Greenhouse environment regulation and control method and system for strawberry flowering influence analysis

Abstract

The invention discloses a greenhouse environment regulation and control method and system for strawberry flowering influence analysis, and relates to the technical field of greenhouse environment regulation and control, and the method comprises the steps: obtaining basic variety information of target strawberries, carrying out the big data retrieval, and obtaining a historical flowering index set and a historical environment parameter set; obtaining a historical flowering quality factor set; obtaining a plurality of cluster historical environment parameter sets; determining environmental parameters in a plurality of cluster sets; obtaining a plurality of target greenhouse environment regulation and control parameters; environment regulation and control are conducted on the multiple test greenhouses, and strawberry flowering influence analysis tests are conducted on the target strawberries in the multiple test greenhouses after regulation and control are completed. The technical problems of low greenhouse environment regulation and control accuracy and poor test reliability during strawberry flowering influence analysis in the prior art are solved, and the technical effects of determining reliable greenhouse environment regulation and control parameters and improving the greenhouse environment regulation and control accuracy are achieved.

Description

Technical Field

[0001] The invention relates to the technical field of greenhouse environment control, and in particular to a greenhouse environment control method and system for strawberry flowering impact analysis. Background Art

[0002] At present, greenhouse environmental control often relies on manual settings and empirical data, which leads to low accuracy in environmental parameter control. Taking temperature and humidity as an example, traditional methods often adjust the parameters of the devices in the greenhouse according to experience. However, the adjusted devices cannot accurately meet the temperature and humidity requirements and need to be adjusted repeatedly, which affects the progress of the experiment. Due to the inaccurate control of greenhouse environmental parameters and the interference of external uncontrollable factors during the growth of strawberries, it is easy to cause inconsistency and non-repeatability of test results. For example, in some experiments, even under the same strawberry variety and planting environment requirements, deviations in greenhouse environmental control may lead to deviations in strawberry flowering, increasing the error and uncertainty of the research results.

[0003] The existing technology has technical problems such as low accuracy of greenhouse environment control and poor test reliability when analyzing the impact of strawberry flowering. Summary of the invention

[0004] The present application provides a greenhouse environment control method and system for strawberry flowering impact analysis, which is used to solve the technical problems of low accuracy of greenhouse environment control and poor test reliability in strawberry flowering impact analysis in the prior art.

[0005] In view of the above problems, the present application provides a greenhouse environment control method and system for strawberry flowering impact analysis.

[0006] In a first aspect of the present application, a greenhouse environment control method for strawberry flowering impact analysis is provided, the method comprising:

[0007] The basic variety information of the target strawberry is obtained, and a preset flowering index is used as a search target. The basic variety information is used as an index to perform a big data search to obtain a historical flowering index set and a historical environmental parameter set, wherein each historical flowering index corresponds to a historical environmental parameter set; a flowering quality evaluation is performed based on the historical flowering index set to obtain a historical flowering quality factor set; the historical flowering quality factor set is traversed to perform a similar clustering analysis to determine a plurality of clustered historical flowering quality factor sets, and the historical environmental parameter set is mapped and clustered based on the plurality of clustered historical flowering quality factor sets to obtain a plurality of clustered historical environmental parameter sets; a parameter centralized search is performed on the plurality of clustered historical environmental parameter sets respectively to determine a plurality of clustered centralized environmental parameters; greenhouse environmental control parameters are configured according to the plurality of clustered centralized environmental parameters to obtain a plurality of target greenhouse environmental control parameters; devices of a plurality of test greenhouses are controlled based on the plurality of target greenhouse environmental control parameters, and a strawberry flowering impact analysis test is performed on the target strawberry in the plurality of test greenhouses after the control is completed.

[0008] The second aspect of the present application provides a greenhouse environment control system for strawberry flowering impact analysis, the system comprising:

[0009] A historical environmental parameter set acquisition module is used to obtain basic variety information of target strawberries, take a preset flowering index as a search target, and perform big data search with the basic variety information as an index to obtain a historical flowering index set and a historical environmental parameter set, wherein each historical flowering index corresponds to a historical environmental parameter set;

[0010] A historical flowering quality factor set acquisition module, used to evaluate the flowering quality based on the historical flowering indicator set to obtain a historical flowering quality factor set;

[0011] a clustering historical environment parameter set acquisition module, used for traversing the historical flowering quality factor set to perform similar clustering analysis, determine multiple clustering historical flowering quality factor sets, and map and cluster the historical environment parameter set based on the multiple clustering historical flowering quality factor sets to obtain multiple clustering historical environment parameter sets;

[0012] A cluster centralized environmental parameter determination module, used to perform a centralized parameter search on the multiple cluster historical environmental parameter sets respectively to determine multiple cluster centralized environmental parameters;

[0013] A greenhouse environment control parameter acquisition module is used to configure greenhouse environment control parameters according to the multiple clustered environmental parameters to obtain multiple target greenhouse environment control parameters;

[0014] The environmental control module is used to control the devices of multiple test greenhouses based on the multiple target greenhouse environmental control parameters, and to perform strawberry flowering influence analysis tests on the target strawberries in the multiple test greenhouses after the control is completed.

[0015] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0016] The present application obtains basic variety information of target strawberries, takes preset flowering indicators as retrieval targets, and performs big data retrieval with basic variety information as index, so as to obtain a historical flowering indicator set and a historical environmental parameter set, wherein each historical flowering indicator corresponds to a historical environmental parameter set, and then performs flowering quality evaluation based on the historical flowering indicator set to obtain a historical flowering quality factor set, and then traverses the historical flowering quality factor set to perform similar clustering analysis to determine multiple clustered historical flowering quality factor sets, and maps and clusters the historical environmental parameter set based on the multiple clustered historical flowering quality factor sets to obtain multiple clustered historical environmental parameter sets, and then performs parameter centralized search on the multiple clustered historical environmental parameter sets to determine multiple clustered centralized environmental parameters, and then performs greenhouse environmental control parameter configuration according to the multiple clustered centralized environmental parameters to obtain multiple target greenhouse environmental control parameters, and then controls the devices of multiple test greenhouses based on the multiple target greenhouse environmental control parameters, and performs strawberry flowering impact analysis tests on target strawberries in the multiple test greenhouses after the control is completed. The technical effect of providing reliable environmental parameters, optimizing the configuration of environmental control parameters for the experimental greenhouse, and improving the accuracy of control has been achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0018] Figure 1 A schematic flow chart of a greenhouse environment control method for strawberry flowering impact analysis provided in an embodiment of the present application;

[0019] Figure 2 A schematic diagram of a process for obtaining environmental parameters in the first cluster in a greenhouse environment control method for strawberry flowering impact analysis provided in an embodiment of the present application;

[0020] Figure 3 A schematic diagram of the structure of a greenhouse environment control system for strawberry flowering impact analysis provided in an embodiment of the present application.

[0021] Explanation of the accompanying drawings: historical environmental parameter set acquisition module 11, historical flowering quality factor set acquisition module 12, cluster historical environmental parameter set acquisition module 13, cluster centralized environmental parameter determination module 14, greenhouse environment control parameter acquisition module 15, environment control module 16. DETAILED DESCRIPTION

[0022] The present application provides a greenhouse environment control method and system for strawberry flowering impact analysis, aiming to solve the technical problems of low accuracy of greenhouse environment control and poor test reliability in strawberry flowering impact analysis in the prior art.

[0023] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0024] It should be noted that the terms "including" and "having" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.

[0025] Embodiment 1, as Figure 1 As shown, the present application provides a greenhouse environment control method for strawberry flowering impact analysis, wherein the method comprises:

[0026] S100: Obtain basic variety information of target strawberries, take a preset flowering index as a search target, perform big data search with the basic variety information as an index, and obtain a historical flowering index set and a historical environmental parameter set, wherein each historical flowering index corresponds to a historical environmental parameter set;

[0027] In a possible embodiment, the target strawberry is any type of strawberry that needs to be analyzed for flowering influence, and may be a red-faced strawberry. The basic variety information is the basic characteristics and genetic information related to the strawberry variety, which is used to help understand the growth characteristics, environmental requirements and flowering characteristics of the variety. The preset flowering index is a set of key indicators for measuring the flowering state of strawberries during the growth process, including indicators such as flowering time, inflorescence number, number of flowers, petal integrity and flowering synchronization.

[0028] Based on the basic variety information of the target strawberry, a search is performed in a pre-stored agricultural data platform or big data system with a preset flowering index as the search target. The agricultural data platform or big data system contains a large amount of historical data, including the flowering index of strawberries and corresponding environmental parameters (such as temperature, humidity, light intensity, etc.). The historical flowering index set and the historical environmental parameter set are obtained according to the search results. Each historical flowering index is associated with a corresponding historical environmental parameter set (such as greenhouse environmental data within a certain time period).

[0029] The retrieved historical flowering index set and the corresponding historical environmental parameter set are stored and sorted as basic data for subsequent analysis. Each set of historical flowering indexes and its corresponding historical environmental parameter set form a one-to-one correspondence in the data, achieving the technical effect of providing basic data support for subsequent flowering quality evaluation and environmental regulation.

[0030] S200: performing a flowering quality evaluation based on the historical flowering indicator set to obtain a historical flowering quality factor set;

[0031] Further, the flower quality evaluation is performed based on the historical flower index set to obtain a historical flower quality factor set. Step S200 in the embodiment of the present application further includes:

[0032] Acquire multiple sample flowering index sets and multiple sample flowering quality factor sets as training sample data;

[0033] The training sample data is divided into a training set and a validation set according to a preset ratio, the training set is used to train a framework built based on a convolutional neural network, the mapping relationship between a flowering quality indicator set and a flowering quality factor is learned, and the validation set is used to supervise the training results until the training converges, thereby obtaining a trained quality evaluation network layer;

[0034] The quality evaluation network layer is used to analyze the historical flowering indicator set to obtain the historical flowering quality factor set.

[0035] In one embodiment, the historical flowering index set reflects the flowering of strawberries under different environmental conditions, and the historical flowering index set is subjected to an in-depth and comprehensive analysis to determine the flowering quality and generate the historical flowering quality factor set. Each historical flowering quality factor reflects the flowering quality of strawberries under a historical environmental parameter, and the higher the historical flowering quality factor, the better the corresponding strawberry flowering quality.

[0036] In one embodiment, multiple sample flowering index sets and multiple sample flowering quality factor sets are obtained as training sample data. In order to construct a quality evaluation network layer, the training sample data is divided into a training set and a validation set according to a preset ratio. Optionally, the preset ratio can be 3:2, 4:1, etc. The training set accounts for the majority (e.g., 80%) and is used to train the model; the validation set accounts for a smaller portion (e.g., 20%) and is used to supervise and verify the training effect of the model. Thus, the overfitting of the network layer is avoided, and the performance of the network layer on new data is evaluated through the validation set to ensure the generalization ability of the model.

[0037] Preferably, a framework built based on a convolutional neural network (CNN) is trained using a training set to automatically learn the complex mapping relationship between flowering quality factors and flowering indicators. Among them, convolutional neural networks are a deep learning method that is suitable for processing input data with spatial structures (such as images or time series data), and can extract high-order features of data through convolution and pooling operations at multiple levels. During the training process, using the data in the training set, CNN will automatically adjust the weights and bias parameters in the network to minimize the prediction error. The input of the network is a set of flowering indicators, and the output is the corresponding flowering quality factor.

[0038] During the training process, the validation set is used to supervise the model in real time. After each training, the validation set data is used to evaluate the performance of the model and check whether the model is overfitting or underfitting. If the error on the validation set continues to decrease, and the errors of the training set and validation set are close, it means that the model has converged and training can be stopped. After training and verification, the CNN model will learn the mapping relationship between flowering indicators and flowering quality factors. After completing the training, the quality evaluation network layer of the model has been optimized to the best state, which can efficiently and accurately analyze and evaluate the flowering indicators of strawberries.

[0039] Furthermore, the trained quality evaluation network layer is used to analyze the historical flowering index set to obtain the historical flowering quality factor set. By using the trained quality evaluation network layer, the historical flowering index set is analyzed to obtain the flowering quality factor corresponding to each historical flowering index. The historical flowering quality factor analysis is realized automatically. The technical effect of providing a basis for environmental regulation of greenhouses is achieved.

[0040] S300: traversing the historical flowering quality factor sets to perform similar clustering analysis, determining multiple clustered historical flowering quality factor sets, and mapping and clustering the historical environment parameter sets based on the multiple clustered historical flowering quality factor sets to obtain multiple clustered historical environment parameter sets;

[0041] In one possible embodiment, cluster analysis is an unsupervised learning algorithm that aims to divide samples into multiple clusters based on their similarities. In this step, an appropriate clustering algorithm (such as K-means, DBSCAN or hierarchical clustering) is used to group historical flowering quality factors. Through cluster analysis, the flowering quality factor set is divided into different sets according to similarity, and each set represents a class of similar flowering quality factor data, which helps to understand the environmental conditions corresponding to different strawberry flowering qualities.

[0042] Optionally, a first historical flowering quality factor is randomly extracted from the historical flowering quality factor set, the difference between the historical flowering quality factor set and the first historical flowering quality factor is traversed and calculated, and the calculated difference is divided by the first historical flowering quality factor, and then the inverse of the ratio is used as the similarity to obtain a first historical similarity set. The first historical similarity set reflects the similarity between the historical flowering quality factors in the historical flowering quality factor set and the first historical flowering quality factor. The historical flowering quality factors in the first historical similarity set that are greater than or equal to a preset similarity threshold are added to the first clustered historical flowering quality factor set. The preset similarity threshold is a minimum historical similarity value that can be classified into the same set and is pre-set by a person skilled in the art.

[0043] Then, the first clustered historical flowering quality factor set is removed from the historical flowering quality factor set, and a second historical flowering quality factor is randomly selected from the removed historical flowering quality factor set, and combined with the preset similarity threshold, a similar clustering analysis is performed according to the same acquisition principle as that for obtaining the first clustered historical flowering quality factor set to obtain a second clustered historical flowering quality factor set. After multiple similar clustering analyses, until each historical flowering quality factor in the historical flowering quality factor set is divided into a clustered historical flowering quality factor set, the analysis is stopped to obtain the multiple clustered historical flowering quality factor sets.

[0044] Since the historical flowering quality factor corresponds to the historical flowering index one-to-one, and the historical flowering index corresponds to the historical environmental parameter one-to-one, the historical flowering quality factor corresponds to the historical environmental parameter as well. The historical environmental parameter set is mapped and clustered according to the multiple clustered historical flowering quality factor sets to obtain the multiple clustered historical environmental parameter sets. Among them, the multiple clustered historical environmental parameter sets reflect the environmental distribution corresponding to strawberry plants with flowering quality within a certain range. Each clustered historical environmental parameter set contains environmental parameter data that matches the flowering quality factor set of the cluster. Thus, the correlation analysis between the flowering quality factor and the environmental parameter is realized, achieving the technical effect of providing data support for environmental regulation of greenhouses.

[0045] S400: performing parameter centralized search on the plurality of cluster historical environment parameter sets respectively to determine a plurality of cluster centralized environment parameters;

[0046] Further, the plurality of cluster historical environment parameter sets are respectively subjected to parameter centralized search to determine a plurality of cluster centralized environment parameters. Step S400 of the embodiment of the present application further includes:

[0047] Randomly extracting a first cluster historical environment parameter set from the multiple cluster historical environment parameter sets, calculating a parameter mean of the first cluster historical environment parameter set, and obtaining a first cluster historical environment parameter mean;

[0048] Taking the mean of the first cluster historical environment parameter as the centralized search starting point, performing a centralized parameter search in the first cluster historical environment parameter set according to a preset search bandwidth to obtain the first cluster centralized environment parameter;

[0049] Mean calculation is performed on the multiple cluster historical environment parameter sets, and the calculation results are used as concentrated search starting points respectively, and parameter concentrated search is performed in the first cluster historical environment parameter set according to the preset search bandwidth to obtain the multiple cluster concentrated environment parameters.

[0050] Further, such as Figure 2 As shown, taking the mean of the first cluster historical environment parameter as the centralized search starting point, performing a centralized search of parameters in the first cluster historical environment parameter set according to a preset search bandwidth to obtain the first cluster centralized environment parameter, step S400 of the embodiment of the present application further includes:

[0051] Taking the first cluster historical environment parameter mean as a centralized search starting point, moving in the first cluster historical environment parameter set according to the preset search bandwidth to obtain iterative first cluster historical environment parameters;

[0052] Constructing the centralized search starting point and the search neighborhood of the iterative first clustering historical environment parameter based on the preset search bandwidth and the first clustering historical environment parameter set respectively, to obtain a centralized search neighborhood and an iterative search neighborhood;

[0053] Determine whether the neighborhood density of the iterative search neighborhood is greater than or equal to the neighborhood density of the centralized search neighborhood. If so, update the iterative first cluster historical environment parameter to the centralized search starting point, and iterate again in the first cluster historical environment parameter set according to the preset search bandwidth until the preset number of iterations is met, and use the iterative first cluster historical environment parameter obtained in the last iteration as the first cluster centralized environment parameter.

[0054] Furthermore, step S400 in the embodiment of the present application further includes:

[0055] If not, the search direction from the centralized search starting point to the iterative first cluster historical environment parameter is added to the disabled search table, and the mean of the first cluster historical environment parameter is used as the centralized search starting point again, and iteration is performed in the first cluster historical environment parameter set according to the preset search bandwidth until the preset number of iterations is met, and the iterative first cluster historical environment parameter obtained in the last iteration is used as the first cluster centralized environment parameter, wherein the disabled search table includes a preset number of iteration disables, and the preset number of iteration disables is the number of times the search direction stored in the disabled search table is prohibited from iteration.

[0056] In a possible embodiment, after obtaining the multiple clustered historical environmental parameter sets, since there are accidental errors or historical environmental parameters with low frequency in each set, it is necessary to conduct parameter centralized searches for them respectively to determine the multiple clustered centralized environmental parameters that can reflect the common parameter conditions of the multiple clustered historical environmental parameter sets. This achieves the technical effect of providing a reliable control target for subsequent greenhouse environmental control.

[0057] In one embodiment, a first cluster historical environment parameter set is randomly selected from the plurality of cluster historical environment parameter sets, and then a parameter mean of the first cluster historical environment parameter set is calculated to obtain the first cluster historical environment parameter mean, wherein the first cluster historical environment parameter mean reflects the general parameter situation of the first cluster historical environment parameter set in consideration of edge values.

[0058] Then, taking the mean of the first cluster historical environment parameter as the starting point of the centralized search, a centralized parameter search is performed in the first cluster historical environment parameter set according to a preset search bandwidth to obtain the first cluster centralized environment parameter. The preset search bandwidth is the distance of a single move when performing an iterative search in the first cluster historical environment parameter set, that is, the difference of the parameter. The first cluster centralized environment parameter is the environment parameter that best represents the parameter situation of the first cluster historical environment parameter set.

[0059] Preferably, the first cluster historical environment parameter mean is used as the concentrated search starting point, and the first cluster historical environment parameter set is moved according to the preset search bandwidth to obtain the iterative first cluster historical environment parameter. The concentrated search neighborhood is constructed with the concentrated search starting point as the center and the preset search bandwidth as the radius, wherein the difference between the first cluster historical environment parameter in the concentrated search neighborhood and the first cluster historical environment parameter mean is less than or equal to the preset search bandwidth. Furthermore, the iterative search neighborhood is constructed with the iterative first cluster historical environment parameter as the center and the preset search bandwidth as the radius, wherein the first cluster historical environment parameter in the iterative search neighborhood and the iterative first cluster historical environment parameter are less than or equal to the preset search bandwidth.

[0060] The number of parameters in the centralized search neighborhood and the iterative search neighborhood is counted respectively to obtain the centralized search neighborhood parameter quantity and the iterative search neighborhood parameter quantity. The ratio of the centralized search neighborhood parameter quantity to the area of ​​the centralized search neighborhood is used as the neighborhood density of the centralized search neighborhood. The ratio of the iterative search neighborhood parameter quantity to the area of ​​the iterative search neighborhood is used as the neighborhood density of the iterative search neighborhood. Wherein, the neighborhood density reflects the intensive degree of data aggregation in each search neighborhood.

[0061] Furthermore, it is determined whether the neighborhood density of the iterative search neighborhood is greater than or equal to the neighborhood density of the concentrated search neighborhood. If so, it indicates that the iterative first cluster historical environment parameter is more representative than the concentrated search starting point. At this time, the iterative first cluster historical environment parameter is updated to the concentrated search starting point, and iterates again in the first cluster historical environment parameter set according to the preset search bandwidth until the preset number of iterations (the maximum number of iterations pre-set by a person skilled in the art) is met, and the iterative first cluster historical environment parameter obtained in the last iteration is used as the first cluster concentrated environment parameter.

[0062] If not, it indicates that the centralized search starting point is more representative than the iterative first cluster historical environment parameter, and the search direction from the centralized search starting point to the iterative first cluster historical environment parameter is added to the disabled search table. The disabled search table includes a preset number of iterative disabled times, which is the number of times the search direction stored in the disabled search table is prohibited from iterating.

[0063] That is, the direction of obtaining the iterative first cluster historical environment parameter will not be used within the preset number of disabled iterations, thereby improving the globality and efficiency of the centralized search. Then, the mean value of the first cluster historical environment parameter is used as the centralized search starting point again, and iteration is performed in the first cluster historical environment parameter set according to the preset search bandwidth until the preset number of iterations is met, and the iterative first cluster historical environment parameter obtained in the last iteration is used as the first cluster centralized environment parameter.

[0064] Based on the same principle as that of obtaining the first cluster centralized environmental parameter, mean calculation is performed on the multiple cluster historical environmental parameter sets, and the calculation results are used as centralized search starting points respectively, and centralized parameter search is performed in the first cluster historical environmental parameter set according to the preset search bandwidth to obtain the multiple cluster centralized environmental parameters.

[0065] By obtaining the environmental parameters of the multiple clusters, the general conditions of the environmental parameters corresponding to different sets of flowering quality factors are determined, providing target environmental parameters for subsequent greenhouse environmental control, thereby achieving the technical effect of providing a basis for determining reliable environmental control parameters.

[0066] S500: configuring greenhouse environmental control parameters according to the plurality of clustered environmental parameters to obtain a plurality of target greenhouse environmental control parameters;

[0067] S600: regulating the devices of multiple test greenhouses based on the multiple target greenhouse environmental control parameters, and conducting a strawberry flowering impact analysis test on the target strawberries in the multiple test greenhouses after the control is completed.

[0068] Further, greenhouse environmental control parameters are configured according to the multiple clustered environmental parameters to obtain multiple target greenhouse environmental control parameters. Step S500 of the embodiment of the present application further includes:

[0069] Retrieve a historical greenhouse environment control log set by using the environmental parameters in the multiple cluster sets as indexes, and obtain multiple historical greenhouse environment control parameter sets and multiple historical greenhouse environment control matching degree sets;

[0070] Traversing the plurality of historical greenhouse environment control matching degree sets to extract the maximum value of the matching value, obtaining the plurality of maximum values ​​of the historical greenhouse environment control matching degrees, and using the historical greenhouse environment control parameters corresponding to the plurality of maximum values ​​of the historical greenhouse environment control matching degrees as the historical greenhouse environment control parameters in multiple directions;

[0071] Taking the multiple directional historical greenhouse environmental control parameters as multiple optimization directions, performing multi-scale directional optimization on the multiple historical greenhouse environmental control parameter sets according to multiple directional optimization scale sets, and obtaining multiple follow-up optimized historical greenhouse environmental control parameter sets;

[0072] Performing control matching degree analysis on the multiple follow-up optimization historical greenhouse environment control parameter sets to obtain multiple follow-up optimization historical greenhouse environment control matching degree sets;

[0073] Compare the multiple follow-up optimization historical greenhouse environment control matching degree sets with the multiple historical greenhouse environment control matching degree maximum values, and when there is a follow-up optimization historical greenhouse environment control matching degree greater than the corresponding multiple historical greenhouse environment control matching degree maximum values, update the multiple follow-up optimization historical greenhouse environment control parameters corresponding to the maximum values ​​in the multiple follow-up optimization historical greenhouse environment control matching degree sets to multiple optimization directions;

[0074] According to the updated multiple optimization directions, multi-scale directional optimization is performed on the multiple sets of follow-optimized historical greenhouse environmental control parameters. After multiple optimizations, until the preset number of optimizations is met, the multiple target greenhouse environmental control parameters are obtained.

[0075] Furthermore, step S500 of the embodiment of the present application further includes:

[0076] The differences between the multiple historical greenhouse environment control matching degree sets and the multiple historical greenhouse environment control matching degree maximum values ​​are calculated respectively, and the calculated results are compared with the multiple historical greenhouse environment control matching degree maximum values, and the product of the ratio and the preset adjustment scale is used as the multiple directional optimization scale sets.

[0077] In a possible embodiment, after obtaining the multiple clustered centralized environmental parameters, the environmental control devices in the multiple test greenhouses are subjected to control parameter analysis to determine multiple target greenhouse environmental control parameters. Wherein, each clustered centralized environmental parameter includes strawberry growth environmental parameters such as temperature, humidity, and sunshine duration. Each target greenhouse environmental control parameter is used to adjust the parameters of the environmental control device in a test greenhouse so that the environment in the test greenhouse can meet the requirements of the corresponding clustered centralized environmental parameters. Preferably, a target greenhouse environmental control parameter includes temperature setting value, power output value, heating time, sunshine time, humidity setting value, air volume, wind speed, light intensity, light wavelength, carbon dioxide gas injection flow rate, etc. Optionally, the environmental control device in each test greenhouse includes heating equipment (such as heater, hot water pipe, electric heater, etc.), cooling equipment (fan, sunshade net, atomization cooling system, etc.), humidification equipment (atomizer, spray system), dehumidification equipment (dehumidifier, air conditioner, ventilation system), light control equipment (LED plant growth lamp, fluorescent lamp, sunshade net) and other devices.

[0078] Preferably, the parameters of the corresponding devices in the multiple test greenhouses are adjusted based on the multiple target greenhouse environmental control parameters, and the devices are started, so that the environments of the multiple test greenhouses meet the requirements of the multiple clustered environmental parameters after operation, and then the strawberry flowering influence analysis test is performed on the target strawberries in the multiple test greenhouses after the control is completed. The technical effect of improving the accuracy of greenhouse environmental control and improving the reliability of control is achieved.

[0079] In one embodiment, the environmental parameters in the multiple clusters are used as indexes to retrieve the historical greenhouse environmental control log sets, and multiple historical greenhouse environmental control parameter sets and multiple historical greenhouse environmental control matching degree sets are obtained. The historical greenhouse environmental control log sets are control records of greenhouse environmental control in historical time. Each historical greenhouse environmental control matching degree reflects the similarity between the environmental parameters that can be achieved after greenhouse control and the corresponding cluster environmental parameters. The higher the historical greenhouse environmental control matching degree, the higher the quality of the corresponding historical greenhouse environmental control parameters.

[0080] Therefore, the maximum values ​​of the matching values ​​in the multiple historical greenhouse environment control matching degree sets are extracted respectively to obtain multiple maximum values ​​of the historical greenhouse environment control matching degrees. The historical greenhouse environment control parameters corresponding to the multiple maximum values ​​of the historical greenhouse environment control matching degrees are used as the historical greenhouse environment control parameters in multiple directions. Among them, the historical greenhouse environment control parameters in multiple directions are the optimal greenhouse environment control parameters of the environmental parameters in each cluster set in the historical time, so they are used as the adjustment optimization direction.

[0081] Optionally, the multiple directional historical greenhouse environment control parameters are used as multiple optimization directions, and the multiple historical greenhouse environment control parameter sets are directional optimized according to different scales according to multiple directional optimization scale sets. That is, according to the degree of difference between the multiple historical greenhouse environment control parameter sets and the corresponding multiple directional historical greenhouse environment control parameters, the multiple historical greenhouse environment control parameter sets are adjusted with different adjustment ranges, so as to obtain multiple follow-up optimized historical greenhouse environment control parameter sets. In other words, the historical greenhouse environment control parameter set is adjusted to the temporary optimal parameter to obtain more greenhouse environment control parameters, so as to provide more data for obtaining the optimal solution.

[0082] Preferably, the difference between the multiple historical greenhouse environment control matching degree sets and the multiple historical greenhouse environment control matching degree maximum values ​​are calculated respectively, and the calculated result is compared with the multiple historical greenhouse environment control matching degree maximum values, and the product of the ratio and the preset adjustment scale (the single parameter adjustment range preset by technicians in this field) is used as the multiple directional optimization scale sets. In other words, the smaller the difference between the historical greenhouse environment control parameters and the corresponding optimization direction, the smaller the corresponding directional optimization scale.

[0083] Furthermore, the matching degree prediction network layer is used to perform control matching degree analysis on the multiple sets of follow-up optimized historical greenhouse environment control parameters to obtain multiple sets of follow-up optimized historical greenhouse environment control matching degrees. Preferably, multiple sample greenhouse environment control parameters and multiple sample greenhouse environment control matching degrees are obtained as training data, and the training data is used to perform supervised training on a framework built based on a feedforward neural network to learn the mapping relationship between greenhouse environment control parameters and greenhouse environment control matching degrees, until the training converges, and the matching degree prediction network layer that has been trained is obtained.

[0084] Compare the sizes of the multiple follow-up optimization historical greenhouse environment control matching degree sets with the maximum values ​​of the multiple historical greenhouse environment control matching degrees. When there is a follow-up optimization historical greenhouse environment control matching degree greater than the corresponding maximum values ​​of the multiple historical greenhouse environment control matching degrees, it indicates that a better solution than the multiple directional historical greenhouse environment control parameters has appeared in the multiple follow-up optimization historical greenhouse environment control matching degree sets after directional optimization. At this time, the multiple follow-up optimization historical greenhouse environment control parameters corresponding to the maximum values ​​in the multiple follow-up optimization historical greenhouse environment control matching degree sets are updated to multiple optimization directions.

[0085] According to the updated multiple optimization directions, multi-scale directional optimization is performed on the multiple sets of follow-optimized historical greenhouse environmental control parameters. After multiple optimizations, until the preset number of optimizations is met, the multiple target greenhouse environmental control parameters are obtained.

[0086] Optionally, when there is no follow-up optimized historical greenhouse environment control matching degree greater than the corresponding maximum value of multiple historical greenhouse environment control matching degrees, the current multiple optimization directions are still maintained, and the multiple follow-up optimized historical greenhouse environment control matching degree sets are continuously directionally optimized according to multiple directional optimization scale sets until the preset number of optimization times is met, and the multiple follow-up optimized historical greenhouse environment control parameters corresponding to the multiple maximum values ​​of the follow-up optimized greenhouse environment control matching degrees during the optimization process are used as the multiple target greenhouse environment control parameters.

[0087] Through multiple rounds of optimization, directional adjustment and matching analysis, it can be ensured that the environment of each test greenhouse can reach the best state and effectively meet the requirements of the corresponding multiple clustered environmental parameters. The technical effect of effectively regulating the environmental conditions of each test greenhouse has been achieved.

[0088] In summary, the embodiments of the present application have at least the following technical effects:

[0089] This application evaluates the flowering quality of the historical flowering index set, then performs similar clustering analysis on the historical flowering quality factor set, and maps and clusters the historical environmental parameter set according to the respective results to obtain multiple clustered historical environmental parameter sets, and then performs parameter centralized search on the multiple clustered historical environmental parameter sets to determine multiple clustered centralized environmental parameters, thereby achieving the determination of the environmental conditions that need to be achieved in different experimental greenhouses, and then configures greenhouse environmental control parameters according to the multiple clustered centralized environmental parameters to obtain multiple target greenhouse environmental control parameters, and then, controls the devices of multiple experimental greenhouses based on the multiple target greenhouse environmental control parameters, and performs strawberry flowering impact analysis tests on target strawberries in the multiple experimental greenhouses after the control is completed. The technical effect of optimizing the configuration of environmental control parameters for the experimental greenhouse and improving the accuracy of greenhouse environmental control is achieved.

[0090] Embodiment 2, based on the same inventive concept as the greenhouse environment control method for strawberry flowering analysis in the previous embodiment, Figure 3 As shown, the present application provides a greenhouse environment control system for strawberry flowering influence analysis, and the system and method embodiments in the present application embodiments are based on the same inventive concept. The system includes:

[0091] The historical environmental parameter set acquisition module 11 is used to obtain the basic variety information of the target strawberry, take the preset flowering index as the retrieval target, and perform big data retrieval with the basic variety information as the index to obtain the historical flowering index set and the historical environmental parameter set, wherein each historical flowering index corresponds to a historical environmental parameter set;

[0092] A historical flowering quality factor set obtaining module 12 is used to evaluate the flowering quality based on the historical flowering indicator set to obtain a historical flowering quality factor set;

[0093] A clustering historical environment parameter set obtaining module 13 is used to traverse the historical flowering quality factor set to perform similar clustering analysis, determine multiple clustering historical flowering quality factor sets, and map and cluster the historical environment parameter set based on the multiple clustering historical flowering quality factor sets to obtain multiple clustering historical environment parameter sets;

[0094] A cluster centralized environment parameter determination module 14 is used to perform a centralized parameter search on the plurality of cluster historical environment parameter sets respectively to determine a plurality of cluster centralized environment parameters;

[0095] A greenhouse environment control parameter acquisition module 15 is used to configure greenhouse environment control parameters according to the multiple clustered environmental parameters to obtain multiple target greenhouse environment control parameters;

[0096] The environmental control module 16 is used to control the devices of multiple test greenhouses based on the multiple target greenhouse environmental control parameters, and perform strawberry flowering influence analysis tests on the target strawberries in the multiple test greenhouses after the control is completed.

[0097] Furthermore, the clustering centralized environment parameter determination module 14 is used to perform the following steps:

[0098] Randomly extracting a first cluster historical environment parameter set from the multiple cluster historical environment parameter sets, calculating a parameter mean of the first cluster historical environment parameter set, and obtaining a first cluster historical environment parameter mean;

[0099] Taking the mean of the first cluster historical environment parameter as the centralized search starting point, performing a centralized parameter search in the first cluster historical environment parameter set according to a preset search bandwidth to obtain the first cluster centralized environment parameter;

[0100] Mean calculation is performed on the multiple cluster historical environment parameter sets, and the calculation results are used as concentrated search starting points respectively, and parameter concentrated search is performed in the first cluster historical environment parameter set according to the preset search bandwidth to obtain the multiple cluster concentrated environment parameters.

[0101] Furthermore, the clustering centralized environment parameter determination module 14 is used to perform the following steps:

[0102] Taking the first cluster historical environment parameter mean as a centralized search starting point, moving in the first cluster historical environment parameter set according to the preset search bandwidth to obtain iterative first cluster historical environment parameters;

[0103] Constructing the centralized search starting point and the search neighborhood of the iterative first clustering historical environment parameter based on the preset search bandwidth and the first clustering historical environment parameter set respectively, to obtain a centralized search neighborhood and an iterative search neighborhood;

[0104] Determine whether the neighborhood density of the iterative search neighborhood is greater than or equal to the neighborhood density of the centralized search neighborhood. If so, update the iterative first cluster historical environment parameter to the centralized search starting point, and iterate again in the first cluster historical environment parameter set according to the preset search bandwidth until the preset number of iterations is met, and use the iterative first cluster historical environment parameter obtained in the last iteration as the first cluster centralized environment parameter.

[0105] Furthermore, the clustering centralized environment parameter determination module 14 is used to perform the following steps:

[0106] If not, the search direction from the centralized search starting point to the iterative first cluster historical environment parameter is added to the disabled search table, and the mean of the first cluster historical environment parameter is used as the centralized search starting point again, and iteration is performed in the first cluster historical environment parameter set according to the preset search bandwidth until the preset number of iterations is met, and the iterative first cluster historical environment parameter obtained in the last iteration is used as the first cluster centralized environment parameter, wherein the disabled search table includes a preset number of iteration disables, and the preset number of iteration disables is the number of times the search direction stored in the disabled search table is prohibited from iteration.

[0107] Furthermore, the greenhouse environment control parameter acquisition module 15 is used to perform the following steps:

[0108] Retrieve a historical greenhouse environment control log set by using the environmental parameters in the multiple cluster sets as indexes, and obtain multiple historical greenhouse environment control parameter sets and multiple historical greenhouse environment control matching degree sets;

[0109] Traversing the plurality of historical greenhouse environment control matching degree sets to extract the maximum value of the matching value, obtaining the plurality of maximum values ​​of the historical greenhouse environment control matching degrees, and using the historical greenhouse environment control parameters corresponding to the plurality of maximum values ​​of the historical greenhouse environment control matching degrees as the historical greenhouse environment control parameters in multiple directions;

[0110] Taking the multiple directional historical greenhouse environmental control parameters as multiple optimization directions, performing multi-scale directional optimization on the multiple historical greenhouse environmental control parameter sets according to multiple directional optimization scale sets, and obtaining multiple follow-up optimized historical greenhouse environmental control parameter sets;

[0111] Performing control matching degree analysis on the multiple follow-up optimization historical greenhouse environment control parameter sets to obtain multiple follow-up optimization historical greenhouse environment control matching degree sets;

[0112] Compare the multiple follow-up optimization historical greenhouse environment control matching degree sets with the multiple historical greenhouse environment control matching degree maximum values, and when there is a follow-up optimization historical greenhouse environment control matching degree greater than the corresponding multiple historical greenhouse environment control matching degree maximum values, update the multiple follow-up optimization historical greenhouse environment control parameters corresponding to the maximum values ​​in the multiple follow-up optimization historical greenhouse environment control matching degree sets to multiple optimization directions;

[0113] According to the updated multiple optimization directions, multi-scale directional optimization is performed on the multiple sets of follow-optimized historical greenhouse environmental control parameters. After multiple optimizations, until the preset number of optimizations is met, the multiple target greenhouse environmental control parameters are obtained.

[0114] Furthermore, the greenhouse environment control parameter acquisition module 15 is used to perform the following steps:

[0115] The differences between the multiple historical greenhouse environment control matching degree sets and the multiple historical greenhouse environment control matching degree maximum values ​​are calculated respectively, and the calculated results are compared with the multiple historical greenhouse environment control matching degree maximum values, and the product of the ratio and the preset adjustment scale is used as the multiple directional optimization scale sets.

[0116] Furthermore, the historical flower quality factor set acquisition module 12 is used to perform the following steps:

[0117] Acquire multiple sample flowering index sets and multiple sample flowering quality factor sets as training sample data;

[0118] The training sample data is divided into a training set and a validation set according to a preset ratio, the training set is used to train a framework built based on a convolutional neural network, the mapping relationship between a flowering quality indicator set and a flowering quality factor is learned, and the validation set is used to supervise the training results until the training converges, thereby obtaining a trained quality evaluation network layer;

[0119] The quality evaluation network layer is used to analyze the historical flowering indicator set to obtain the historical flowering quality factor set. It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-tasking and parallel processing are also possible or may be advantageous.

[0120] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0121] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A greenhouse environment control method for strawberry flowering influence analysis, characterized in that: The method comprises: Obtain basic variety information of target strawberries, take a preset flowering index as a search target, perform big data search with the basic variety information as an index, and obtain a historical flowering index set and a historical environmental parameter set, wherein each historical flowering index corresponds to a historical environmental parameter set; Performing a flowering quality evaluation based on the historical flowering indicator set to obtain a historical flowering quality factor set; Traversing the historical flowering quality factor sets to perform similar clustering analysis, determine multiple clustered historical flowering quality factor sets, and map and cluster the historical environment parameter sets based on the multiple clustered historical flowering quality factor sets to obtain multiple clustered historical environment parameter sets; Performing parameter centralized search on the multiple cluster historical environment parameter sets respectively to determine multiple cluster centralized environment parameters; Configure greenhouse environmental control parameters according to the multiple clustered environmental parameters to obtain multiple target greenhouse environmental control parameters; The devices of the multiple test greenhouses are regulated based on the multiple target greenhouse environmental control parameters, and a strawberry flowering impact analysis test is performed on the target strawberries in the multiple test greenhouses after the regulation is completed.

2. The greenhouse environment control method for strawberry flowering impact analysis according to claim 1, characterized in that: Performing parameter centralized search on the multiple cluster historical environment parameter sets respectively to determine multiple cluster centralized environment parameters, including: Randomly extracting a first cluster historical environment parameter set from the multiple cluster historical environment parameter sets, calculating a parameter mean of the first cluster historical environment parameter set, and obtaining a first cluster historical environment parameter mean; Taking the mean of the first cluster historical environment parameter as the centralized search starting point, performing a centralized parameter search in the first cluster historical environment parameter set according to a preset search bandwidth to obtain the first cluster centralized environment parameter; Mean calculation is performed on the multiple cluster historical environment parameter sets, and the calculation results are used as concentrated search starting points respectively, and parameter concentrated search is performed in the first cluster historical environment parameter set according to the preset search bandwidth to obtain the multiple cluster concentrated environment parameters.

3. The greenhouse environment control method for strawberry flowering impact analysis as claimed in claim 2, characterized in that: Taking the first cluster historical environment parameter mean as the centralized search starting point, performing a centralized search in the first cluster historical environment parameter set according to a preset search bandwidth to obtain the first cluster centralized environment parameter, including: Taking the first cluster historical environment parameter mean as a centralized search starting point, moving in the first cluster historical environment parameter set according to the preset search bandwidth to obtain iterative first cluster historical environment parameters; Constructing the centralized search starting point and the search neighborhood of the iterative first clustering historical environment parameter based on the preset search bandwidth and the first clustering historical environment parameter set respectively, to obtain a centralized search neighborhood and an iterative search neighborhood; Determine whether the neighborhood density of the iterative search neighborhood is greater than or equal to the neighborhood density of the centralized search neighborhood. If so, update the iterative first cluster historical environment parameter to the centralized search starting point, and iterate again in the first cluster historical environment parameter set according to the preset search bandwidth until the preset number of iterations is met, and use the iterative first cluster historical environment parameter obtained in the last iteration as the first cluster centralized environment parameter.

4. The greenhouse environment control method for strawberry flowering impact analysis according to claim 3, characterized in that: include: If not, the search direction from the centralized search starting point to the iterative first cluster historical environment parameter is added to the disabled search table, and the mean of the first cluster historical environment parameter is used as the centralized search starting point again, and iteration is performed in the first cluster historical environment parameter set according to the preset search bandwidth until the preset number of iterations is met, and the iterative first cluster historical environment parameter obtained in the last iteration is used as the first cluster centralized environment parameter, wherein the disabled search table includes a preset number of iteration disables, and the preset number of iteration disables is the number of times the search direction stored in the disabled search table is prohibited from iteration.

5. The greenhouse environment control method for strawberry flowering impact analysis according to claim 1, characterized in that: The greenhouse environment control parameters are configured according to the multiple clustered environmental parameters to obtain multiple target greenhouse environment control parameters, including: Retrieve a historical greenhouse environment control log set by using the environmental parameters in the multiple cluster sets as indexes, and obtain multiple historical greenhouse environment control parameter sets and multiple historical greenhouse environment control matching degree sets; Traversing the plurality of historical greenhouse environment control matching degree sets to extract the maximum value of the matching value, obtaining the plurality of maximum values ​​of the historical greenhouse environment control matching degrees, and using the historical greenhouse environment control parameters corresponding to the plurality of maximum values ​​of the historical greenhouse environment control matching degrees as the historical greenhouse environment control parameters in multiple directions; Taking the multiple directional historical greenhouse environmental control parameters as multiple optimization directions, performing multi-scale directional optimization on the multiple historical greenhouse environmental control parameter sets according to multiple directional optimization scale sets, and obtaining multiple follow-up optimized historical greenhouse environmental control parameter sets; Performing control matching degree analysis on the multiple follow-up optimization historical greenhouse environment control parameter sets to obtain multiple follow-up optimization historical greenhouse environment control matching degree sets; Compare the multiple follow-up optimization historical greenhouse environment control matching degree sets with the multiple historical greenhouse environment control matching degree maximum values, and when there is a follow-up optimization historical greenhouse environment control matching degree greater than the corresponding multiple historical greenhouse environment control matching degree maximum values, update the multiple follow-up optimization historical greenhouse environment control parameters corresponding to the maximum values ​​in the multiple follow-up optimization historical greenhouse environment control matching degree sets to multiple optimization directions; According to the updated multiple optimization directions, multi-scale directional optimization is performed on the multiple sets of follow-optimized historical greenhouse environmental control parameters. After multiple optimizations, until the preset number of optimizations is met, the multiple target greenhouse environmental control parameters are obtained.

6. The greenhouse environment control method for strawberry flowering impact analysis according to claim 5, characterized in that: The differences between the multiple historical greenhouse environment control matching degree sets and the multiple historical greenhouse environment control matching degree maximum values ​​are calculated respectively, and the calculated results are compared with the multiple historical greenhouse environment control matching degree maximum values, and the product of the ratio and the preset adjustment scale is used as the multiple directional optimization scale sets.

7. The greenhouse environment control method for strawberry flowering impact analysis according to claim 1, characterized in that: The flowering quality evaluation is performed based on the historical flowering index set to obtain a historical flowering quality factor set, including: Acquire multiple sample flowering index sets and multiple sample flowering quality factor sets as training sample data; The training sample data is divided into a training set and a validation set according to a preset ratio, the training set is used to train a framework built based on a convolutional neural network, the mapping relationship between a flowering quality indicator set and a flowering quality factor is learned, and the validation set is used to supervise the training results until the training converges, thereby obtaining a trained quality evaluation network layer; The quality evaluation network layer is used to analyze the historical flowering indicator set to obtain the historical flowering quality factor set.

8. A greenhouse environment control system for strawberry flowering influence analysis, characterized in that: The system is used to execute the greenhouse environment control method for strawberry flowering impact analysis according to any one of claims 1 to 7, and the system comprises: A historical environmental parameter set acquisition module is used to obtain basic variety information of target strawberries, take a preset flowering index as a search target, and perform big data search with the basic variety information as an index to obtain a historical flowering index set and a historical environmental parameter set, wherein each historical flowering index corresponds to a historical environmental parameter set; A historical flowering quality factor set acquisition module, used to evaluate the flowering quality based on the historical flowering indicator set to obtain a historical flowering quality factor set; a clustering historical environment parameter set acquisition module, used for traversing the historical flowering quality factor set to perform similar clustering analysis, determine multiple clustering historical flowering quality factor sets, and map and cluster the historical environment parameter set based on the multiple clustering historical flowering quality factor sets to obtain multiple clustering historical environment parameter sets; A cluster centralized environmental parameter determination module, used to perform a centralized parameter search on the multiple cluster historical environmental parameter sets respectively to determine multiple cluster centralized environmental parameters; A greenhouse environment control parameter acquisition module is used to configure greenhouse environment control parameters according to the multiple clustered environmental parameters to obtain multiple target greenhouse environment control parameters; The environmental control module is used to control the devices of multiple test greenhouses based on the multiple target greenhouse environmental control parameters, and to perform strawberry flowering influence analysis tests on the target strawberries in the multiple test greenhouses after the control is completed.

Images