Greenhouse environment regulation method and system for strawberry flowering effect analysis

Abstract

The application discloses a greenhouse environment regulation method and system for strawberry flowering influence analysis, and relates to the technical field of greenhouse environment regulation.The method comprises the following steps: obtaining basic variety information of target strawberries, performing big data retrieval, obtaining a historical flowering index set and a historical environment parameter set;obtaining a historical flowering quality factor set;obtaining a plurality of clustered historical environment parameter sets;determining a plurality of clustered environment parameters;obtaining a plurality of target greenhouse environment regulation parameters;regulating the environment of a plurality of test greenhouses; and performing strawberry flowering influence analysis tests on the target strawberries in the plurality of test greenhouses after the regulation is completed.The application solves the technical problems of low accuracy of greenhouse environment regulation and poor reliability of tests in the prior art during strawberry flowering influence analysis, and achieves the technical effects of determining reliable greenhouse environment regulation parameters and improving the accuracy of greenhouse environment regulation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of greenhouse environment regulation, and particularly relates to a greenhouse environment regulation method and system for analysis of strawberry flowering influence. BACKGROUND

[0002] At present, greenhouse environment regulation often relies on manual setting and experience data, which leads to low accuracy of environment parameter regulation. Taking temperature and humidity as examples, the traditional method often adjusts the parameters of the device in the greenhouse according to experience, however, the adjusted device cannot accurately meet the requirements of temperature and humidity and needs to be adjusted repeatedly for many times, which affects the test progress. Due to the inaccurate regulation of the environment parameters of the greenhouse, and the interference of external uncontrollable factors in the growth process of strawberries, the test results are inconsistent and non-reproducible. For example, in some tests, even if the same strawberry variety and planting environment requirements are met, the deviation of the greenhouse environment regulation may lead to the deviation of strawberry flowering, which increases the error and uncertainty of the research results.

[0003] The prior art has the technical problems of low accuracy of greenhouse environment regulation and poor reliability of test when analyzing the influence of strawberry flowering. SUMMARY

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

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

[0006] In a first aspect, the present application provides a greenhouse environment regulation method for analysis of strawberry flowering influence, which comprises:

[0007] The basic variety information of the target strawberry is acquired, a preset flowering index is taken as a retrieval target, big data retrieval is performed by taking the basic variety information as an index, a historical flowering index set and a historical environment parameter set are obtained, wherein each historical flowering index corresponds to a historical environment parameter set; flowering quality evaluation is performed based on the historical flowering index set, a historical flowering quality factor set is obtained; same type clustering analysis is performed by traversing the historical flowering quality factor set, a plurality of clustered historical flowering quality factor sets are determined, and the historical environment parameter set is mapped and clustered based on the plurality of clustered historical flowering quality factor sets, a plurality of clustered historical environment parameter sets are obtained; parameter set searching is respectively performed on the plurality of clustered historical environment parameter sets, a plurality of clustered set environment parameters are determined; greenhouse environment regulation parameter configuration is performed according to the plurality of clustered set environment parameters, a plurality of target greenhouse environment regulation parameters are obtained; the devices of a plurality of test greenhouse are regulated based on the plurality of target greenhouse environment regulation parameters, and strawberry flowering influence analysis tests are performed on the target strawberry in the plurality of test greenhouses after the regulation is completed.

[0008] In a second aspect of the present application, a greenhouse environment regulation system for strawberry flowering influence analysis is provided, and the system comprises:

[0009] The historical environment parameter set acquisition module is configured to acquire the basic variety information of the target strawberry, take a preset flowering index as a retrieval target, perform big data retrieval by taking the basic variety information as an index, and obtain a historical flowering index set and a historical environment parameter set, wherein each historical flowering index corresponds to a historical environment parameter set.

[0010] The historical flowering quality factor set acquisition module is configured to perform flowering quality evaluation based on the historical flowering index set, and obtain a historical flowering quality factor set.

[0011] The clustered historical environment parameter set acquisition module is configured to perform same type clustering analysis by traversing the historical flowering quality factor set, determine a plurality of clustered historical flowering quality factor sets, and perform mapping and clustering on the historical environment parameter set based on the plurality of clustered historical flowering quality factor sets, and obtain a plurality of clustered historical environment parameter sets.

[0012] The clustered set environment parameter determination module is configured to perform parameter set searching on the plurality of clustered historical environment parameter sets respectively, and determine a plurality of clustered set environment parameters.

[0013] The greenhouse environment regulation parameter acquisition module is configured to perform greenhouse environment regulation parameter configuration according to the plurality of clustered set environment parameters, and obtain a plurality of target greenhouse environment regulation parameters.

[0014] An environment regulation module is configured to regulate the devices of the plurality of test greenhouse based on the plurality of target greenhouse environment regulation parameters, and perform strawberry flowering influence analysis test on the target strawberries in the plurality of test greenhouses after the regulation is completed.

[0015] The one or more technical solutions provided in the present application have at least the following technical effects or advantages:

[0016] In the present application, the basic variety information of the target strawberries is acquired, the preset flowering index is taken as the retrieval target, and big data retrieval is performed with the basic variety information as the index to obtain a historical flowering index set and a historical environment parameter set, wherein each historical flowering index corresponds to a historical environment parameter set. Then, the flowering quality evaluation is performed based on the historical flowering index set to obtain a historical flowering quality factor set. Furthermore, the same type clustering analysis is performed on the historical flowering quality factor set to determine a plurality of clustered historical flowering quality factor sets. The historical environment parameter set is mapped and clustered based on the plurality of clustered historical flowering quality factor sets to obtain a plurality of clustered historical environment parameter sets. The parameter set search is performed on the plurality of clustered historical environment parameter sets respectively to determine a plurality of clustered environment parameters. Then, the greenhouse environment regulation parameter configuration is performed according to the plurality of clustered environment parameters to obtain a plurality of target greenhouse environment regulation parameters. Furthermore, the devices of the plurality of test greenhouses are regulated based on the plurality of target greenhouse environment regulation parameters, and the target strawberries are subjected to strawberry flowering influence analysis test in the plurality of test greenhouses after the regulation is completed. The technical effect of providing reliable environment parameters for optimizing the configuration of the environment regulation parameters of the test greenhouses and improving the regulation accuracy is achieved. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative effort.

[0018] Figure 1 The greenhouse environment regulation method flowchart for strawberry flowering influence analysis provided in the embodiments of the present application is shown in the figure.

[0019] Figure 2 The flowchart for obtaining the first clustered environment parameters in the greenhouse environment regulation method for strawberry flowering influence analysis provided in the embodiments of the present application is shown in the figure.

[0020] Figure 3 The greenhouse environment regulation system structure diagram for strawberry flowering influence analysis provided in the embodiments of the present application is shown in the figure.

[0021] The reference signs are explained as follows: a historical environment parameter set obtaining module 11, a historical flowering quality factor set obtaining module 12, a clustered historical environment parameter set obtaining module 13, a clustered concentrated environment parameter determining module 14, a greenhouse environment regulation parameter obtaining module 15, and an environment regulation module 16. DETAILED DESCRIPTION

[0022] The present application provides a greenhouse environment regulation method and system for strawberry flowering influence analysis, which is used to solve the technical problems of low accuracy of greenhouse environment regulation and poor reliability of test in the prior art when analyzing the influence of strawberry flowering.

[0023] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0024] It should be noted that the terms "comprising" and "having" are intended to cover non-exclusive inclusion, for example, a process, method, system, product or server including a series of steps or units does not have to be limited to those clearly listed steps or units, but can include other steps or modules that are not clearly listed or inherent to the process, method, product or device.

[0025] In one embodiment, as shown in the accompanying drawings, the present application provides a greenhouse environment regulation method for strawberry flowering influence analysis, wherein the method comprises: Figure 1

[0026] S100: obtaining basic variety information of target strawberries, taking preset flowering indicators as retrieval targets, and performing big data retrieval on the basic variety information to obtain a historical flowering indicator set and a historical environment parameter set, wherein each historical flowering indicator corresponds to a historical environment parameter set;

[0027] In one possible embodiment, the target strawberries are any type of strawberries that need to be analyzed for flowering influence, which can be red strawberry. The basic variety information is 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 indicators are a group of key indicators for measuring the flowering state of strawberries during growth, including flowering time, inflorescence number, flower number, petal integrity and flowering synchrony.

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

[0029] The retrieved set of historical flowering indexes and corresponding set of historical environmental parameters are stored and sorted as basic data for subsequent analysis. Each set of historical flowering indexes and its corresponding set of historical environmental parameters 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: Based on the set of historical flowering indexes, a set of historical flowering quality factors is obtained by evaluating the flowering quality.

[0031] Further, based on the set of historical flowering indexes, a set of historical flowering quality factors is obtained by evaluating the flowering quality. The step S200 of the embodiment of the application further includes:

[0032] A plurality of sample sets of flowering indexes and a plurality of sample sets of flowering quality factors are obtained 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 based on a convolutional neural network, learn the mapping relationship between the set of flowering quality indexes and the set of flowering quality factors, and supervise the training result using the validation set until the training converges, obtaining a trained quality evaluation network layer.

[0034] The set of historical flowering indexes is analyzed using the quality evaluation network layer to obtain the set of historical flowering quality factors.

[0035] In one embodiment, the set of historical flowering indexes reflects the flowering of strawberries under different environmental conditions. The set of historical flowering indexes is comprehensively analyzed to determine the flowering quality and generate the set of historical flowering quality factors. Each historical flowering quality factor reflects the flowering quality of strawberries under a historical environmental parameter. The higher the historical flowering quality factor, the better the corresponding flowering quality of strawberries.

[0036] In an embodiment, a plurality of sample flowering indicator sets and a plurality of sample flowering quality factor sets are obtained as training sample data. In order to construct the 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 part (e.g., 20%), and is used to supervise and verify the training effect of the model. Thus, the network layer overfitting is avoided, the performance of the network layer on new data is evaluated through the validation set, and the generalization ability of the model is ensured.

[0037] Preferably, the framework based on a convolutional neural network (CNN) is trained using the training set to automatically learn the complex mapping relationship between the flowering quality factors and the flowering indicators. The convolutional neural network is a deep learning method suitable for processing input data with spatial structure (such as image or time series data), and can extract high-order features of the data through multiple levels of convolution and pooling operations. During the training process, the CNN automatically adjusts the weights and bias parameters in the network to minimize the prediction error using the data in the training set. The input of the network is the flowering indicator set, 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 performance of the model is evaluated using the validation set data to check whether the model is overfitting or underfitting. If the error on the validation set continues to decrease, and the training set and validation set errors are close, it indicates that the model has converged and the training can be stopped. After training and verification, the CNN model learns the mapping relationship between the flowering indicators and the flowering quality factors. After the training is completed, the quality evaluation network layer of the model has been optimized to the best state, and can efficiently and accurately analyze and evaluate the strawberry flowering indicators.

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

[0040] S300: performing same-class clustering analysis on the historical flowering quality factor set to determine a plurality of clustered historical flowering quality factor sets, and performing mapping clustering on the historical environmental parameter set based on the plurality of clustered historical flowering quality factor sets to obtain a plurality of clustered historical environmental parameter sets;

[0041] In one possible embodiment, the clustering analysis is an unsupervised learning algorithm aiming to divide samples into multiple clusters according to their similarities. In this step, the historical flowering quality factors are grouped using a suitable clustering algorithm such as K-means, DBSCAN or hierarchical clustering. Through clustering analysis, the set of flowering quality factors is divided into different sets according to similarities, and each set represents a class of similar flowering quality factor data, which helps to understand the environmental conditions corresponding to different strawberry flowering quality.

[0042] Optionally, a first historical flowering quality factor is randomly selected from the set of historical flowering quality factors, the differences between the set of historical flowering quality factors and the first historical flowering quality factor are calculated, the calculated differences are divided by the first historical flowering quality factor, and the reciprocal of the ratio is taken 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 set of historical flowering quality factors 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 value are added to a first clustered historical flowering quality factor set. The preset similarity threshold value is the minimum historical similarity value that can be divided into the same set and is set by a person skilled in the art in advance.

[0043] Further, the first clustered historical flowering quality factor set is removed from the set of historical flowering quality factors, a second historical flowering quality factor is randomly selected from the set of historical flowering quality factors after the removal, and the same clustering analysis is performed according to the same principle of obtaining the first clustered historical flowering quality factor set in combination with the preset similarity threshold value to obtain a second clustered historical flowering quality factor set. After multiple similar clustering analyses, each historical flowering quality factor in the set of historical flowering quality factors is divided into a clustered historical flowering quality factor set, the analysis is stopped, and the multiple clustered historical flowering quality factor sets are obtained.

[0044] Since the historical flowering quality factors and the historical flowering indicators are one-to-one corresponding, and the historical flowering indicators and the historical environmental parameters are one-to-one corresponding, the historical flowering quality factors and the historical environmental parameters are also one-to-one corresponding. The set of historical environmental parameters is mapped and clustered according to the multiple clustered historical flowering quality factor sets to obtain multiple clustered historical environmental parameter sets. The multiple clustered historical environmental parameter sets reflect the environmental distribution of strawberry plants corresponding to the flowering quality within a certain range. Each clustered historical environmental parameter set contains environmental parameter data matching the flowering quality factor set of the cluster. Thus, the correlation analysis between the flowering quality factors and the environmental parameters is realized, and the technical effect of providing data support for the environmental regulation of the greenhouse is achieved.

[0045] S400: respectively on the plurality of clustering historical environment parameter set parameter set search, determine a plurality of clustering set environment parameters;

[0046] Further, respectively on the plurality of clustering historical environment parameter set parameter set search, determine a plurality of clustering set environment parameters, the embodiment step S400 of the application further comprises:

[0047] Randomly extract a first clustering historical environment parameter set from the plurality of clustering historical environment parameter set, calculate the parameter mean of the first clustering historical environment parameter set, and obtain the first clustering historical environment parameter mean;

[0048] The first clustering historical environment parameter mean is taken as the starting point of the centralized search, and the parameter centralized search is carried out in the first clustering historical environment parameter set according to the preset search bandwidth, and the first clustering set environment parameter is obtained.

[0049] The plurality of clustering historical environment parameter sets are calculated, and the calculation results are respectively taken as the starting point of the centralized search, and the parameter centralized search is carried out in the first clustering historical environment parameter set according to the preset search bandwidth, and the plurality of clustering set environment parameters are obtained.

[0050] Further, as shown in Figure 2 The first clustering historical environment parameter mean is taken as the starting point of the centralized search, and the parameter centralized search is carried out in the first clustering historical environment parameter set according to the preset search bandwidth, and the first clustering set environment parameter is obtained.

[0051] The first clustering historical environment parameter mean is taken as the starting point of the centralized search, and the parameter centralized search is carried out in the first clustering historical environment parameter set according to the preset search bandwidth, and the first clustering set environment parameter is obtained.

[0052] Respectively based on the preset search bandwidth and the first clustering historical environment parameter set, the search neighborhood of the centralized search starting point and the iteration first clustering historical environment parameter is constructed, and the centralized search neighborhood and the iteration search neighborhood are obtained.

[0053] Determine whether the neighborhood density of the iteration search neighborhood is greater than or equal to the neighborhood density of the centralized search neighborhood, if yes, update the iteration first clustering historical environment parameter to the centralized search starting point, and iterate again in the first clustering historical environment parameter set according to the preset search bandwidth, until the preset iteration number is met, and the iteration first clustering historical environment parameter obtained by the last iteration is taken as the first clustering set environment parameter.

[0054] Further, the step S400 of the embodiment of the application further includes:

[0055] If no, the search direction from the centralized search starting point to the iterated first clustering historical environment parameter is added into the disabled search table, the first clustering historical environment parameter mean is taken as the centralized search starting point again, and iteration is performed in the first clustering historical environment parameter set according to the preset search bandwidth until a preset iteration number is met, and the iterated first clustering historical environment parameter obtained in the last iteration is taken as the first clustering centralized environment parameter. The disabled search table includes a preset iteration disabled number, and the preset iteration disabled number is the number of times of iteration disabled of the search direction stored in the disabled search table.

[0056] In one possible embodiment, after the plurality of clustering historical environment parameter sets are obtained, since there are accidental errors or less frequent historical environment parameters in each set, it is necessary to respectively perform parameter centralized search on them to determine the plurality of clustering centralized environment parameters capable of reflecting the general parameter condition of the plurality of clustering historical environment parameter sets. The technical effect of providing a reliable control target for subsequent greenhouse environment control is achieved.

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

[0058] Further, the first clustering historical environment parameter mean is taken as the centralized search starting point, and parameter centralized search is performed in the first clustering historical environment parameter set according to a preset search bandwidth to obtain a first clustering centralized environment parameter. The preset search bandwidth is the distance of single movement during iteration search in the first clustering historical environment parameter set, that is, the difference of parameters. The first clustering centralized environment parameter is the environment parameter that can best represent the parameter condition of the first clustering historical environment parameter set.

[0059] Preferably, the first clustering historical environmental parameter mean value is taken as a central search starting point, and the first clustering historical environmental parameter set is moved according to the preset search bandwidth to obtain an iterative first clustering historical environmental parameter. The central search neighborhood is constructed with the central search starting point as the center and the preset search bandwidth as the radius, wherein the difference between the first clustering historical environmental parameter in the central search neighborhood and the first clustering historical environmental parameter mean value is less than or equal to the preset search bandwidth. Further, the iterative search neighborhood is constructed with the iterative first clustering historical environmental parameter as the center and the preset search bandwidth as the radius, wherein the difference between the first clustering historical environmental parameter in the iterative search neighborhood and the iterative first clustering historical environmental parameter is less than or equal to the preset search bandwidth.

[0060] The number of parameters in the central search neighborhood and the iterative search neighborhood is counted respectively to obtain the central search neighborhood parameter quantity and the iterative search neighborhood parameter quantity. The ratio of the central search neighborhood parameter quantity to the area of the central search neighborhood is taken as the neighborhood density of the central search neighborhood. The ratio of the iterative search neighborhood parameter quantity to the area of the iterative search neighborhood is taken as the neighborhood density of the iterative search neighborhood. The neighborhood density reflects the degree of data aggregation in each search neighborhood.

[0061] Further, it is judged whether the neighborhood density of the iterative search neighborhood is greater than or equal to the neighborhood density of the central search neighborhood. If yes, it indicates that the iterative first clustering historical environmental parameter is more representative than the central search starting point. At this time, the iterative first clustering historical environmental parameter is updated as the central search starting point, and the first clustering historical environmental parameter set is iterated again according to the preset search bandwidth until the preset iteration number (the maximum iteration number preset by the person skilled in the art) is met. The iterative first clustering historical environmental parameter obtained in the last iteration is taken as the first clustering central environmental parameter.

[0062] If not, it indicates that the central search starting point is more representative than the iterative first clustering historical environmental parameter. At this time, the search direction from the central search starting point to the iterative first clustering historical environmental parameter is added to the disabled search table. The disabled search table includes a preset iteration disabled number, and the preset iteration disabled number is the number of times of iteration disabled for the search direction stored in the disabled search table.

[0063] That is, the direction of obtaining the iterated first clustering historical environment parameter will not be used within the preset iteration disabling times, thereby improving the globality and efficiency of the centralized search. Further, the first clustering historical environment parameter mean is taken as the starting point of the centralized search again, and iteration is performed in the first clustering historical environment parameter set according to the preset search bandwidth until the preset iteration times are met, and the iterated first clustering historical environment parameter obtained in the last iteration is taken as the first clustering centralized environment parameter.

[0064] Based on the same principle as obtaining the first clustering centralized environment parameter, the mean value of the plurality of clustering historical environment parameter sets is calculated, and the calculation results are respectively taken as the starting point of the centralized search, and the parameter centralized search is performed in the first clustering historical environment parameter set according to the preset search bandwidth, and the plurality of clustering centralized environment parameters are obtained.

[0065] By obtaining the plurality of clustering centralized environment parameters, the general situation of the environment parameters corresponding to different flowering quality factor sets is determined, which provides target environment parameters for subsequent greenhouse environment regulation and control, and thus the technical effect of providing a basis for determining reliable environment regulation and control parameters is achieved.

[0066] S500: configuring a greenhouse environment regulation and control parameter according to the plurality of clustering centralized environment parameters to obtain a plurality of target greenhouse environment regulation and control parameters;

[0067] S600: regulating and controlling devices of a plurality of test greenhouses based on the plurality of target greenhouse environment regulation and control parameters, and performing a strawberry flowering influence analysis test on the target strawberries in the plurality of test greenhouses after the regulation and control is completed.

[0068] Further, a plurality of target greenhouse environment regulation and control parameters are obtained by configuring a greenhouse environment regulation and control parameter according to the plurality of clustering centralized environment parameters. The embodiment S500 of the present application further comprises:

[0069] Respectively taking the plurality of clustering centralized environment parameters as indexes, a historical greenhouse environment regulation and control log set is searched to obtain a plurality of historical greenhouse environment regulation and control parameter sets and a plurality of historical greenhouse environment regulation and control matching degree sets;

[0070] The plurality of historical greenhouse environment regulation and control matching degree sets are traversed to extract a maximum matching value to obtain a plurality of historical greenhouse environment regulation and control matching degree maximum values, and historical greenhouse environment regulation and control parameters corresponding to the plurality of historical greenhouse environment regulation and control matching degree maximum values are taken as a plurality of direction historical greenhouse environment regulation and control parameters;

[0071] The multiple historical greenhouse environment regulation parameters are optimized in multiple scales and directions according to the multiple directional optimization scale sets, and multiple follow-up optimized historical greenhouse environment regulation parameter sets are obtained.

[0072] The multiple follow-up optimized historical greenhouse environment regulation parameter sets are analyzed for regulation matching degrees, and multiple follow-up optimized historical greenhouse environment regulation matching degree sets are obtained.

[0073] The multiple follow-up optimized historical greenhouse environment regulation matching degree sets are compared with the multiple historical greenhouse environment regulation matching degree maximum values, and when there is a follow-up optimized historical greenhouse environment regulation matching degree greater than the corresponding multiple historical greenhouse environment regulation matching degree maximum values, the multiple follow-up optimized historical greenhouse environment regulation parameters corresponding to the maximum value in the multiple follow-up optimized historical greenhouse environment regulation matching degree sets are updated as the multiple optimization directions.

[0074] According to the updated multiple optimization directions, the multiple follow-up optimized historical greenhouse environment regulation parameter sets are optimized in multiple scales and directions, and after multiple optimizations, the multiple target greenhouse environment regulation parameters are obtained until a preset optimization number is met.

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

[0076] The difference between the multiple historical greenhouse environment regulation matching degree sets and the multiple historical greenhouse environment regulation matching degree maximum values is calculated respectively, and the calculation result is divided by the multiple historical greenhouse environment regulation matching degree maximum values. The product of the ratio and a preset adjustment scale is taken as the multiple directional optimization scale sets.

[0077] In one possible embodiment, after the environmental parameters in the multiple clusters are obtained, the environmental regulation devices in the multiple test greenhouses are analyzed for regulation parameters, and the multiple target greenhouse environment regulation parameters are determined. Each cluster of environmental parameters includes temperature, humidity, sunshine duration, and other strawberry growth environment parameters. Each target greenhouse environment regulation parameter is used to adjust the parameters of the environmental regulation devices in a test greenhouse, so that the environment in the test greenhouse can meet the requirements of the corresponding cluster of environmental parameters. Preferably, a target greenhouse environment regulation parameter includes a temperature set value, a power output value, a heating time, a sunshine time, a humidity set value, an air volume, an air speed, an illumination intensity, a light wavelength, a carbon dioxide gas injection flow rate, and the like. Optionally, the environmental regulation devices in each test greenhouse include heating equipment (such as a warm air blower, a hot water pipeline, an electric heater, etc.), cooling equipment (a fan, a sunshade net, a fogging cooling system, etc.), humidifying equipment (a fogger, a spraying system), dehumidifying equipment (a dehumidifier, an air conditioner, a ventilation system), light regulation equipment (an LED plant growth lamp, a fluorescent lamp, a sunshade net), and the like.

[0078] Preferably, based on the plurality of target greenhouse environment control parameters, the parameters of the corresponding devices in the plurality of test greenhouses are adjusted, and the devices are started, so that after running, the environment of the plurality of test greenhouses meets the requirements of the plurality of cluster concentrated environment parameters, and then the target strawberries in the plurality of test greenhouses after control are subjected to strawberry flowering influence analysis test. The technical effects of improving the accuracy of greenhouse environment control and improving the reliability of control are achieved.

[0079] In one embodiment, the historical greenhouse environment control log set is searched with the plurality of cluster concentrated environment parameters as indexes respectively, to obtain a plurality of historical greenhouse environment control parameter sets and a plurality of historical greenhouse environment control matching degree sets. The historical greenhouse environment control log set is the control record when the greenhouse is controlled in the historical time. Each historical greenhouse environment control matching degree reflects the similarity between the environment parameters that the greenhouse can achieve after control and the corresponding cluster concentrated environment parameters. The higher the historical greenhouse environment control matching degree is, the higher the quality of the corresponding historical greenhouse environment control parameters is.

[0080] Therefore, the maximum values of the matching values in the plurality of historical greenhouse environment control matching degree sets are extracted respectively, to obtain a plurality of historical greenhouse environment control matching degree maximum values. The historical greenhouse environment control parameters corresponding to the plurality of historical greenhouse environment control matching degree maximum values are taken as a plurality of directional historical greenhouse environment control parameters. The plurality of directional historical greenhouse environment control parameters are the optimal greenhouse environment control parameters of each cluster concentrated environment parameter in the historical time, and thus are taken as the adjustment and optimization direction.

[0081] Optionally, the plurality of directional historical greenhouse environment control parameters are taken as a plurality of optimization directions, and the plurality of historical greenhouse environment control parameter sets are directionally optimized according to a plurality of directional optimization scale sets respectively according to different scales. That is, according to the difference between the plurality of historical greenhouse environment control parameter sets and the corresponding plurality of directional historical greenhouse environment control parameters, the plurality of historical greenhouse environment control parameter sets are adjusted at different adjustment amplitudes respectively, to obtain a plurality of follow-up optimization historical greenhouse environment control parameter sets. That is, the historical greenhouse environment control parameter sets are adjusted to the temporary optimal parameters, to obtain more greenhouse environment control parameters, and to provide more data for obtaining the optimal solution.

[0082] Preferably, the difference between the multiple sets of historical greenhouse environment control matching degrees and the multiple maximum values ​​of historical greenhouse environment control matching degrees is calculated respectively. The calculated result is then divided by the multiple maximum values ​​of historical greenhouse environment control matching degrees. The product of this ratio and a preset adjustment scale (a single parameter adjustment range preset by those skilled in the art) is used as a multiple set of directional optimization scales. 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 matching degree analysis on the multiple sets of historical greenhouse environment control parameters that follow optimization, thereby obtaining multiple sets of matching degrees for historical greenhouse environment control. Preferably, multiple sample greenhouse environment control parameters and multiple sample greenhouse environment control matching degrees are obtained as training data. The framework based on the feedforward neural network is trained under supervision using the training data to learn the mapping relationship between the greenhouse environment control parameters and the greenhouse environment control matching degree until the training converges, thus obtaining the trained matching degree prediction network layer.

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

[0085] Based on the updated multiple optimization directions, the multiple sets of historical greenhouse environment control parameters following optimization are subjected to multi-scale directional optimization. After multiple optimizations, until the preset number of optimizations is met, the multiple target greenhouse environment control parameters are obtained.

[0086] Optionally, when there is no matching degree of the following optimized historical greenhouse environment control that is greater than the maximum value of the matching degree of the corresponding multiple historical greenhouse environment control, the current multiple optimization directions are maintained, and the multiple sets of following optimized historical greenhouse environment control matching degrees are directionally optimized according to multiple sets of directional optimization scales until the preset number of optimizations is met. The multiple following optimized historical greenhouse environment control parameters corresponding to the maximum value of the matching degree of the multiple following optimized greenhouse environment control during the optimization process are used as the multiple target greenhouse environment control parameters.

[0087] Through multiple rounds of optimization, directional adjustment and matching degree analysis, it can be ensured that the environment of each test greenhouse can reach the best state, effectively meeting the requirements of multiple cluster concentrated environmental parameters. The technical effect of effectively regulating the environmental conditions of each test greenhouse is achieved.

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

[0089] The present application obtains a set of historical flowering indicators, evaluates the flowering quality, then performs same type clustering analysis on the set of historical flowering quality factors, and maps the set of historical environmental parameters according to the respective results to obtain a plurality of clustered sets of historical environmental parameters. Then, the plurality of clustered sets of historical environmental parameters are searched for parameters to determine a plurality of clustered sets of environmental parameters, which realizes the determination of the environmental conditions required by different test greenhouses. Then, the greenhouse environmental control parameters are configured according to the plurality of clustered sets of environmental parameters to obtain a plurality of target greenhouse environmental control parameters. Further, based on the plurality of target greenhouse environmental control parameters, the devices of the plurality of test greenhouses are regulated. The strawberry flowering influence analysis test is performed on the target strawberries in the plurality of test greenhouses after the regulation is completed. The technical effect of optimizing and configuring the environmental control parameters of the test greenhouse to improve the accuracy of the greenhouse environmental control is achieved.

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

[0091] A historical environmental parameter set obtaining module 11 is configured to obtain the basic variety information of target strawberries, search a big data with the preset flowering indicators as the search target, and index the basic variety information to obtain a set of historical flowering indicators and a set of historical environmental parameters. Each historical flowering indicator corresponds to a set of historical environmental parameters.

[0092] A historical flowering quality factor set obtaining module 12 is configured to evaluate the flowering quality based on the set of historical flowering indicators to obtain a set of historical flowering quality factors.

[0093] A clustered historical environmental parameter set obtaining module 13 is configured to perform same type clustering analysis on the set of historical flowering quality factors to determine a plurality of clustered sets of historical flowering quality factors, and perform mapping clustering on the set of historical environmental parameters based on the plurality of clustered sets of historical flowering quality factors to obtain a plurality of clustered sets of historical environmental parameters.

[0094] The clustering concentrated environment parameter determination module 14 is configured to perform parameter clustering search on the plurality of clustering historical environment parameter sets respectively, and determine a plurality of clustering concentrated environment parameters.

[0095] The greenhouse environment regulation parameter obtaining module 15 is configured to configure greenhouse environment regulation parameters according to the plurality of clustering concentrated environment parameters, and obtain a plurality of target greenhouse environment regulation parameters.

[0096] The environment regulation module 16 is configured to regulate the devices of the plurality of test greenhouses based on the plurality of target greenhouse environment regulation parameters, and perform strawberry flowering influence analysis test on the target strawberries in the plurality of test greenhouses after the regulation is completed.

[0097] Further, the clustering concentrated environment parameter determination module 14 is configured to perform the following steps:

[0098] Randomly extract a first clustering historical environment parameter set from the plurality of clustering historical environment parameter sets, calculate the parameter mean value of the first clustering historical environment parameter set, and obtain a first clustering historical environment parameter mean value;

[0099] Take the first clustering historical environment parameter mean value as a clustering search starting point, perform parameter clustering search in the first clustering historical environment parameter set according to a preset search bandwidth, and obtain a first clustering concentrated environment parameter;

[0100] Perform mean value calculation on the plurality of clustering historical environment parameter sets, and take the calculation results as clustering search starting points respectively, perform parameter clustering search in the first clustering historical environment parameter set according to the preset search bandwidth, and obtain the plurality of clustering concentrated environment parameters.

[0101] Further, the clustering concentrated environment parameter determination module 14 is configured to perform the following steps:

[0102] Take the first clustering historical environment parameter mean value as a clustering search starting point, and move in the first clustering historical environment parameter set according to the preset search bandwidth, to obtain an iterative first clustering historical environment parameter;

[0103] Construct search neighborhoods of the clustering search starting point and 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 clustering search neighborhood and an iterative search neighborhood;

[0104] determining whether the neighborhood density of the iterative search neighborhood is greater than or equal to the neighborhood density of the concentrated search neighborhood, if yes, updating the iterative first clustering historical environment parameter as a concentrated search starting point, and then performing iteration in the first clustering historical environment parameter set according to the preset search bandwidth again, until a preset iteration number is met, and taking the iterative first clustering historical environment parameter obtained in the last iteration as the first clustering concentrated environment parameter.

[0105] Further, the clustering concentrated environment parameter determination module 14 is configured to perform the following steps:

[0106] If no, adding a search direction from the concentrated search starting point to the iterative first clustering historical environment parameter into a disabled search table, taking the first clustering historical environment parameter mean value as the concentrated search starting point again, performing iteration in the first clustering historical environment parameter set according to the preset search bandwidth, until a preset iteration number is met, and taking the iterative first clustering historical environment parameter obtained in the last iteration as the first clustering concentrated environment parameter, wherein the disabled search table includes a preset iteration disabled number, and the preset iteration disabled number is a number of times of iteration disabled of the search direction stored in the disabled search table.

[0107] Further, the greenhouse environment regulation parameter obtaining module 15 is configured to perform the following steps:

[0108] Respectively taking the plurality of clustering concentrated environment parameters as indexes, searching the historical greenhouse environment regulation log set to obtain a plurality of historical greenhouse environment regulation parameter sets and a plurality of historical greenhouse environment regulation matching degree sets;

[0109] Extracting a maximum value of the matching value from the plurality of historical greenhouse environment regulation matching degree sets to obtain a plurality of historical greenhouse environment regulation matching degree maximum values, and taking historical greenhouse environment regulation parameters corresponding to the plurality of historical greenhouse environment regulation matching degree maximum values as a plurality of directional historical greenhouse environment regulation parameters;

[0110] Taking the plurality of directional historical greenhouse environment regulation parameters as a plurality of optimization directions, performing multi-scale directional optimization on the plurality of historical greenhouse environment regulation parameter sets according to a plurality of directional optimization scale sets to obtain a plurality of follow-up optimization historical greenhouse environment regulation parameter sets;

[0111] Performing regulation matching degree analysis on the plurality of follow-up optimization historical greenhouse environment regulation parameter sets to obtain a plurality of follow-up optimization historical greenhouse environment regulation matching degree sets;

[0112] The plurality of follow-up optimization historical greenhouse environment regulation matching degree sets are compared with the plurality of historical greenhouse environment regulation matching degree maximum values, and when there is a follow-up optimization historical greenhouse environment regulation matching degree greater than the plurality of historical greenhouse environment regulation matching degree maximum values, the plurality of follow-up optimization historical greenhouse environment regulation matching degree maximum value corresponding plurality of follow-up optimization historical greenhouse environment regulation matching degree is updated to the plurality of optimization directions.

[0113] According to the updated plurality of optimization directions, the plurality of follow-up optimization historical greenhouse environment regulation parameter sets are subjected to multi-scale directional optimization, and after multiple optimizations, until the preset optimization number is satisfied, the plurality of target greenhouse environment regulation parameters are obtained.

[0114] Further, the greenhouse environment regulation parameter obtaining module 15 is configured to perform the following steps:

[0115] The difference between the plurality of historical greenhouse environment regulation matching degree sets and the plurality of historical greenhouse environment regulation matching degree maximum values is calculated respectively, and the calculation result is compared with the plurality of historical greenhouse environment regulation matching degree maximum values, and the product of the ratio and the preset adjustment scale is taken as the plurality of directional optimization scale sets.

[0116] Further, the historical flowering quality factor set obtaining module 12 is configured to perform the following steps:

[0117] Obtain a plurality of sample flowering index sets and a plurality of 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 framework constructed based on the convolutional neural network is trained using the training set, the mapping relationship between the flowering quality index set and the flowering quality factor is learned, and the training result is supervised using the validation set, until the training converges, and the quality evaluation network layer is obtained after the training is completed;

[0119] The quality evaluation network layer is used to analyze the historical flowering index set, and the historical flowering quality factor set is obtained. 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 describes a specific embodiment of the present application. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or may be advantageous.

[0120] The above only describes the preferred embodiments of the present application, and does not limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0121] The specification and drawings are, of course, to be regarded in an illustrative rather than a restrictive sense. It is to be understood that any such modifications, variations, combinations or equivalents that fall within the scope of the application are intended to be embraced herein.

Claims

1. A method for controlling the environment of a greenhouse tunnel for the analysis of the effect of flowering in strawberries, characterized by, The method comprises: obtaining the basic variety information of the target strawberry, taking the preset flowering index as the retrieval target, and performing big data retrieval on the basic variety information to obtain a historical flowering index set and a historical environment parameter set, wherein each historical flowering index corresponds to a historical environment parameter set; performing flowering quality evaluation based on the historical flowering index set to obtain a historical flowering quality factor set; performing same-type clustering analysis on the historical flowering quality factor set to determine a plurality of clustered historical flowering quality factor sets, and performing mapping clustering on the historical environment parameter set based on the plurality of clustered historical flowering quality factor sets to obtain a plurality of clustered historical environment parameter sets; performing parameter set search on the plurality of clustered historical environment parameter sets respectively to determine a plurality of clustered environment parameters; performing greenhouse environment control parameter configuration according to the plurality of clustered environment parameters to obtain a plurality of target greenhouse environment control parameters; controlling the devices of a plurality of test greenhouse based on the plurality of target greenhouse environment control parameters, and performing strawberry flowering influence analysis test on the target strawberry in the plurality of test greenhouses after the control is completed; wherein the greenhouse environment control parameter configuration according to the plurality of clustered environment parameters to obtain the plurality of target greenhouse environment control parameters comprises: respectively taking the plurality of clustered environment parameters as indexes, performing retrieval on a historical greenhouse environment control log set to obtain a plurality of historical greenhouse environment control parameter sets and a plurality of historical greenhouse environment control matching degree sets; performing maximum matching value extraction on the plurality of historical greenhouse environment control matching degree sets to obtain a plurality of historical greenhouse environment control matching degree maximum values, and taking the historical greenhouse environment control parameters corresponding to the plurality of historical greenhouse environment control matching degree maximum values as a plurality of directional historical greenhouse environment control parameters; taking the plurality of directional historical greenhouse environment control parameters as a plurality of optimization directions, performing multi-scale directional optimization on the plurality of historical greenhouse environment control parameter sets according to a plurality of directional optimization scale sets to obtain a plurality of follow-up optimization historical greenhouse environment control parameter sets; performing control matching degree analysis on the plurality of follow-up optimization historical greenhouse environment control parameter sets to obtain a plurality of follow-up optimization historical greenhouse environment control matching degree sets; comparing the plurality of follow-up optimization historical greenhouse environment control matching degree sets with the plurality of historical greenhouse environment control matching degree maximum values, when there is a follow-up optimization historical greenhouse environment control matching degree greater than the corresponding plurality of historical greenhouse environment control matching degree maximum values, updating the plurality of follow-up optimization historical greenhouse environment control parameters corresponding to the maximum value in the plurality of follow-up optimization historical greenhouse environment control matching degree sets to the plurality of optimization directions; performing multi-scale directional optimization on the plurality of follow-up optimization historical greenhouse environment control parameter sets according to the updated plurality of optimization directions, and performing optimization for multiple times until a preset optimization number is satisfied to obtain the plurality of target greenhouse environment control parameters; wherein the flowering quality evaluation based on the historical flowering index set to obtain the historical flowering quality factor set comprises: Obtaining a plurality of sample flowering index sets and a plurality of sample flowering quality factor sets as training sample data; According to a preset proportion, the training sample data is divided into a training set and a validation set, the framework based on the convolutional neural network is trained by using the training set, the mapping relationship between the flowering quality index set and the flowering quality factor is learned, and the training result is supervised by using the validation set until the training converges, and the quality evaluation network layer is obtained. The quality evaluation network layer is used to analyze the historical flowering index set, and the historical flowering quality factor set is obtained.

2. The method for controlling the environment of a greenhouse for the analysis of the effect of strawberry flowering according to claim 1, wherein, The plurality of clustered historical environmental parameter sets are searched respectively, and a plurality of clustered environmental parameters are determined, including: A first clustered historical environmental parameter set is randomly extracted from the plurality of clustered historical environmental parameter sets, the parameter mean of the first clustered historical environmental parameter set is calculated, and the first clustered historical environmental parameter mean is obtained. The first clustered historical environmental parameter mean is taken as the starting point of the centralized search, and the parameter centralized search is performed in the first clustered historical environmental parameter set according to the preset search bandwidth, and the first clustered environmental parameter is obtained. The plurality of clustered historical environmental parameter sets are calculated, and the calculation results are taken as the starting points of the centralized search respectively, and the parameter centralized search is performed in the first clustered historical environmental parameter set according to the preset search bandwidth, and the plurality of clustered environmental parameters are obtained.

3. The method for controlling the environment of a greenhouse for the analysis of the effect of strawberry flowering according to claim 2, wherein, The first clustered historical environmental parameter mean is taken as the starting point of the centralized search, and the parameter centralized search is performed in the first clustered historical environmental parameter set according to the preset search bandwidth, and the first clustered environmental parameter is obtained, including: The first clustered historical environmental parameter mean is taken as the starting point of the centralized search, and the parameter centralized search is performed in the first clustered historical environmental parameter set according to the preset search bandwidth, and the first clustered environmental parameter is obtained, including: The search neighborhood of the centralized search starting point and the iterative first clustered historical environmental parameter is constructed based on the preset search bandwidth and the first clustered historical environmental parameter set respectively, and the centralized search neighborhood and the iterative search neighborhood are obtained. If the neighborhood density of the iterative search neighborhood is greater than or equal to the neighborhood density of the centralized search neighborhood, the iterative first clustered historical environmental parameter is updated as the centralized search starting point, and the iteration is performed again in the first clustered historical environmental parameter set according to the preset search bandwidth until the preset iteration number is met, and the iterative first clustered historical environmental parameter obtained by the last iteration is taken as the first clustered environmental parameter.

4. The method for controlling the environment of a greenhouse for the analysis of the effect of strawberry flowering according to claim 3, wherein, Including: If not, the search direction from the centralized search starting point to the iterated first clustering historical environment parameter is added to the disabled search table, the first clustering historical environment parameter mean is taken as the centralized search starting point again, and iteration is performed in the first clustering historical environment parameter set according to the preset search bandwidth until a preset iteration number is met, and the iterated first clustering historical environment parameter obtained in the last iteration is taken as the first clustering centralized environment parameter. The disabled search table includes a preset iteration disabled number, and the preset iteration disabled number is the number of times of iteration disabled for the search direction stored in the disabled search table.

5. The method for controlling the environment of a greenhouse for the analysis of the effect of strawberry flowering according to claim 1, wherein, The difference between the plurality of historical greenhouse environment regulation matching degree sets and the plurality of historical greenhouse environment regulation matching degree maximum values is calculated respectively, and the calculation result is compared with the plurality of historical greenhouse environment regulation matching degree maximum values. The product of the ratio and a preset adjustment scale is taken as a plurality of directional optimization scale sets.

6. A greenhouse environment control system for analysis of strawberry flowering effect, characterized in that, The system is used to execute the greenhouse environment regulation method for strawberry flowering influence analysis of any one of claims 1-5, and the system comprises: A historical environment parameter set obtaining module is configured to obtain basic variety information of a target strawberry, take a preset flowering index as a retrieval target, and perform big data retrieval by taking the basic variety information as an index to obtain a historical flowering index set and a historical environment parameter set. Each historical flowering index corresponds to a historical environment parameter set. A historical flowering quality factor set obtaining module is configured to perform flowering quality evaluation based on the historical flowering index set to obtain a historical flowering quality factor set. A clustering historical environment parameter set obtaining module is configured to perform same-type clustering analysis by traversing the historical flowering quality factor set to determine a plurality of clustering historical flowering quality factor sets, and perform mapping clustering on the historical environment parameter set based on the plurality of clustering historical flowering quality factor sets to obtain a plurality of clustering historical environment parameter sets. A clustering centralized environment parameter determining module is configured to perform parameter centralized search on the plurality of clustering historical environment parameter sets respectively to determine a plurality of clustering centralized environment parameters. A greenhouse environment regulation parameter obtaining module is configured to perform greenhouse environment regulation parameter configuration according to the plurality of clustering centralized environment parameters to obtain a plurality of target greenhouse environment regulation parameters. An environment regulation module is configured to regulate devices of a plurality of test greenhouses based on the plurality of target greenhouse environment regulation parameters, and perform strawberry flowering influence analysis test on the target strawberry in the plurality of test greenhouses after the regulation is completed.

Images