An automatic light supplement control method and system for vegetable cultivation

Abstract

The application relates to the technical field of vegetable cultivation, and discloses an automatic light supplement control method and system for vegetable cultivation, which comprises the following steps: determining the growth state of a vegetable crop based on crop image information and a crop growth model, judging whether to execute a light supplement strategy according to a solar elevation angle, determining a backtracking light supplement strategy or an initial light supplement strategy according to the number of occurrences of the growth state in a historical light supplement database, and determining target light supplement data of the vegetable crop according to the number of light supplement data; when the initial light supplement strategy is determined, target light supplement data of the vegetable crop is determined based on a recurrent neural network model, historical light supplement records are analyzed to determine a light supplement swing factor of a light supplement lamp, the target light supplement data is adjusted based on the light supplement swing factor, and the vegetable crop is light supplemented according to the adjusted target light supplement data. The crop growth model and the light supplement swing factor ensure the reliability of light supplement.

Description

Technical Field

[0001] This invention relates to the field of vegetable cultivation technology, and more specifically, to an automated supplemental lighting control method and system for vegetable cultivation. Background Technology

[0002] Light is a crucial factor affecting the photosynthesis, growth cycle, yield, and quality of vegetables. Vegetable growth requires specific conditions such as photosynthetically active radiation and light duration. For example, leafy vegetables need 8-10 hours of moderate light daily to promote leaf differentiation. However, natural light is affected by factors such as season, region, and weather. Therefore, supplemental lighting has become an important piece of equipment for vegetable cultivation in greenhouses and other facilities.

[0003] When it comes to vegetable cultivation, supplemental lighting is usually set based on human experience. These human-set parameters depend on the grower's experience. Due to differences in individual understanding, the parameters set by different growers do not match the actual needs of the vegetables. Relying on human experience makes it impossible to dynamically adjust the supplemental lighting according to the vegetable cultivation situation. The lack of a feedback adjustment mechanism results in insufficient reliability in vegetable cultivation.

[0004] Therefore, it is necessary to design an automated supplemental lighting control method and system for vegetable cultivation to solve the problems existing in the current technology. Summary of the Invention

[0005] In view of this, the present invention proposes an automated supplemental lighting control method and system for vegetable cultivation, aiming to solve the problems that the parameters set by different growers do not match the actual needs of vegetables, that it is impossible to dynamically adjust the supplemental lights according to the vegetable cultivation conditions based on human experience, and that the lack of a feedback adjustment mechanism results in insufficient reliability of vegetable cultivation.

[0006] In one aspect, the present invention proposes an automated supplemental lighting control method for vegetable cultivation, comprising:

[0007] Obtain crop image information of vegetable crops, determine the growth status of vegetable crops based on the crop image information and crop growth model, and determine whether to implement supplemental lighting strategy based on solar altitude angle;

[0008] When it is determined that the supplemental lighting strategy should be executed, the historical supplemental lighting database is searched based on the growth state, and the retrospective supplemental lighting strategy or the initial supplemental lighting strategy is determined according to the number of times the growth state appears in the historical supplemental lighting database.

[0009] When the retrospective supplemental lighting strategy is determined, the corresponding supplemental lighting data is determined in the historical supplemental lighting database based on the growth status, and the target supplemental lighting data of the vegetable crop is determined according to the quantity of the supplemental lighting data. When the initial supplemental lighting strategy is determined, the target supplemental lighting data of the vegetable crop is determined based on the recurrent neural network model.

[0010] Historical supplemental lighting records of the supplemental lighting lamps are collected, and the supplemental lighting swing factor of the supplemental lighting lamps is determined by analyzing the historical supplemental lighting records. The target supplemental lighting data is adjusted based on the supplemental lighting swing factor, and the vegetable crops are supplemented with light according to the adjusted target supplemental lighting data.

[0011] Furthermore, when determining the growth status of the vegetable crop based on the crop image information and the crop growth model, the process includes:

[0012] Obtain the growth dataset of the vegetable crop, divide the growth dataset into a training set and a test set, find the model parameters according to grid search, and build a convolutional neural network model based on the model parameters;

[0013] The convolutional neural network model is trained based on the training set, the test set is substituted into the trained convolutional neural network model and the prediction accuracy is determined, and the crop growth model is determined based on the prediction accuracy.

[0014] Furthermore, when determining the growth status of the vegetable crop based on the crop image information and the crop growth model, the method further includes:

[0015] When the prediction accuracy is greater than or equal to the prediction accuracy threshold, the currently trained convolutional neural network model is determined as the crop growth model.

[0016] When the prediction accuracy is less than the prediction accuracy threshold, the magnitude of the gradient direction change of the currently trained convolutional neural network model is reduced, and training continues.

[0017] The crop image information is substituted into the crop growth model to determine the growth status of the vegetable crop.

[0018] Furthermore, when determining whether to implement a supplemental lighting strategy based on the solar altitude angle, the following steps are taken:

[0019] A solar altitude angle threshold is preset, and the solar altitude angle threshold is compared with the solar altitude angle.

[0020] When the solar altitude angle is greater than or equal to the solar altitude angle threshold, it is determined that the supplementary lighting strategy will not be executed.

[0021] When the solar altitude angle is less than the solar altitude angle threshold, the supplementary lighting strategy is executed.

[0022] The solar altitude angle threshold is determined based on the growth state.

[0023] Furthermore, when determining the retrospective lighting strategy or the initial lighting strategy based on the frequency of occurrence of the growth state in the historical lighting database, the process includes:

[0024] The historical supplementary lighting database includes several historical growth states, several historical supplementary lighting data, and the number of occurrences of each historical growth state. Each historical growth state corresponds to a historical supplementary lighting data. The historical supplementary lighting data includes historical supplementary lighting intensity and historical supplementary lighting angle.

[0025] When the growth state has the same historical growth state in the historical supplementary lighting database, and the corresponding occurrence count is greater than or equal to the occurrence count threshold, it is determined to be the retrospective supplementary lighting strategy.

[0026] If the growth state does not have the same historical growth state in the historical supplementary lighting database, or if the growth state has the same historical growth state in the historical supplementary lighting database and the corresponding occurrence count is less than the occurrence count threshold, then it is determined as the initial supplementary lighting strategy.

[0027] Furthermore, when determining the corresponding supplemental lighting data in the historical supplemental lighting database based on the growth status, and determining the target supplemental lighting data for the vegetable crop based on the quantity of the supplemental lighting data, the process includes:

[0028] Extract the historical growth states that are the same as those in the historical supplemental lighting database, and determine the average value of the historical supplemental lighting data corresponding to each historical growth state as the target supplemental lighting data for the vegetable crop.

[0029] Furthermore, when determining the target supplemental lighting data for the vegetable crop based on a recurrent neural network model, the process includes:

[0030] A recurrent neural network model was pre-trained, and the solar illumination trajectory was determined based on a solar tracking algorithm;

[0031] The input layer of the recurrent neural network model is used to receive the solar illumination trajectory and growth status;

[0032] The hidden layers of the recurrent neural network model include a first LSTM layer, a second LSTM layer, a first fully connected layer, and a second fully connected layer.

[0033] The first LSTM layer has 128 neurons, the second LSTM layer has 64 neurons, and the first and second fully connected layers are used to compress features.

[0034] The output layer of the recurrent neural network model includes a supplementary light intensity output layer and a supplementary light angle output layer;

[0035] The number of neurons in the supplementary light intensity output layer and the supplementary light angle output layer is 1.

[0036] The target supplementary lighting data for the vegetable crop is determined based on the supplementary light intensity output layer and the supplementary light angle output layer.

[0037] Furthermore, when collecting historical supplementary lighting records and analyzing these records to determine the supplementary lighting oscillation factor, the process includes:

[0038] Determine the swaying behavior and non-swaying behavior in the historical supplementary lighting records, and obtain the number of swaying behaviors and the number of non-swaying behaviors;

[0039] Obtain the ratio m between the number of swings and the number of non-swings;

[0040] The first preset fill light swing factor, the second preset fill light swing factor, and the third preset fill light swing factor are preset.

[0041] When m≤1, the first preset fill light swing factor is determined as the fill light swing factor of the fill light;

[0042] When 1 < m ≤ 1.5, the second preset fill light swing factor is determined as the fill light swing factor of the fill light;

[0043] When 1.5 < m, the third preset fill light swing factor is determined as the fill light swing factor of the fill light;

[0044] The first preset fill light swing factor is less than the second preset fill light swing factor, and the second preset fill light swing factor is less than the third preset fill light swing factor.

[0045] Furthermore, when adjusting the target illumination data based on the illumination swing factor, the following steps are included:

[0046] The target illumination data is directly proportional to the illumination swing factor.

[0047] Compared with existing technologies, the advantages of this invention are as follows: It determines the growth status of vegetable crops based on crop image information and crop growth models, ensuring the matching degree between supplemental lighting parameters and the growth needs of vegetable crops. This avoids the disconnect between supplemental lighting and the actual needs of vegetable crops caused by relying on human experience. It determines whether to implement a supplemental lighting strategy by using the solar altitude angle, achieving dynamic response to environmental changes. It determines a retrospective or initial supplemental lighting strategy based on the frequency of occurrence of the growth status in the historical supplemental lighting database. By reusing mature experience or using a recurrent neural network model to determine target supplemental lighting data, it ensures the stability of supplemental lighting under normal growth conditions while adapting to special or new growth conditions, ensuring the flexibility and adaptability of supplemental lighting. It collects historical supplemental lighting records to determine the supplemental lighting swing factor, forming a closed-loop feedback of growth status monitoring, supplemental lighting execution, historical record analysis, and parameter adjustment. It dynamically controls the supplemental lighting effect based on the growth status of vegetable crops and the fluctuation range of supplemental lighting, improving the reliability of supplemental lighting and providing stable photosynthetically effective radiation and illumination duration for vegetable crops.

[0048] On the other hand, this application also provides an automated supplemental lighting control system for vegetable cultivation, used to apply the above-mentioned automated supplemental lighting control method for vegetable cultivation, including:

[0049] The acquisition and judgment module is configured to acquire crop image information of vegetable crops, determine the growth status of vegetable crops based on the crop image information and crop growth model, and determine whether to implement a supplementary lighting strategy based on the solar altitude angle.

[0050] The strategy determination module is configured to, when it is determined that the supplemental lighting strategy should be executed, search the historical supplemental lighting database based on the growth state, and determine the retrospective supplemental lighting strategy or the initial supplemental lighting strategy based on the number of times the growth state appears in the historical supplemental lighting database.

[0051] The supplemental lighting determination module is configured to, when the retrospective supplemental lighting strategy is determined, determine the corresponding supplemental lighting data in the historical supplemental lighting database based on the growth status, and determine the target supplemental lighting data of the vegetable crop based on the quantity of the supplemental lighting data; when the initial supplemental lighting strategy is determined, determine the target supplemental lighting data of the vegetable crop based on a recurrent neural network model.

[0052] The supplemental lighting control module is configured to collect historical supplemental lighting records of the supplemental lighting lamp, analyze the historical supplemental lighting records to determine the supplemental lighting swing factor of the supplemental lighting lamp, adjust the target supplemental lighting data based on the supplemental lighting swing factor, and supplement the vegetable crop with supplemental lighting according to the adjusted target supplemental lighting data.

[0053] It is understandable that the above-mentioned automated supplemental lighting control method and system for vegetable cultivation have the same beneficial effects, and will not be elaborated further here. Attached Figure Description

[0054] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0055] Figure 1 A flowchart illustrating an automated supplemental lighting control method for vegetable cultivation, provided in an embodiment of the present invention;

[0056] Figure 2 This is a functional block diagram of an automated supplemental lighting control system for vegetable cultivation, provided as an embodiment of the present invention. Detailed Implementation

[0057] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the disclosure to those skilled in the art. It should be noted that, unless otherwise specified, embodiments and features in the embodiments of the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0058] In some embodiments of this application, see Figure 1 As shown, an automated supplemental lighting control method for vegetable cultivation includes:

[0059] S100: Acquire crop image information of vegetable crops, determine the growth status of vegetable crops based on crop image information and crop growth model, and determine whether to implement supplemental lighting strategy based on solar altitude angle.

[0060] S200: When it is determined to execute the supplemental lighting strategy, the historical supplemental lighting database is searched based on the growth status, and the retrospective supplemental lighting strategy or the initial supplemental lighting strategy is determined according to the number of times the growth status appears in the historical supplemental lighting database.

[0061] S300: When the retrospective supplemental lighting strategy is determined, the corresponding supplemental lighting data is determined in the historical supplemental lighting database based on the growth status, and the target supplemental lighting data for vegetable crops is determined according to the number of supplemental lighting data. When the initial supplemental lighting strategy is determined, the target supplemental lighting data for vegetable crops is determined based on the recurrent neural network model.

[0062] S400: Collects historical supplemental lighting records of supplemental lights, analyzes these records to determine the supplemental lighting swing factor, adjusts the target supplemental lighting data based on the swing factor, and provides supplemental lighting to vegetable crops according to the adjusted target supplemental lighting data.

[0063] Specifically, vegetable crops are cultivated in greenhouses and other facilities. High-definition cameras and other image acquisition devices are used to collect crop images. These images reflect visual characteristics such as plant height, number of leaves, and stem thickness. A crop growth model is used to extract the growth status of the vegetables from these images. The growth status distinguishes different stages of the vegetables, such as the seedling stage, vegetative growth stage, and reproductive growth stage, and includes data such as leaf color, degree of expansion, and leaf thickness. The growth status directly reflects the growth of the vegetables, providing a data basis for matching appropriate supplemental lighting parameters. The solar altitude angle is a key factor determining the intensity and duration of natural sunlight received by greenhouses and other facilities. When the solar altitude angle is higher, such as at noon, the path of sunlight through the atmosphere is shorter, and the energy loss due to absorption and scattering of light by the atmosphere is less. This results in higher photosynthetically active radiation reaching the vegetable surface, i.e., the intensity of light available for vegetable photosynthesis. Conversely, when the solar altitude angle is lower, such as in the early morning, evening, or winter, the path of light through the atmosphere is longer, energy loss increases, and the actual photosynthetically active radiation reaching the crop will decrease, possibly even falling below the minimum threshold required for the current growth stage of the vegetables. By monitoring the solar altitude angle, we can determine the actual supply capacity of natural sunlight and analyze whether it meets the needs of vegetables for photosynthetically active radiation and light duration under the current growth state, thereby deciding whether to implement supplemental lighting strategies. When it is determined that a supplemental lighting strategy needs to be implemented, the system first searches the historical supplemental lighting database. Based on the matching between the current growth status and the records in the historical supplemental lighting database, a retrospective supplemental lighting strategy or an initial supplemental lighting strategy is dynamically adopted. If a retrospective supplemental lighting strategy is determined, it means that there are a certain number of historical records for the growth status in the historical supplemental lighting database. Therefore, the corresponding supplemental lighting data is determined in the historical supplemental lighting database, and the target supplemental lighting data for the vegetable crop is determined based on the number of supplemental lighting data. If an initial supplemental lighting strategy is determined, it means that there are few or no historical records for the growth status in the historical supplemental lighting database. Directly using the data in the historical supplemental lighting database is prone to adjustment deviations. Therefore, a recurrent neural network model is used to generate target supplemental lighting data by combining the temporal pattern of vegetable growth and the photosynthetic demand characteristics. Furthermore, the supplemental lighting in greenhouses and other facilities can sway due to weather conditions such as sandstorms and precipitation. By collecting historical supplemental lighting records and analyzing the swaying patterns in these records, a supplemental lighting sway factor reflecting the stability of the supplemental lighting can be determined. The target supplemental lighting data can then be fine-tuned based on this sway factor, and finally, supplemental lighting can be applied to vegetable crops according to the adjusted target supplemental lighting data, ensuring the reliability of the supplemental lighting.

[0064] Understandably, individual differences in perception can lead to a disconnect between the settings of supplemental lighting and the actual needs of vegetables. However, by determining the growth status of vegetables through crop image information and crop growth models, subjective biases are eliminated. Combining the solar altitude angle to determine the necessity of supplemental lighting effectively avoids energy waste when natural light is sufficient. At the same time, it also prevents the omission of supplemental lighting when light is insufficient, which would affect the growth of vegetables. The historical supplemental lighting database not only utilizes effective past data but also covers growth states with no or limited historical data through recurrent neural networks, further improving the reliability of supplemental lighting. A feedback adjustment mechanism is established based on the supplemental lighting swing factor to avoid parameter deviations caused by environmental changes in supplemental lighting, improve the accuracy of supplemental lighting, and reduce reliance on human experience. This ensures the photosynthesis of vegetables, and even if natural light is affected by season, region, and weather, stable supplemental lighting can ensure the growth of vegetables.

[0065] In some embodiments of this application, when determining the growth status of vegetable crops based on crop image information and crop growth models, the process includes: acquiring a growth dataset of vegetable crops and dividing the growth dataset into a training set and a test set; finding model parameters using grid search; establishing a convolutional neural network model based on the model parameters; training the convolutional neural network model based on the training set; substituting the test set into the trained convolutional neural network model and determining the prediction accuracy; and determining the crop growth model based on the prediction accuracy.

[0066] In some embodiments of this application, when determining the growth status of vegetable crops based on crop image information and crop growth models, the method further includes: when the prediction accuracy is greater than or equal to the prediction accuracy threshold, the currently trained convolutional neural network model is determined as the crop growth model; when the prediction accuracy is less than the prediction accuracy threshold, the magnitude of the change in the gradient direction of the currently trained convolutional neural network model is reduced, and training continues, and the crop image information is substituted into the crop growth model to determine the growth status of vegetable crops.

[0067] Specifically, a growth dataset of vegetable crops is obtained. The growth dataset includes multi-dimensional images of vegetable crops at different growth stages and in different health conditions, such as full-cycle images from the seedling stage to the reproductive stage, as well as close-up images of leaves in different states such as normal, yellowing, and wilting. The growth dataset is divided into a training set for model training and a test set for performance verification, usually in a ratio of 7:3. Model parameters suitable for image feature analysis are selected through grid search. These parameters include learning rate, activation function, and regularization parameters. A convolutional neural network (CNN) model is then built based on these parameters. CNN models excel at extracting detailed features from images, which aligns with the processing needs of crop image information. The model is then trained using a training set. After each training iteration, a test set is input into the model to calculate the prediction accuracy. Prediction accuracy is a crucial factor in evaluating model performance, and a prediction accuracy threshold of 0.85 is preferred. If the prediction accuracy reaches or exceeds the threshold, the model's performance is considered relatively stable, training can be stopped, and the trained model is designated as the crop growth model. If the prediction accuracy threshold is not reached, the model's gradient direction is reduced (the learning rate is improved to mitigate parameter fluctuations during training), and training continues until the prediction accuracy reaches the target. Finally, the growth status of the vegetable crop is determined based on the crop growth model.

[0068] Understandably, grid search can find the optimal model parameters through filtering, avoiding the subjective bias of human experience, while the application of convolutional neural network models can specifically extract crop image information, effectively capturing the growth status of vegetable crops, thereby ensuring the accuracy of supplemental lighting control.

[0069] In some embodiments of this application, when determining whether to execute a supplementary lighting strategy based on the solar altitude angle, the method includes: presetting a solar altitude angle threshold and comparing the solar altitude angle threshold with the solar altitude angle. When the solar altitude angle is greater than or equal to the solar altitude angle threshold, it is determined that the supplementary lighting strategy will not be executed. When the solar altitude angle is less than the solar altitude angle threshold, it is determined that the supplementary lighting strategy will be executed. The solar altitude angle threshold is determined based on the growth state.

[0070] Specifically, a solar altitude angle threshold is preset based on the current growth state of the vegetable crop. Since the needs of vegetables for photosynthetically effective radiation and light duration differ at different growth states, for example, the light requirement during the reproductive growth stage is usually higher than that during the seedling stage. Moreover, natural light is affected by the solar cycle. When the solar altitude angle is larger, such as at noon, the path of sunlight through the atmosphere is shorter, and the energy loss of light through absorption and scattering by the atmosphere is less, so the light intensity available for vegetable photosynthesis is higher. Conversely, when the solar altitude angle is smaller, such as in the early morning, evening, or winter, the path of light through the atmosphere is longer, the energy loss is increased, and the actual photosynthetically effective radiation intensity reaching the crop will be reduced. Therefore, a solar altitude angle threshold that is suitable for the needs of vegetables at different growth stages is matched. The solar altitude angle threshold is bound to the growth state, but no specific limitation is made on the solar altitude angle threshold here. The solar altitude angle in the environment is monitored in real time and compared with the solar altitude angle threshold. If the current solar altitude angle is greater than or equal to the solar altitude angle threshold, it means that the natural light intensity, effective radiation and duration can meet the needs of the current growth state of the vegetables, and it is determined that no supplemental lighting strategy will be implemented. If the current solar altitude angle is less than the solar altitude angle threshold, it means that the natural light supply is insufficient and cannot support the vegetables to carry out photosynthesis efficiently, and it is determined that a supplemental lighting strategy will be implemented.

[0071] Understandably, traditional methods either rely solely on subjective human judgment regarding whether supplemental lighting is needed, without considering the solar altitude angle, or use fixed solar altitude angle thresholds, ignoring the different needs of vegetables at different growth stages. This leads to energy waste from supplemental lighting even when natural sunlight is sufficient, or missed supplemental lighting when sunlight is insufficient, thus affecting vegetable growth. Linking the solar altitude angle threshold to the growth stage allows for the determination of supplemental lighting timing to simultaneously consider both the natural light supply capacity and the actual needs of the crop, reducing ineffective energy consumption from supplemental lighting and preventing growth stagnation caused by insufficient sunlight, further ensuring the effectiveness of supplemental lighting control.

[0072] In some embodiments of this application, when determining a retrospective lighting strategy or an initial lighting strategy based on the number of times a growth state appears in a historical lighting database, the following steps are taken: The historical lighting database includes several historical growth states, several historical lighting data, and the number of times each historical growth state appears, and each historical growth state corresponds to a historical lighting data. The historical lighting data includes historical lighting intensity and historical lighting angle. When the same historical growth state exists in the historical lighting database and the corresponding number of appearances is greater than or equal to a threshold number of appearances, a retrospective lighting strategy is determined. When the same historical growth state does not exist in the historical lighting database, or when the same historical growth state exists in the historical lighting database and the corresponding number of appearances is less than a threshold number of appearances, an initial lighting strategy is determined.

[0073] In some embodiments of this application, when determining the corresponding supplementary lighting data in the historical supplementary lighting database based on the growth status, and determining the target supplementary lighting data of the vegetable crop based on the quantity of supplementary lighting data, the method includes: extracting the same historical growth status in the historical supplementary lighting database, and determining the average value of the historical supplementary lighting data corresponding to each historical growth status as the target supplementary lighting data of the vegetable crop.

[0074] Specifically, the historical supplemental lighting database stores multiple sets of historical information, including the historical growth status of the vegetable crop and the historical supplemental lighting data corresponding to each historical growth status, as well as the number of times each historical growth status appears in the historical supplemental lighting database. When determining the supplemental lighting strategy, the current growth status of the vegetable crop is compared with the historical growth status in the historical supplemental lighting database. The preferred threshold for the number of occurrences is 5. If the same historical growth status exists and the number of occurrences of the historical growth status is greater than or equal to the occurrence threshold, the retrospective supplemental lighting strategy is adopted. If no matching historical growth status is found, or if the matching historical growth status is found but the number of occurrences is less than the occurrence threshold, the initial supplemental lighting strategy is adopted. When the retrospective supplemental lighting strategy is adopted, all historical growth statuses that are the same as the current growth status are extracted from the historical supplemental lighting database. Then, the average value of all historical supplemental lighting data corresponding to these historical growth statuses is calculated. Finally, the average value is determined as the target supplemental lighting data for the current vegetable crop. The target supplemental lighting data is the supplemental lighting intensity and supplemental lighting angle of the supplemental light.

[0075] Understandably, traditional methods rely on human experience to set the intensity and angle of supplemental lighting, without utilizing historical data. Furthermore, different growers with varying levels of experience can lead to deviations in the set parameters. However, by matching historical supplemental lighting databases with growth status, retrospective supplemental lighting strategies can rely on objective data to directly determine target supplemental lighting data, thereby ensuring the reliability and consistency of supplemental lighting operations. Through data-driven automated adjustments, reliance on human experience and intuition is reduced, lowering the uncertainty and risk of supplemental lighting deviations caused by human judgment, thus improving the automation level and accuracy of supplemental lighting control.

[0076] In some embodiments of this application, when determining the target supplemental lighting data for vegetable crops based on a recurrent neural network model, the process includes: pre-training the recurrent neural network model and determining the solar illumination trajectory based on a solar tracking algorithm. The input layer of the recurrent neural network model is used to receive the solar illumination trajectory and growth status. The hidden layers of the recurrent neural network model include a first LSTM layer, a second LSTM layer, a first fully connected layer, and a second fully connected layer. The first LSTM layer has 128 neurons, and the second LSTM layer has 64 neurons. The first and second fully connected layers are used to compress features. The output layer of the recurrent neural network model includes a supplemental lighting intensity output layer and a supplemental lighting angle output layer. The supplemental lighting intensity output layer and the supplemental lighting angle output layer each have 1 neuron. The target supplemental lighting data for vegetable crops is determined based on the supplemental lighting intensity output layer and the supplemental lighting angle output layer.

[0077] Specifically, if no identical historical growth state is found, or if the number of occurrences of the identical historical growth state is less than a threshold, an initial supplemental lighting strategy is adopted. This initial supplemental lighting strategy is based on a recurrent neural network (RNN) model to determine the target supplemental lighting data for the vegetable crop. The RNN model is pre-trained, and the training process is consistent with that of a convolutional neural network (CNN) model, so it will not be repeated here. The RNN model has the ability to integrate temporal information with crop growth requirements. It uses a solar tracking algorithm to capture and determine the sun's trajectory at different times in real time. This trajectory can intuitively reflect the dynamic changes in natural light. The process of capturing and determining the sun's trajectory at different times using a solar tracking algorithm is lengthy and well-established, and will not be described in detail here. The solar illumination trajectory and the current growth status of the vegetables are input into the recurrent neural network model. After receiving these two types of information, the input layer passes the information to the hidden layer. Two LSTM layers are responsible for processing the temporal features contained in the solar illumination trajectory (such as the changing trend of the sun's position throughout the day), accurately capturing the fluctuation pattern of natural light over time. Then, two fully connected layers integrate and compress the features processed by the LSTM layers, removing redundant information to simplify the calculation. The output layer is divided into a supplementary light intensity output layer and a supplementary light angle output layer, which output the model's prediction results for the supplementary light parameters, respectively. The target supplementary light data can be determined based on the results of these two output layers.

[0078] Understandably, the trajectory of sunlight has certain temporal characteristics. Recurrent neural network models are good at processing temporal information and can effectively correlate changes in light intensity at different times with the crop's supplemental lighting needs. By inputting the trajectory of sunlight and the crop's growth status, the model can balance the natural light supply with the actual needs of the crop, avoiding supplemental lighting deviations caused by single-factor judgments. This reduces the waste of supplemental lighting energy and enhances the reliability and adaptability of automated supplemental lighting control.

[0079] In some embodiments of this application, when collecting historical supplementary lighting records and analyzing these records to determine the supplementary lighting swing factor, the process includes: determining swinging behavior and non-swinging behavior in the historical supplementary lighting records, obtaining the number of swinging behaviors and the number of non-swinging behaviors, obtaining the ratio m between the number of swinging behaviors and the number of non-swinging behaviors, and pre-setting a first preset supplementary lighting swing factor, a second preset supplementary lighting swing factor, and a third preset supplementary lighting swing factor. When m ≤ 1, the first preset supplementary lighting swing factor is determined as the supplementary lighting swing factor of the supplementary lighting; when 1 < m ≤ 1.5, the second preset supplementary lighting swing factor is determined as the supplementary lighting swing factor of the supplementary lighting; when 1.5 < m, the third preset supplementary lighting swing factor is determined as the supplementary lighting swing factor of the supplementary lighting. The first preset supplementary lighting swing factor is less than the second preset supplementary lighting swing factor, and the second preset supplementary lighting swing factor is less than the third preset supplementary lighting swing factor.

[0080] Specifically, the first preset fill light swing factor is preferably 1.1, the second preset fill light swing factor is preferably 1.3, and the third preset fill light swing factor is preferably 1.5. All past fill light records are collected, and swing behavior and non-swing behavior are distinguished from these records. Swing behavior refers to unstable fluctuations in the fill light angle during the fill light process, such as wind disturbances caused by sandstorms or rainfall affecting the fill light angle. Non-swing behavior refers to situations where the fill light angle remains stable without significant fluctuations. The number of swing behaviors and the number of non-swing behaviors are counted, and the ratio m is calculated. A larger ratio indicates a higher degree of environmental swaying affecting the fill light. When m ≤ 1, it indicates that the fill light is subjected to... If the degree of swaying due to environmental influence is minimized, then the minimum first preset supplementary light sway factor is determined as the supplementary light sway factor. When 1 < m ≤ 1.5, it indicates that the degree of swaying due to environmental influence is moderate, so the second preset supplementary light sway factor is determined as the supplementary light sway factor. When 1.5 < m, it indicates that the degree of swaying due to environmental influence is relatively severe, indicating that there is a serious deficiency in supplementary light irradiation for vegetable crops, so the maximum third preset supplementary light sway factor is determined as the supplementary light sway factor to enhance the supplementary light angle and intensity.

[0081] It is understandable that when supplemental lighting is actually working, it is affected by factors such as ambient airflow, and the supplemental lighting parameters will deviate from the ideal supplemental lighting conditions. By distinguishing between swaying behavior and non-swaying behavior and calculating the ratio m, the instability of the target supplemental lighting data is effectively quantified. Then, different preset supplemental lighting swaying factors are used to compensate for the corresponding fluctuation levels, ensuring the actual supplemental lighting state of the supplemental lighting, making up for the lack of fluctuation correction in traditional supplemental lighting, reducing the impact of actual supplemental lighting deviation on vegetable growth, further improving the accuracy and reliability of supplemental lighting control, and thus ensuring that vegetables obtain stable photosynthetically effective radiation.

[0082] In some embodiments of this application, when adjusting the target supplementary lighting data based on the supplementary lighting swing factor, the target supplementary lighting data is proportional to the supplementary lighting swing factor.

[0083] Specifically, the target supplementary lighting data is adjusted based on the supplementary lighting swing factor. Assuming the supplementary lighting intensity is L, the supplementary lighting angle is H, and the supplementary lighting swing factor is P, the adjusted target supplementary lighting data is determined to be L*P and H*P. When a wider supplementary lighting angle and stronger supplementary lighting intensity are needed to compensate for fluctuations in the supplementary lighting, the adjusted target supplementary lighting data will increase synchronously as the supplementary lighting swing factor increases. By establishing a direct proportional relationship between the target supplementary lighting data and the supplementary lighting swing factor, precise control of the supplementary lighting intensity and angle is achieved, ensuring the reliability of the supplementary lighting.

[0084] In summary, the beneficial effects of this invention are as follows: It determines the growth status of vegetable crops based on crop image information and crop growth models, ensuring the matching degree between supplemental lighting parameters and the growth needs of vegetable crops. This avoids the disconnect between supplemental lighting and the actual needs of vegetable crops caused by relying on human experience. It determines whether to implement a supplemental lighting strategy by using the solar altitude angle, achieving dynamic response to environmental changes. It determines a retrospective or initial supplemental lighting strategy based on the frequency of occurrence of the growth status in the historical supplemental lighting database. By reusing mature experience or using a recurrent neural network model to determine target supplemental lighting data, it ensures the stability of supplemental lighting under normal growth conditions while adapting to special or new growth conditions, ensuring the flexibility and adaptability of supplemental lighting. Collecting historical supplemental lighting records from the supplemental lights determines the supplemental lighting swing factor, forming a closed-loop feedback loop of growth status monitoring, supplemental lighting execution, historical record analysis, and parameter adjustment. It dynamically controls the supplemental lighting effect based on the growth status of the vegetable crops and the fluctuation range of supplemental lighting, improving the reliability of supplemental lighting and providing stable photosynthetically effective radiation and illumination duration for vegetable crops.

[0085] In another preferred embodiment based on the above embodiments, see [reference] Figure 2 As shown, this embodiment provides an automated supplemental lighting control system for vegetable cultivation, which applies the above-described automated supplemental lighting control method for vegetable cultivation, including:

[0086] The acquisition and judgment module is configured to acquire crop image information of vegetable crops, determine the growth status of vegetable crops based on crop image information and crop growth model, and determine whether to implement supplemental lighting strategy based on solar altitude angle.

[0087] The strategy determination module is configured to, when it is determined to execute the supplementary lighting strategy, search the historical supplementary lighting database based on the growth status, and determine the retrospective supplementary lighting strategy or the initial supplementary lighting strategy based on the number of times the growth status appears in the historical supplementary lighting database.

[0088] The supplemental lighting determination module is configured to determine the corresponding supplemental lighting data in the historical supplemental lighting database based on the growth status when the retrospective supplemental lighting strategy is determined, and to determine the target supplemental lighting data of the vegetable crop based on the number of supplemental lighting data. When the initial supplemental lighting strategy is determined, the target supplemental lighting data of the vegetable crop is determined based on the recurrent neural network model.

[0089] The supplemental lighting control module is configured to collect historical supplemental lighting records of the supplemental lighting lamps, analyze the historical supplemental lighting records to determine the supplemental lighting swing factor of the supplemental lighting lamps, adjust the target supplemental lighting data based on the supplemental lighting swing factor, and provide supplemental lighting to the vegetable crops according to the adjusted target supplemental lighting data.

[0090] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program goods. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program goods embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0091] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program goods according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0092] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0093] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0094] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. An automated supplemental lighting control method for vegetable cultivation, characterized in that, include: Obtain crop image information of vegetable crops, determine the growth status of vegetable crops based on the crop image information and crop growth model, and determine whether to implement supplemental lighting strategy based on solar altitude angle; When it is determined that the supplemental lighting strategy should be executed, the historical supplemental lighting database is searched based on the growth state, and the retrospective supplemental lighting strategy or the initial supplemental lighting strategy is determined according to the number of times the growth state appears in the historical supplemental lighting database. When the retrospective supplemental lighting strategy is determined, the corresponding supplemental lighting data is determined in the historical supplemental lighting database based on the growth status, and the target supplemental lighting data of the vegetable crop is determined according to the quantity of the supplemental lighting data. When the initial supplemental lighting strategy is determined, the target supplemental lighting data of the vegetable crop is determined based on the recurrent neural network model. Collect historical supplementary lighting records of supplementary lights, analyze the historical supplementary lighting records to determine the supplementary lighting swing factor of the supplementary lights, adjust the target supplementary lighting data based on the supplementary lighting swing factor, and supplement the vegetable crops with supplementary lighting according to the adjusted target supplementary lighting data; When collecting historical supplemental lighting records and analyzing these records to determine the supplemental lighting swing factor, the process includes: Determine the swaying behavior and non-swaying behavior in the historical supplementary lighting records, and obtain the number of swaying behaviors and the number of non-swaying behaviors; Obtain the ratio m between the number of swings and the number of non-swings; The first preset fill light swing factor, the second preset fill light swing factor, and the third preset fill light swing factor are preset. When m≤1, the first preset fill light swing factor is determined as the fill light swing factor of the fill light; When 1 < m ≤ 1.5, the second preset fill light swing factor is determined as the fill light swing factor of the fill light; When 1.5 < m, the third preset fill light swing factor is determined as the fill light swing factor of the fill light; The first preset fill light swing factor is smaller than the second preset fill light swing factor, and the second preset fill light swing factor is smaller than the third preset fill light swing factor; When adjusting the target supplementary lighting data based on the supplementary lighting swing factor, the target supplementary lighting data is proportional to the supplementary lighting swing factor.

2. The automated supplemental lighting control method for vegetable cultivation according to claim 1, characterized in that, Determining the growth status of the vegetable crop based on the crop image information and the crop growth model includes: Obtain the growth dataset of the vegetable crop, divide the growth dataset into a training set and a test set, find the model parameters according to grid search, and build a convolutional neural network model based on the model parameters; The convolutional neural network model is trained based on the training set, the test set is substituted into the trained convolutional neural network model and the prediction accuracy is determined, and the crop growth model is determined based on the prediction accuracy.

3. The automated supplemental lighting control method for vegetable cultivation according to claim 2, characterized in that, When determining the growth status of the vegetable crop based on the crop image information and the crop growth model, the method further includes: When the prediction accuracy is greater than or equal to the prediction accuracy threshold, the currently trained convolutional neural network model is determined as the crop growth model. When the prediction accuracy is less than the prediction accuracy threshold, the magnitude of the gradient direction change of the currently trained convolutional neural network model is reduced, and training continues. The crop image information is substituted into the crop growth model to determine the growth status of the vegetable crop.

4. The automated supplemental lighting control method for vegetable cultivation according to claim 3, characterized in that, When determining whether to implement supplemental lighting based on the solar altitude angle, the following should be included: A solar altitude angle threshold is preset, and the solar altitude angle threshold is compared with the solar altitude angle. When the solar altitude angle is greater than or equal to the solar altitude angle threshold, it is determined that the supplementary lighting strategy will not be executed. When the solar altitude angle is less than the solar altitude angle threshold, the supplementary lighting strategy is executed. The solar altitude angle threshold is determined based on the growth state.

5. The automated supplemental lighting control method for vegetable cultivation according to claim 4, characterized in that, When determining the retrospective lighting strategy or the initial lighting strategy based on the frequency of occurrence of the growth state in the historical lighting database, the following steps are included: The historical supplementary lighting database includes several historical growth states, several historical supplementary lighting data, and the number of occurrences of each historical growth state. Each historical growth state corresponds to a historical supplementary lighting data. The historical supplementary lighting data includes historical supplementary lighting intensity and historical supplementary lighting angle. When the growth state has the same historical growth state in the historical supplementary lighting database, and the corresponding occurrence count is greater than or equal to the occurrence count threshold, it is determined to be the retrospective supplementary lighting strategy. If the growth state does not have the same historical growth state in the historical supplementary lighting database, or if the growth state has the same historical growth state in the historical supplementary lighting database and the corresponding occurrence count is less than the occurrence count threshold, then it is determined as the initial supplementary lighting strategy.

6. The automated supplemental lighting control method for vegetable cultivation according to claim 5, characterized in that, When determining the corresponding supplemental lighting data in the historical supplemental lighting database based on the growth status, and determining the target supplemental lighting data for the vegetable crop based on the quantity of the supplemental lighting data, the process includes: Extract the historical growth states that are the same as those in the historical supplemental lighting database, and determine the average value of the historical supplemental lighting data corresponding to each historical growth state as the target supplemental lighting data for the vegetable crop.

7. The automated supplemental lighting control method for vegetable cultivation according to claim 6, characterized in that, When determining the target supplemental lighting data for the vegetable crop based on a recurrent neural network model, the following are included: A recurrent neural network model was pre-trained, and the solar illumination trajectory was determined based on a solar tracking algorithm; The input layer of the recurrent neural network model is used to receive the solar illumination trajectory and growth status; The hidden layers of the recurrent neural network model include a first LSTM layer, a second LSTM layer, a first fully connected layer, and a second fully connected layer. The first LSTM layer has 128 neurons, the second LSTM layer has 64 neurons, and the first and second fully connected layers are used to compress features. The output layer of the recurrent neural network model includes a supplementary light intensity output layer and a supplementary light angle output layer; The number of neurons in the supplementary light intensity output layer and the supplementary light angle output layer is 1. The target supplementary lighting data for the vegetable crop is determined based on the supplementary light intensity output layer and the supplementary light angle output layer.

8. An automated supplemental lighting control system for vegetable cultivation, used in applying the automated supplemental lighting control method for vegetable cultivation as described in any one of claims 1-7, characterized in that, include: The acquisition and judgment module is configured to acquire crop image information of vegetable crops, determine the growth status of vegetable crops based on the crop image information and crop growth model, and determine whether to implement a supplementary lighting strategy based on the solar altitude angle. The strategy determination module is configured to, when it is determined that the supplemental lighting strategy should be executed, search the historical supplemental lighting database based on the growth state, and determine the retrospective supplemental lighting strategy or the initial supplemental lighting strategy based on the number of times the growth state appears in the historical supplemental lighting database. The supplemental lighting determination module is configured to, when the retrospective supplemental lighting strategy is determined, determine the corresponding supplemental lighting data in the historical supplemental lighting database based on the growth status, and determine the target supplemental lighting data of the vegetable crop based on the quantity of the supplemental lighting data; when the initial supplemental lighting strategy is determined, determine the target supplemental lighting data of the vegetable crop based on a recurrent neural network model. The supplemental lighting control module is configured to collect historical supplemental lighting records of the supplemental lighting lamp, analyze the historical supplemental lighting records to determine the supplemental lighting swing factor of the supplemental lighting lamp, adjust the target supplemental lighting data based on the supplemental lighting swing factor, and supplement the vegetable crop with supplemental lighting according to the adjusted target supplemental lighting data.

Images