Intelligent crop growth environment monitoring and control method and system based on the Internet of Things

Abstract

The present invention relates to the field of smart agricultural technology, specifically to an intelligent crop growth environment monitoring and control method and system based on the Internet of Things. The present invention sets corresponding initial acquisition parameters for each acquisition environment, collects crop growth information according to the initial acquisition parameters of each acquisition environment, then screens out unqualified crop growth information, then inputs the unqualified crop growth information into an acquisition parameter model to train the acquisition parameter model, obtains a trained acquisition parameter model, and finally continuously optimizes the initial acquisition parameters in combination with the predicted crop information output by the trained acquisition parameter model to obtain optimized acquisition parameters, and collects crop growth information according to the optimized acquisition parameters. The present application continuously optimizes the acquisition parameters to increase the acquisition frequency of unqualified crop growth information, so as to collect more abnormal data and facilitate finding the root cause of unqualified crop growth information.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent traffic control, and in particular to an intelligent crop growth environment monitoring and regulation method and system based on the Internet of Things. Background Art

[0002] Modern agricultural production has increasingly higher requirements for environmental conditions, especially in terms of refined management of temperature, humidity, light, soil composition, etc. Traditional manual monitoring methods are inefficient and difficult to ensure data accuracy. With the development of Internet of Things technology, it has become a trend to improve the crop growth environment through real-time monitoring and automatic control.

[0003] A Chinese patent with publication number CN116449738A discloses a smart agricultural crop monitoring and control system based on the Internet of Things. By setting cameras and alarms around the crops, the cameras can monitor and record the surrounding conditions of the crops. If animals are found stealing and destroying the crops, the alarm will sound to drive them away, thereby achieving the purpose of protecting the crops. However, the prior art generally adopts equal interval sampling during collection, resulting in incomplete information collection for abnormal time periods, making the collected crop information incomplete. Summary of the invention

[0004] The purpose of the present invention is to address the problems existing in the background technology and to propose an intelligent crop growth environment monitoring and control method and system based on the Internet of Things.

[0005] The technical solution of the present invention:

[0006] On the one hand, a smart crop growth environment monitoring and control method based on the Internet of Things is provided, including:

[0007] Setting corresponding initial collection parameters for each collection environment, and collecting crop growth information in each collection environment according to the initial collection parameters of each collection environment;

[0008] Screen out unqualified crop growth information based on crop growth information;

[0009] Create acquisition parameter model;

[0010] Inputting unqualified crop growth information into a collection parameter model to train the collection parameter model to obtain a trained collection parameter model; the trained collection parameter model has the ability to output predicted unqualified crop growth information according to the input collection environment and initial collection parameters, and the predicted unqualified crop growth information output by the trained collection parameter model is recorded as predicted crop information;

[0011] The initial acquisition parameters are continuously optimized in combination with the predicted crop information output by the trained acquisition parameter model to obtain optimized acquisition parameters, and crop growth information is collected based on the optimized acquisition parameters.

[0012] Preferably, corresponding initial collection parameters are set for each collection environment, and crop growth information in each collection environment is collected according to the initial collection parameters of each collection environment, including:

[0013] Create a crop growth information table;

[0014] Select a collection environment;

[0015] Set the initial collection parameters of the collection environment; the initial collection parameters include the initial collection period and the initial collection node;

[0016] Collecting a plurality of crop growth information based on the initial collection node within the initial collection period, and putting all the collected crop growth information into a crop growth information table; the crop growth information includes environmental humidity, environmental temperature, and soil pH value;

[0017] Return to select a collection environment until all collection environments are selected, and obtain the crop growth information of each collection environment.

[0018] Preferably, screening out unqualified crop growth information based on crop growth information includes:

[0019] Select a collection environment;

[0020] Setting a crop growth standard threshold range for each initial collection node of the collection environment; the crop growth standard threshold range includes a humidity standard threshold range, a temperature standard threshold range and a pH value standard threshold range;

[0021] Select an initial collection period;

[0022] Randomly select a crop growth information within the initial collection period from the crop growth information table;

[0023] Determining whether the crop growth information is within a crop growth standard threshold range;

[0024] If the crop growth information is within the crop growth standard threshold range, the crop growth information is qualified;

[0025] Return to randomly select a crop growth information within the initial collection period from the crop growth information table until all crop growth information within the initial collection period has been selected;

[0026] Return to select an initial collection cycle until all initial collection cycles are selected;

[0027] Return to select a collection environment until all collection environments are selected.

[0028] Preferably, screening out unqualified crop growth information based on the crop growth information further includes:

[0029] If at least one of the crop growth information is not within the crop growth standard threshold range, the crop growth information is unqualified;

[0030] Return to randomly select a crop growth information within the initial collection period from the crop growth information table until all crop growth information within the initial collection period has been selected;

[0031] Return to select an initial collection cycle until all initial collection cycles are selected;

[0032] Return to select a collection environment until all collection environments have been selected;

[0033] Filter out all unqualified crop growth information contained in the collection environment.

[0034] Preferably, determining whether the crop growth information is within a crop growth standard threshold range includes:

[0035] Determine whether the ambient humidity is within the humidity standard threshold range;

[0036] If the ambient humidity is within the humidity standard threshold range, then determining whether the ambient temperature is within the temperature standard threshold range;

[0037] If the ambient temperature is within the temperature standard threshold range, it is determined whether the soil pH value is within the pH standard threshold range.

[0038] Preferably, the unqualified crop growth information is input into the acquisition parameter model to train the acquisition parameter model, and the trained acquisition parameter model is obtained, including:

[0039] Obtaining a collection environment corresponding to each piece of unqualified crop growth information and an initial collection node corresponding to each piece of unqualified crop growth information; recording one piece of unqualified crop growth information, the collection environment corresponding to the unqualified crop growth information, and the initial collection node corresponding to the unqualified crop growth information as a collection sample;

[0040] Create acquisition parameter model;

[0041] All collected samples are divided into training sets and test sets in random proportions;

[0042] Inputting the acquisition samples in the training set into the acquisition parameter model in sequence to train the acquisition parameter model, thereby obtaining a trained acquisition parameter model;

[0043] The collected samples in the test set are input into the trained collection parameter model in sequence to verify whether the trained collection parameter model is trained.

[0044] Preferably, the acquisition samples in the training set are sequentially input into the acquisition parameter model to train the acquisition parameter model, and the trained acquisition parameter model is obtained, including:

[0045] Select a collection environment and obtain all collection samples in the collection environment;

[0046] Count the number of unqualified crop growth information included in each collection cycle under the collection environment;

[0047] The collection cycles are sorted from large to small according to the number of times of unqualified crop growth information contained in each collection cycle, and the first N collection cycles are selected; the first N unqualified collection cycles are recorded as adjustment cycles;

[0048] Return to select a collection environment and obtain all the collection samples in the collection environment until all the collection environments in the training set are selected, and obtain multiple adjustment cycles for each collection environment.

[0049] Preferably, the initial acquisition parameters are continuously optimized in combination with the predicted crop information output by the trained acquisition parameter model to obtain optimized acquisition parameters, and the crop growth information is collected according to the optimized acquisition parameters, including:

[0050] Randomly select a collection environment;

[0051] Randomly select an adjustment period under the acquisition environment and obtain the start node and end node of the adjustment period;

[0052] Based on the start node and the end node of the adjustment cycle, a new acquisition node is created by formula 1;

[0053] (0,n)Formula 1;

[0054] in, is the i-th new acquisition node, is the starting node of the adjustment cycle. is the end node of the adjustment cycle, and n is the adjustment parameter;

[0055] Each new acquisition node is sequentially inserted into the adjustment cycle to optimize the initial acquisition parameters;

[0056] Return to randomly select an adjustment cycle until all adjustment cycles in the acquisition environment are selected, and obtain the acquisition parameters after the acquisition environment is optimized;

[0057] Return to randomly select a collection environment until all collection environments are selected, and obtain the optimized collection parameters of each collection environment.

[0058] Preferably, the initial acquisition parameters are continuously optimized in combination with the predicted crop information output by the trained acquisition parameter model to obtain optimized acquisition parameters, and the crop growth information is collected according to the optimized acquisition parameters, further comprising:

[0059] Select a collection environment;

[0060] Obtaining the optimized acquisition parameters of the acquisition environment;

[0061] Collect the crop growth information in the collection environment based on the optimized collection parameters, and put all the collected crop growth information into the crop growth information table;

[0062] Return to select a collection environment until all collection environments are selected.

[0063] On the other hand, the present application also provides an intelligent crop growth environment monitoring and control system based on the Internet of Things, including:

[0064] A collection component, the collection component includes a plurality of sensors, and the collection component is used to collect the ambient temperature, ambient humidity and soil pH value of the crop growth environment;

[0065] A processing component is communicatively connected with the acquisition component, and the processing component includes a central processing unit, a data analysis module and an automatic control module. The data analysis module analyzes the acquired data and triggers corresponding devices of the automatic control module to adjust initial acquisition parameters.

[0066] Compared with the prior art, the above technical solution of the present invention has the following beneficial technical effects:

[0067] By setting corresponding initial collection parameters for each collection environment, the crop growth information in each collection environment is collected according to the initial collection parameters of each collection environment, and then unqualified crop growth information is screened out based on the crop growth information, and then the unqualified crop growth information is input into the collection parameter model to train the collection parameter model to obtain the trained collection parameter model, and finally the initial collection parameters are continuously optimized in combination with the predicted crop information output by the trained collection parameter model to obtain the optimized collection parameters, and the crop growth information is collected according to the optimized collection parameters. The present application improves the collection frequency of unqualified crop growth information by continuously optimizing the collection parameters, so as to collect more abnormal data and facilitate finding the root cause of the unqualified crop growth information. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 This is a flow chart of the intelligent crop growth environment monitoring and control method based on the Internet of Things proposed by the present invention;

[0069] Figure 2 This is a schematic diagram of the structure of the intelligent crop growth environment monitoring and control system based on the Internet of Things proposed by the present invention.

[0070] Reference numerals:

[0071] 100, acquisition component; 101, sensor; 200, processing component; 201, central processing unit;

[0072] 202. Data analysis module; 203. Automatic control module. DETAILED DESCRIPTION

[0073] Embodiment 1, as Figure 1 As shown, the intelligent crop growth environment monitoring and control method based on the Internet of Things proposed by the present invention includes:

[0074] S100, setting corresponding initial collection parameters for each collection environment, and collecting crop growth information in each collection environment according to the initial collection parameters of each collection environment;

[0075] Specifically, the collection environment can be divided according to the different types of crops planted therein, or a separate collection space can be used as an independent collection environment; when dividing the collection environment, each collection environment should be able to reflect the growth information characteristics of the crops it contains.

[0076] For example, there is a collection area in which crops A, B and C are planted, and there are 10 independent planting spaces for crop A, 8 independent planting spaces for crop B, and 13 independent planting spaces for crop C. When dividing the collection environment, the collection area can be divided into three collection environments, namely, A collection environment, B collection environment and C collection environment, according to the types of crops. Each independent planting space can also be used as an independent collection environment, and the collection area can be divided into 31 collection environments, namely, A-1 collection environment to A-10 collection environment, B-1 collection environment to B-8 collection environment, and C-1 collection environment to C-13 collection environment.

[0077] S200, screening out unqualified crop growth information based on the crop growth information;

[0078] S300, creating an acquisition parameter model;

[0079] S400, inputting unqualified crop growth information into a collection parameter model to train the collection parameter model, and obtaining a trained collection parameter model; the trained collection parameter model has the ability to output predicted unqualified crop growth information according to the input collection environment and initial collection parameters, and the predicted unqualified crop growth information output by the trained collection parameter model is recorded as predicted crop information;

[0080] S500, continuously optimizing the initial acquisition parameters in combination with the predicted crop information output by the trained acquisition parameter model to obtain optimized acquisition parameters, and acquiring crop growth information according to the optimized acquisition parameters.

[0081] In the present invention, corresponding initial collection parameters are set for each collection environment, and crop growth information in each collection environment is collected according to the initial collection parameters of each collection environment. Then, unqualified crop growth information is screened out based on the crop growth information, and then the unqualified crop growth information is input into a collection parameter model to train the collection parameter model to obtain a trained collection parameter model. Finally, the initial collection parameters are continuously optimized in combination with the predicted crop information output by the trained collection parameter model to obtain optimized collection parameters, and crop growth information is collected according to the optimized collection parameters. The present application improves the collection frequency of unqualified crop growth information by continuously optimizing the collection parameters, so as to collect more abnormal data and facilitate finding the root cause of the unqualified crop growth information.

[0082] In an optional embodiment, the S100 includes:

[0083] S110, creating a crop growth information table;

[0084] S120, selecting a collection environment;

[0085] S130, setting initial acquisition parameters of the acquisition environment; the initial acquisition parameters include an initial acquisition cycle and an initial acquisition node;

[0086] S140, collecting a plurality of crop growth information based on the initial collection node within the initial collection period, and putting all the collected crop growth information into a crop growth information table; the crop growth information includes environmental humidity, environmental temperature, and soil pH value;

[0087] Specifically, one crop growth information should include multiple test items such as environmental humidity, environmental temperature, and soil pH value, that is, "one crop growth information" is actually the data of multiple test items collected in one collection;

[0088] S150, return to S120, until all the collection environments are selected, and the crop growth information of each collection environment is obtained.

[0089] It should be noted that, since each collection environment is independent, when setting the initial collection parameters for each collection environment, the initial collection parameters should also be set separately for each collection environment.

[0090] Regardless of the initial collection parameters or the optimized collection parameters, the collection nodes include a start point and an end point, that is, the collection node is actually a period of time between the start point and the end point, and the span of the collection cycle should at least include a complete growth cycle of the crop. Therefore, multiple collection nodes can be set within a collection cycle to ensure that the complete growth information of the crop during the growth process can be collected.

[0091] For example, for crops A and B, the growth cycle of crop A is 28 days, and the growth cycle of crop B is 15 days. Then for the collection environment of crop A, its initial collection parameters can be set accordingly: the initial collection cycle is 30 days, and the starting collection node of the initial collection node is set to 6:00 to 8:00 and 16:00 to 18:00 every day in the initial collection cycle; for the collection environment of crop B, its initial collection parameters can be set accordingly: the initial collection cycle is 18 days, and the initial collection nodes are 7:00 to 9:00 and 17:00 to 19:00 every day in the initial collection cycle.

[0092] In an optional embodiment, the step S200 includes:

[0093] S210, selecting a collection environment;

[0094] S220, setting a crop growth standard threshold range for each initial collection node of the collection environment; the crop growth standard threshold range includes a humidity standard threshold range, a temperature standard threshold range and a pH value standard threshold range;

[0095] Specifically, the smaller the crop growth standard threshold range is set, the stricter the detection of crop growth information is;

[0096] As crops grow, the crop growth information collected at each collection node will change with the growth of crops, so the crop growth standard threshold range should also be set corresponding to each collection node;

[0097] S230, selecting an initial collection cycle;

[0098] S240, randomly selecting a piece of crop growth information within the initial collection period from the crop growth information table;

[0099] S250, determining whether the crop growth information is within a crop growth standard threshold range;

[0100] S260, if the crop growth information is within the crop growth standard threshold range, the crop growth information is qualified;

[0101] S270, returning to randomly selecting a piece of crop growth information within the initial collection period from the crop growth information table until all crop growth information within the initial collection period has been selected;

[0102] S280, returning to select an initial collection cycle until all initial collection cycles are selected;

[0103] S290, returning to select a collection environment until all collection environments are selected;

[0104] Specifically, in the actual collection process, multiple collection environments may be collecting at the same time, rather than sequentially.

[0105] It should be noted that, because the actual process of collecting crop growth information is accompanied by changes in the environment and climate, there will be a certain error between the actual collected crop growth information and the standard value. Therefore, a standard threshold range for crop growth is set instead of a standard value in order to retain a certain error.

[0106] When collecting crop growth information, the collection environment is often determined first, and then the crop growth information corresponding to each initial collection node is collected in turn under the collection environment. In this way, multiple crop growth information can be collected within an initial collection cycle, and similarly, crop growth information under multiple collection environments can be collected.

[0107] After the crop growth information is collected, the crop growth information is compared with the crop growth standard threshold range to determine whether there is unqualified crop growth information in the collection cycle. Only when the crop growth information corresponding to each collection node in a collection cycle is within the crop growth standard threshold range, the collection cycle is qualified. Only when all collection cycles in a collection environment are qualified, the collection environment is qualified.

[0108] In an optional embodiment, the S200 further includes:

[0109] S291, if at least one of the crop growth information is not within the crop growth standard threshold range, the crop growth information is unqualified;

[0110] S292, returning to randomly select a piece of crop growth information within the initial collection period from the crop growth information table until all crop growth information within the initial collection period has been selected;

[0111] S293, returning to select an initial collection cycle until all initial collection cycles are selected;

[0112] S294, returning to select a collection environment until all collection environments are selected;

[0113] S295, screening out unqualified crop growth information contained in all the collection environments;

[0114] It should be noted that a crop growth information contains multiple detection items. As long as at least one detection item in a crop growth information fails, the crop growth information is unqualified. When the initial collection parameters are subsequently optimized, the initial collection node and initial collection cycle corresponding to the crop growth information need to be optimized.

[0115] In an optional embodiment, the S250 includes:

[0116] S251, determining whether the ambient humidity is within a humidity standard threshold range;

[0117] S252, if the ambient humidity is within the humidity standard threshold range, determining whether the ambient temperature is within the temperature standard threshold range;

[0118] S253, if the ambient temperature is within the temperature standard threshold range, determining whether the soil pH value is within the pH standard threshold range;

[0119] Optionally, the crop growth information is not limited to the three detection items of ambient humidity, ambient humidity and soil pH value, and other detection items such as light intensity may also be set. Each time a detection item is set, it is necessary to add the detection of the detection item when executing the step of "determining whether the crop growth information is within the standard threshold range of crop growth".

[0120] It should be noted that the three judgment steps in this embodiment do not have a chronological order. Step S251-Step S253 are merely a specific embodiment provided for step S250. In the actual implementation process, each detection item in the crop growth information is equally important. As long as any detection item fails, the corresponding crop growth information is unqualified.

[0121] In an optional embodiment, the S400 includes:

[0122] S410, obtaining a collection environment corresponding to each piece of unqualified crop growth information and an initial collection node corresponding to each piece of unqualified crop growth information; recording one piece of unqualified crop growth information, the collection environment corresponding to the unqualified crop growth information, and the initial collection node corresponding to the unqualified crop growth information as a collection sample;

[0123] S420, creating an acquisition parameter model;

[0124] S430, dividing all collected samples into a training set and a test set in a random ratio;

[0125] S440, inputting the collection samples in the training set into the collection parameter model in sequence to train the collection parameter model, and obtaining a trained collection parameter model;

[0126] S450, inputting the collected samples in the test set into the trained collection parameter model in sequence to verify whether the trained collection parameter model is trained;

[0127] Specifically, step S450 includes:

[0128] S451, setting a pass rate threshold;

[0129] S452, randomly selecting unqualified crop growth information and a collection environment corresponding to the unqualified crop growth information from the collection samples in the training set;

[0130] S453, inputting the unqualified crop growth information and the collection environment corresponding to the unqualified crop growth information into the trained collection parameter model;

[0131] S454, obtaining the predicted acquisition node output by the trained acquisition parameter model;

[0132] S455, determining whether the predicted collection node is the same as the initial collection node corresponding to the unqualified crop growth information in the training set;

[0133] S456, if the predicted collection node is the same as the initial collection node corresponding to the unqualified crop growth information in the training set, the qualified number n is set to n+1, and the process returns to S452 until all the collection samples in the training set are selected;

[0134] Specifically, the initial qualified number = 0;

[0135] S457, determining whether n / N is greater than or equal to a pass rate threshold;

[0136] Specifically, n is the qualified number, and N is the total number of unqualified crop growth information in the training set;

[0137] S458, if n / N is greater than or equal to the pass rate threshold, the acquisition parameter model training is completed;

[0138] S489, if n / N is less than the pass rate threshold, return to S430 until n / N is greater than or equal to the pass rate threshold;

[0139] It should be noted that after executing step S250, unqualified crop growth information can be obtained, and then the unqualified crop growth information is used to train the acquisition parameter model, so that the trained acquisition parameter model has the ability to output predicted unqualified crop growth information based on the input acquisition environment and initial acquisition parameters.

[0140] In an optional embodiment, the S440 includes:

[0141] S441, selecting a collection environment and obtaining all collection samples in the collection environment;

[0142] S442, counting the number of times of unqualified crop growth information included in each collection cycle in the collection environment;

[0143] S443, sorting the collection cycles from large to small according to the number of times of unqualified crop growth information contained in each collection cycle, selecting the first N collection cycles; and recording the first N unqualified collection cycles as adjustment cycles;

[0144] S444, returning to S441, until all the acquisition environments in the training set are selected, and multiple adjustment cycles for each acquisition environment are obtained;

[0145] It should be noted that by sorting the number of times of unqualified crop growth information from large to small, several initial collection cycles that are most in need of optimization are selected.

[0146] Optionally, for an initial collection period in which no unqualified crop growth information appears, the initial collection period may be recorded as a trusted collection period, and the number of initial collection nodes included in the trusted collection period may be reduced, thereby improving collection efficiency;

[0147] Specifically, the following steps are performed to reduce the initial collection nodes in the trust collection cycle:

[0148] K050, select a trust collection cycle;

[0149] K051, obtain all initial collection nodes of the trust collection cycle;

[0150] K052, sort all initial collection nodes in order according to the start time of the collection node;

[0151] K053, starting from the first initial collection node, delete an initial collection node every K initial collection nodes;

[0152] K054, return to select a trust collection cycle until all trust collection cycles are selected and completed.

[0153] In an optional embodiment, the S500 includes:

[0154] S510, randomly selecting a collection environment;

[0155] Specifically, the collection environment selected in step S510 is a collection environment including an adjustment cycle. If the collection environment does not include an adjustment cycle, then the collection environment does not need to add a new collection node;

[0156] S520, randomly selecting an adjustment period in the acquisition environment and obtaining a start node and an end node of the adjustment period;

[0157] S530, creating a new acquisition node based on the start node and the end node of the adjustment cycle by using Formula 1;

[0158] (0,n)Formula 1;

[0159] in, is the i-th new acquisition node, is the starting node of the adjustment cycle. is the end node of the adjustment cycle, and n is the adjustment parameter;

[0160] Specifically, the larger the adjustment parameter is set, the smaller the interval between new acquisition nodes will be;

[0161] Optionally, in addition to inserting a new acquisition node in the adjustment cycle, a new acquisition sub-node may be inserted in each acquisition node, thereby increasing the acquisition frequency of the acquisition node;

[0162] Create a new collection subnode by performing the following steps:

[0163] K100, randomly selects a collection node corresponding to the unqualified crop growth information from the adjustment period;

[0164] K200, based on the start collection node and the end collection node of the collection node, a new collection sub-node is created by formula 2;

[0165] Formula 1;

[0166] in, is the i-th new collection child node, is the end collection node of the i-th new collection sub-node, is the starting collection node of the i-th new collection sub-node, is the start collection node of this collection node, is the end collection node of the collection node, and m is the adjustment sub-parameter;

[0167] K300, inserts each new acquisition sub-node into the acquisition node in sequence to optimize the acquisition node;

[0168] K400, return to K100, until all the collection nodes corresponding to the unqualified crop growth information in the adjustment period are selected;

[0169] S540, inserting each new acquisition node into the adjustment cycle in sequence to optimize the initial acquisition parameters;

[0170] S550, returning to randomly select an adjustment cycle until all adjustment cycles in the acquisition environment are selected, and obtaining the acquisition parameters after the acquisition environment is optimized;

[0171] S560, return to S510, until all collection environments are selected, and the optimized collection parameters of each collection environment are obtained.

[0172] It should be noted that since the collection environment includes an adjustment cycle, it proves that there is unqualified crop growth information in the collection environment. In order to find out the problem of this unqualified crop growth information, the number of collection nodes in the adjustment cycle is increased, thereby increasing the number of collection times in a collection cycle, so as to increase the amount of collected crop growth information, and then increase the amount of available data, so as to analyze the root cause of the unqualified crop growth information.

[0173] In addition, this embodiment also provides specific steps for inserting a new collection sub-node in each collection node. According to K100-K400, multiple new collection sub-nodes can be inserted into the collection nodes corresponding to all unqualified crop growth information in one adjustment cycle, thereby further improving the collection frequency of the collection nodes.

[0174] In an optional embodiment, the S500 further includes:

[0175] S560, selecting a collection environment;

[0176] S570, obtaining the collection parameters after the collection environment is optimized;

[0177] S580, collecting crop growth information in the collection environment based on the optimized collection parameters, and putting all collected crop growth information into a crop growth information table;

[0178] S590, return to S560, until all collection environments are selected;

[0179] It should be noted that after obtaining the optimized collection parameters through the above-mentioned embodiment, it is necessary to collect the crop growth information of each collection environment according to the optimized collection parameters, so as to improve the monitoring efficiency of the time nodes where unqualified crop growth information is likely to appear.

[0180] Embodiment 2, as Figure 2 As shown, the intelligent crop growth environment monitoring and control system based on the Internet of Things proposed by the present invention includes a collection component 100 and a processing component 200.

[0181] The collection component 100 includes multiple sensors 101, and the collection component 100 is used to collect the ambient temperature, ambient humidity and soil pH value of the crop growth environment; the processing component 200 is communicated with the collection component 100, and the processing component 200 includes a central processing unit 201, a data analysis module 202 and an automatic control module 203. The data analysis module 202 analyzes the collected data and triggers the corresponding equipment of the automatic control module 203 to adjust the collection parameters.

[0182] Specifically, the multiple sensors 101 in the present application are respectively configured as sensors 101 of different models to collect data of different detection items in the crop growth information.

[0183] It should be noted that in the present application, multiple sensors 101 are used to collect data of different detection items in the crop growth information, and all the collected data are transmitted to the processing component 200. The central processing unit 201 in the processing component 200 executes the intelligent crop growth environment monitoring and control method based on the Internet of Things described in Example 1, thereby controlling the data analysis module 202 to analyze the collected data input by the collection component 100, and adjusts the automatic control module 203 according to the analysis results to adjust the initial collection parameters to obtain optimized collection parameters.

[0184] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited thereto, and various changes can be made within the knowledge scope of technicians in the relevant technical field without departing from the purpose of the present invention.

Claims

1. An intelligent crop growth environment monitoring and control method based on the Internet of Things, characterized in that: include: Setting corresponding initial collection parameters for each collection environment, and collecting crop growth information in each collection environment according to the initial collection parameters of each collection environment; Screen out unqualified crop growth information based on crop growth information; Create acquisition parameter model; Inputting unqualified crop growth information into a collection parameter model to train the collection parameter model to obtain a trained collection parameter model; the trained collection parameter model has the ability to output predicted unqualified crop growth information according to the input collection environment and initial collection parameters, and the predicted unqualified crop growth information output by the trained collection parameter model is recorded as predicted crop information; The step of inputting unqualified crop growth information into a collection parameter model to train the collection parameter model and obtain a trained collection parameter model comprises: Select a collection environment and obtain all collection samples in the collection environment; Count the number of unqualified crop growth information included in each collection cycle under the collection environment; The collection cycles are sorted from large to small according to the number of times of unqualified crop growth information contained in each collection cycle, and the first N collection cycles are selected; the first N unqualified collection cycles are recorded as adjustment cycles; Return to select a collection environment and obtain all collection samples in the collection environment until all collection environments in the training set are selected, and obtain multiple adjustment cycles for each collection environment; The initial acquisition parameters are continuously optimized based on the predicted crop information output by the trained acquisition parameter model to obtain optimized acquisition parameters, and crop growth information is collected based on the optimized acquisition parameters; The method of continuously optimizing the initial acquisition parameters based on the predicted crop information output by the trained acquisition parameter model to obtain optimized acquisition parameters, and acquiring crop growth information based on the optimized acquisition parameters, includes: Randomly select a collection environment; Randomly select an adjustment period under the acquisition environment and obtain the start node and end node of the adjustment period; Based on the start node and the end node of the adjustment cycle, a new acquisition node is created by formula 1; (0,n) Formula 1; in, is the i-th new acquisition node, is the starting node of the adjustment cycle. is the end node of the adjustment cycle, and n is the adjustment parameter; Each new acquisition node is sequentially inserted into the adjustment cycle to optimize the initial acquisition parameters; Return to randomly select an adjustment cycle until all adjustment cycles in the acquisition environment are selected, and obtain the acquisition parameters after the acquisition environment is optimized; Return to randomly select a collection environment until all collection environments are selected, and obtain the optimized collection parameters of each collection environment.

2. The method for monitoring and controlling the growth environment of crops based on the Internet of Things according to claim 1, characterized in that: Set corresponding initial collection parameters for each collection environment, and collect crop growth information in each collection environment according to the initial collection parameters of each collection environment, including: Create a crop growth information table; Select a collection environment; Set the initial collection parameters of the collection environment; the initial collection parameters include the initial collection period and the initial collection node; Collecting a plurality of crop growth information based on the initial collection node within the initial collection period, and putting all the collected crop growth information into a crop growth information table; the crop growth information includes environmental humidity, environmental temperature, and soil pH value; Return to select a collection environment until all collection environments are selected, and obtain the crop growth information of each collection environment.

3. The method for monitoring and controlling the growth environment of crops based on the Internet of Things according to claim 2 is characterized in that: Based on the crop growth information, unqualified crop growth information is screened out, including: Select a collection environment; Setting a crop growth standard threshold range for each initial collection node of the collection environment; the crop growth standard threshold range includes a humidity standard threshold range, a temperature standard threshold range and a pH value standard threshold range; Select an initial collection period; Randomly select a crop growth information within the initial collection period from the crop growth information table; Determining whether the crop growth information is within a crop growth standard threshold range; If the crop growth information is within the crop growth standard threshold range, the crop growth information is qualified; Return to randomly select a crop growth information within the initial collection period from the crop growth information table until all crop growth information within the initial collection period has been selected; Return to select an initial collection cycle until all initial collection cycles are selected; Return to select a collection environment until all collection environments are selected.

4. The method for monitoring and controlling the growth environment of crops based on the Internet of Things according to claim 3 is characterized in that: Screening out unqualified crop growth information based on crop growth information also includes: If at least one of the crop growth information is not within the crop growth standard threshold range, the crop growth information is unqualified; Return to randomly select a crop growth information within the initial collection period from the crop growth information table until all crop growth information within the initial collection period has been selected; Return to select an initial collection cycle until all initial collection cycles are selected; Return to select a collection environment until all collection environments have been selected; Filter out all unqualified crop growth information contained in the collection environment.

5. The method for monitoring and controlling the growth environment of crops based on the Internet of Things according to claim 4 is characterized in that: Determining whether the crop growth information is within the crop growth standard threshold range includes: Determine whether the ambient humidity is within the humidity standard threshold range; If the ambient humidity is within the humidity standard threshold range, then determining whether the ambient temperature is within the temperature standard threshold range; If the ambient temperature is within the temperature standard threshold range, it is determined whether the soil pH value is within the pH standard threshold range.

6. The method for monitoring and controlling the intelligent crop growth environment based on the Internet of Things according to claim 5 is characterized in that: Inputting unqualified crop growth information into the acquisition parameter model to train the acquisition parameter model, and obtaining a trained acquisition parameter model, including: Obtaining a collection environment corresponding to each piece of unqualified crop growth information and an initial collection node corresponding to each piece of unqualified crop growth information; recording one piece of unqualified crop growth information, the collection environment corresponding to the unqualified crop growth information, and the initial collection node corresponding to the unqualified crop growth information as a collection sample; Create acquisition parameter model; All collected samples are divided into training sets and test sets in random proportions; Inputting the acquisition samples in the training set into the acquisition parameter model in sequence to train the acquisition parameter model, thereby obtaining a trained acquisition parameter model; The collected samples in the test set are input into the trained collection parameter model in sequence to verify whether the trained collection parameter model is trained.

7. The method for monitoring and controlling the growth environment of crops based on the Internet of Things according to claim 6, characterized in that: The initial acquisition parameters are continuously optimized in combination with the predicted crop information output by the trained acquisition parameter model to obtain optimized acquisition parameters, and crop growth information is collected according to the optimized acquisition parameters, which also includes: Select a collection environment; Obtaining the optimized acquisition parameters of the acquisition environment; Collect the crop growth information in the collection environment based on the optimized collection parameters, and put all the collected crop growth information into the crop growth information table; Return to select a collection environment until all collection environments are selected.

8. The intelligent crop growth environment monitoring and control system based on the Internet of Things is characterized by: include: A collection component, the collection component includes a plurality of sensors, and the collection component is used to collect the ambient temperature, ambient humidity and soil pH value of the crop growth environment; A processing component, wherein the processing component is communicatively connected with the acquisition component, the processing component comprises a central processing unit, a data analysis module and an automatic control module, the data analysis module analyzes the acquired data and triggers the corresponding device of the automatic control module to adjust the initial acquisition parameters, and the intelligent crop growth environment monitoring and control method based on the Internet of Things as described in any one of claims 1 to 7 is executed through the processing component.

Images