conversion efficiency of the photovoltaic panel; G is the incident solar radiation intensity; A is the effective area of ​​the photovoltaic panel; θ is the angle between the incident ray and the normal to the photovoltaic panel;

[0089] f(.) is the synthesis function; C is the environmental data collected by the environmental sensor; D represents the decision suggestions of the analysis unit; S is the self-learning and optimization of the control unit; M represents the adjustment of the photovoltaic panel position by the actuator;

[0090] The actuator adjusts the position M of the photovoltaic panel based on the y generated by the prediction model. The actuator adjusts the position M of the photovoltaic panel based on the y generated by the prediction model (i.e., the predicted optimal angle).

[0091] The formula comprehensively considers the physical characteristics of the photovoltaic system itself, as well as the impact of predictive feedback using a large-scale semantic model of artificial intelligence on system performance. By maximizing the value of f(C,D,S,M), the optimal control effect can be obtained, thereby improving the overall efficiency of photovoltaic power generation.

[0092] In summary, this invention achieves adaptive control of the photovoltaic support structure: the system can calculate the optimal angle of the photovoltaic panel based on environmental data to achieve adaptive control and maximize the efficiency of solar energy utilization.

[0093] It achieves continuous learning and self-optimization: The analysis unit can utilize the decision given by the cloud analysis server using the big data model of artificial intelligence semantics, and continuously optimize itself through continuous learning of environmental data, constantly improving decision-making and control strategies to adapt to constantly changing environmental conditions and improve the system's adaptability and efficiency.

[0094] It should be further explained that the execution unit can adjust the position of the photovoltaic panel according to the instructions of the control unit. In specific implementation, the execution unit can be composed of multiple motors, electric motors, telescopic rods, etc., to control the angle and position of the photovoltaic panel.

[0095] Example 2

[0096] This embodiment proposes a photovoltaic support control method, which is used in an artificial intelligence-based adaptive control system for photovoltaic supports described in Embodiment 1 above. Figure 5 As shown, the method includes:

[0097] S1, the environmental monitoring unit collects real-time environmental data and transmits it to the analysis unit, control unit and evaluation unit;

[0098] S2, the analysis unit generates decision recommendations based on historical data, real-time data, future predictions, and control unit status;

[0099] S3, the control unit receives the decision from the analysis unit and, combined with its own learning and optimization results, calculates the optimal angle of the photovoltaic panel;

[0100] S4, The evaluation unit evaluates the adjusted performance based on the information from the analysis unit and the control unit;

[0101] S5, the control unit sends the final adjustment information to the execution unit, and the execution unit controls the photovoltaic panel to adjust its angle.

[0102] In this embodiment, predictive feedback is implemented: by analyzing and predicting environmental data through the analysis unit and control unit, the system can make accurate decisions, thereby achieving predictive feedback, adjusting the angle of the photovoltaic panels in advance to adapt to future weather changes, and improving the stability and power generation efficiency of the system.

[0103] Example 3

[0104] This embodiment proposes a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method proposed in Embodiment 2 above.

[0105] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An adaptive control system for photovoltaic brackets based on artificial intelligence, characterized in that: It includes a control unit, an analysis unit, an environmental monitoring unit, an evaluation unit, and an execution unit; The environmental detection unit is used to detect surrounding environmental data and feed the data back to the analysis unit, evaluation unit, and control unit. The analysis unit is used to receive data from the environmental monitoring unit and generate decision recommendations; The control unit is used to receive data from the environmental detection unit, extract environmental features, and obtain adjustment data based on the decision suggestions of the analysis unit. The execution unit is used to receive data from the control unit and adjust the position and angle of the photovoltaic panel; The evaluation unit is able to evaluate the overall performance based on the adjustment data of the control unit and the decision suggestions of the analysis unit; The control unit includes a collection module, a control module, and a prediction module; The collection module is used to receive information from the environmental detection unit and extract data features according to the convolutional neural network feature extraction algorithm; The prediction module is used to input data features into the feature calculation prediction model to obtain prediction results; The control module is used to send adjustment data to the execution unit to adjust the angle of the photovoltaic panel; The feature calculation and prediction model is derived using the following formula: In the formula: y is the predicted result, which is the optimal tilt angle of the photovoltaic panel; x1, x2, ..., xn are the extracted feature values; β1, β2, ..., βn and b are the weights and biases of the model, which are learned by training the model; The analysis unit includes a decision-making module and a storage module for storing historical data; The analysis unit is used to obtain decision recommendations using the following formula: In the formula: D is the decision structure; H is historical data and environmental data; C is real-time environmental data; P(C) is the prediction of the future environment; F(.) is the decision function. The deep neural network learns the relationship between H, C, P(C) and S, and predicts the optimal D based on the current input.

2. The adaptive control system for photovoltaic brackets based on artificial intelligence according to claim 1, characterized in that: The feature extraction formula of the convolutional neural network feature extraction algorithm is as follows: In the formula: conv is the convolution operation; σ is the ReLU function; * is the convolution operation; x is the input data; w and b are the model weights and biases, which are learned by training the network.

3. The adaptive control system for photovoltaic brackets based on artificial intelligence according to claim 1, characterized in that: The analysis unit also includes an optimization module, which obtains the final decision by combining neural networks and support vector machines, and through voting or weighted averaging.

4. The adaptive control system for photovoltaic brackets based on artificial intelligence according to claim 3, characterized in that: The evaluation unit obtains the evaluation data using the following formula: In the formula: Eout represents the output power of the photovoltaic system; η is the energy conversion efficiency of the photovoltaic panel; G is the incident solar radiation intensity; A is the effective area of ​​the photovoltaic panel; θ is the angle between the incident ray and the normal to the photovoltaic panel; f() is the synthesis function; C is the environmental data collected by the environmental sensors; D represents the decision suggestions of the analysis unit; S represents the self-learning and optimization of the control unit; M represents the adjustment of the photovoltaic panel position by the actuator. The actuator adjusts the position M of the photovoltaic panel based on y generated by the prediction model.

5. A method for regulating a photovoltaic support structure, characterized in that, The method is applied to the AI-based adaptive control system for photovoltaic brackets described in any one of claims 1 to 4: The control method includes the following steps: S1, the environmental monitoring unit collects real-time environmental data and transmits it to the analysis unit, control unit and evaluation unit; S2, the analysis unit generates decision recommendations based on historical data, real-time data, future predictions, and control unit status; S3, the control unit receives the decision from the analysis unit and, combined with its own learning and optimization results, calculates the optimal angle of the photovoltaic panel; S4, The evaluation unit evaluates the adjusted performance based on the information from the analysis unit and the control unit; S5, the control unit sends the final adjustment information to the execution unit, and the execution unit controls the photovoltaic panel to adjust its angle.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the method described in claim 5.

Images