A greenhouse temperature automatic control system based on temperature early warning

By introducing a temperature warning system in the greenhouse, using temperature frequent matching and temperature curve prediction, the temperature control parameters are adjusted in real time, the problem of temperature regulation hysteresis in the traditional greenhouse temperature control system is solved, and more efficient temperatures are achieved control.

CN118760277BActive Publication Date: 2024-11-15GUANGDONG WANQINGYANG AGRI DEV CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411253518.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-09
Publication Date
2024-11-15
Estimated Expiration
2044-09-09

AI Technical Summary

Technical Problem

Due to the real-time monitoring of the temperature control system in traditional greenhouses, the temperature regulation is not timely, and there is a problem of temperature regulation hysteresis.

Method used

Design a greenhouse temperature automatic control system based on temperature warning, adopting greenhouse temperature prediction components and temperature configuration components, and extracting the abnormal prediction time zone through temperature frequent matching, temperature collection, control parameter adjustment and temperature curve prediction, and according to the outside of the shed Adjust the control parameters of the temperature control port with the temperature information in the shed to achieve pre-emptive temperature control.

Benefits of technology

The hysteresis of temperature regulation is reduced, the prerequisite for temperature control in the shed is improved, and the temperature remains stable within the crop's favorite range.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118760277B_ABST
    Figure CN118760277B_ABST
Patent Text Reader

Abstract

The present invention relates to a greenhouse temperature automatic control system based on temperature early warning, and relates to the field of temperature control, including: receiving information on planted crop varieties and crop planting duration information, and obtaining an expected range of ambient temperature; when the time series information of the greenhouse monitoring temperature belongs to the expected range of ambient temperature, and the time series information of the outside greenhouse monitoring temperature does not belong to the outside greenhouse reference temperature range, through a greenhouse temperature prediction component, obtaining a preset time zone greenhouse temperature prediction curve, extracting an abnormal prediction time zone in which the preset time zone greenhouse temperature prediction curve does not belong to the expected range of ambient temperature; when the duration of the abnormal prediction time zone is greater than or equal to the abnormal duration threshold, obtaining a plurality of temperature control ports recommended air supply control parameters, and initializing a temperature control device to control ambient temperature. The technical problem of hysteresis in the prior art due to real-time monitoring of greenhouse temperature and real-time adjustment is solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of temperature control, and in particular to an automatic temperature control system for a greenhouse based on temperature early warning. Background Art

[0002] Temperature control can ensure that the temperature in the greenhouse is maintained within the range preferred by crops. However, the temperature in the greenhouse is affected by external factors, which requires constant adjustment of the temperature control device to keep the temperature in the greenhouse constant. In traditional monitoring, the real-time temperature in the greenhouse is collected through temperature sensors. When an abnormality occurs, the temperature control device is started to adjust it, which has serious lag and leads to the problem of untimely temperature adjustment. Summary of the invention

[0003] The present invention aims to solve the technical problem in the prior art of hysteresis due to real-time monitoring of the temperature in the greenhouse and then real-time adjustment, and provides a greenhouse temperature automatic control system based on temperature early warning to solve the problem.

[0004] The technical solution of the present invention to solve the above technical problems is as follows:

[0005] In the first aspect, the present invention provides a greenhouse temperature automatic control system based on temperature early warning, which is deployed in a temperature control device automatic controller, the temperature control device automatic controller includes a greenhouse temperature prediction component and a greenhouse temperature configuration component, the temperature control device automatic controller and the temperature control device are communicatively connected, and the temperature control device has a plurality of temperature control ports in the greenhouse, including:

[0006] A temperature frequency matching unit is used to receive information about planted crop varieties and crop planting duration information to perform temperature frequency matching and obtain an expected range of ambient temperature;

[0007] The temperature acquisition unit is used to receive the time series information of the temperature monitored inside the greenhouse through the internal temperature sensor, and receive the time series information of the temperature monitored outside the greenhouse and the reference temperature range outside the greenhouse through the external temperature sensor;

[0008] A control parameter retrieving unit, used to obtain a number of temperature control port air supply control parameters through a temperature control device;

[0009] A temperature curve prediction unit is used to process the time series information of the temperature monitoring inside the greenhouse, the time series information of the temperature monitoring outside the greenhouse and the air supply control parameters of the temperature regulating port through a greenhouse temperature prediction component to obtain a greenhouse temperature prediction curve in a preset time zone when the time series information of the temperature monitoring inside the greenhouse belongs to the expected ambient temperature range and the time series information of the temperature monitoring outside the greenhouse does not belong to the reference temperature range outside the greenhouse;

[0010] An abnormal time zone extraction unit is used to extract the abnormal prediction time zone in which the temperature prediction curve in the greenhouse in the preset time zone does not belong to the expected range of the ambient temperature;

[0011] A control parameter adjustment unit, for adjusting the air supply control parameters of the temperature regulating ports according to the time series information of the temperature monitored outside the greenhouse and the time series information of the temperature monitored inside the greenhouse through a greenhouse temperature configuration component and the greenhouse temperature prediction component when the time length of the abnormal prediction time zone is greater than or equal to the abnormal time length threshold, to obtain the recommended air supply control parameters of the temperature regulating ports;

[0012] The temperature control unit in the greenhouse is used to initialize the temperature control device to control the ambient temperature according to the recommended air supply control parameters of the several temperature control ports, and at the same time update the reference temperature range outside the greenhouse according to the timing information of the temperature monitored outside the greenhouse.

[0013] In a second aspect, the present invention provides an electronic device, comprising:

[0014] Memory for storing computer software programs;

[0015] The processor is used to read and execute the computer software program, thereby realizing the greenhouse temperature automatic control system based on temperature warning described in the first aspect.

[0016] In a third aspect, the present invention provides a non-transitory computer-readable storage medium, in which a computer software program is stored. When the computer software program is executed by a processor, the greenhouse temperature automatic control system based on temperature warning described in the first aspect is implemented.

[0017] The beneficial effects of the present invention are as follows: by analyzing plant varieties and growth stages through big data, an expected range of environmental temperatures preferred by crops is configured; temperature sensors inside and outside the greenhouse are then used to collect temperature monitoring time series information inside and outside the greenhouse; if the greenhouse monitoring temperature time series information belongs to the expected range of environmental temperatures, and the outside monitoring temperature time series information does not belong to the outside reference temperature range, the greenhouse temperature prediction component is used to process the greenhouse monitoring temperature time series information, the outside monitoring temperature time series information and the several temperature regulating port air supply control parameters to obtain a greenhouse temperature prediction curve for a preset time zone; then, according to the greenhouse temperature prediction curve for the preset time zone, ; Extract the abnormal prediction time zone in which the greenhouse temperature prediction curve in the preset time zone does not belong to the expected range of the ambient temperature; when the duration of the abnormal prediction time zone is greater than or equal to the abnormal duration threshold, the several temperature control port air supply control parameters are adjusted according to the external greenhouse monitoring temperature timing information and the internal greenhouse monitoring temperature timing information through the greenhouse temperature configuration component and the greenhouse temperature prediction component, and several temperature control port recommended air supply control parameters are obtained. The temperature control device is initialized according to the several temperature control port recommended air supply control parameters to control the ambient temperature, and the external greenhouse reference temperature range is updated according to the external greenhouse monitoring temperature timing information. The greenhouse temperature prediction component is used to predict the change trend of the greenhouse temperature, and then the temperature control port control parameters are adjusted according to the change trend, so as to reduce the hysteresis of temperature regulation and achieve the technical effect of improving the predictability of greenhouse temperature control. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A schematic diagram of the structure of a greenhouse temperature automatic control system based on temperature early warning provided by the present invention;

[0019] Figure 2 A schematic diagram of the construction of a greenhouse temperature prediction component for automatic control of greenhouse temperature based on temperature early warning provided by the present invention;

[0020] Figure 3 A schematic diagram of the structure of an electronic device provided by the present invention;

[0021] Figure 4 A schematic diagram of the structure of a computer-readable storage medium provided by the present invention.

[0022] In the accompanying drawings, the components represented by the reference numerals are listed as follows:

[0023] Electronic device 500 , memory 510 , processor 520 , computer program 511 , computer readable storage medium 600 , computer program 611 . DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0025] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0026] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or explanation". Any embodiment described as "for example" in the present invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any technician in the field to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in the present invention.

[0027] Embodiment 1:

[0028] like Figure 1 As shown, the embodiment of the present invention provides a greenhouse temperature automatic control system based on temperature warning, which is deployed in a temperature control device automatic controller, the temperature control device automatic controller includes a greenhouse temperature prediction component and a greenhouse temperature configuration component, the temperature control device automatic controller is connected to the temperature control device in communication, and the temperature control device has several temperature control ports in the greenhouse, including:

[0029] Specifically, the temperature control device automatic controller is software used to control the temperature control device. The temperature control device has multiple air supply outlets in the greenhouse, that is, several temperature control outlets, which can supply airflow with adjustable direction, speed and temperature to regulate the temperature in the greenhouse.

[0030] A temperature frequency matching unit is used to receive information about planted crop varieties and crop planting duration information to perform temperature frequency matching and obtain an expected range of ambient temperature;

[0031] Specifically, the crop variety information is used to characterize the types of crops planted in the greenhouse; the crop planting time information is used to characterize the planting time of crops planted in the greenhouse from sowing to the current time; the expected ambient temperature range refers to the ideal temperature of the crop variety information at the current time, which can ensure the healthy growth of crops.

[0032] The embodiment of the present application implements the configuration of the expected range of ambient temperature through a temperature frequency matching algorithm, as detailed as follows:

[0033] Further, receiving the crop variety information and the crop planting time information to perform temperature frequency matching and obtain the expected range of the ambient temperature, including the steps of:

[0034] Input the planting crop variety information into a crop growth cycle calibration table, and output a crop growth stage sequence, wherein the crop growth stage sequence has a crop planting time label sequence;

[0035] According to the crop planting time label sequence and the crop growth stage sequence, the crop planting time information is calibrated to obtain the crop growth calibration stage;

[0036] Positive samples are collected according to the planted crop variety information and the crop growth calibration stage to obtain a crop cultivation temperature record time series data set;

[0037] Traversing the crop cultivation temperature record time series data set to perform neighborhood temperature aggregation to obtain a plurality of crop cultivation characteristic temperatures of the crop planting time information;

[0038] A central trend analysis is performed on the multiple crop cultivation characteristic temperatures to construct the expected range of the environmental temperature.

[0039] Specifically, the crop growth cycle calibration table is a data table of crop planting time labels and crop growth stages that are calibrated in advance by greenhouse managers for crop types in the greenhouse. The crop types in the crop growth cycle calibration table can be indexed by planting crop variety information, thereby extracting the one-to-one corresponding crop planting time label sequence and crop growth stage sequence. Furthermore, the crop planting time label sequence is indexed according to the crop planting time information, thereby extracting the matching crop growth stage and storing it as the crop growth calibration stage.

[0040] Furthermore, samples of healthy growth of planted crop varieties during the crop growth calibration phase are collected to obtain a crop cultivation temperature record time series data set for the corresponding phase. Any crop cultivation temperature record time series data stores the cultivation temperature time series record data of a sample during the crop growth calibration phase. First, neighborhood temperature aggregation is performed on each crop cultivation temperature record time series data set to obtain the neighborhood temperature aggregation result of each crop cultivation temperature record time series data, and then multiple crop cultivation characteristic temperatures that meet the crop planting time information are extracted. Furthermore, a central trend analysis is performed on multiple crop cultivation characteristic temperatures to generate an expected range of ambient temperature.

[0041] Preferably, the central tendency analysis algorithm process is as follows:

[0042] Step 1: sorting multiple crop cultivation characteristic temperatures in ascending order to obtain a crop cultivation characteristic temperature sorting result;

[0043] Step 2: Use 0.25*the total number of the characteristic temperatures for growing crops to round up to get the first serial number; then use 0.75*the total number of the characteristic temperatures for growing crops to round down to get the second serial number;

[0044] Step 3: taking values ​​from the crop cultivation characteristic temperature sorting results according to the first sequence number and the second sequence number to obtain the first sequence number characteristic value and the second sequence number characteristic value;

[0045] Step 4: Subtract the eigenvalue of the first sequence number from the eigenvalue of the second sequence number to obtain the eigenvalue difference;

[0046] Step 5: Use the first sequence number eigenvalue - 1.5 * eigenvalue difference to get the first outlier boundary, and use the second sequence number eigenvalue + 1.5 * eigenvalue difference to get the second outlier boundary;

[0047] Step 6: extracting a set of cultivation characteristic temperatures greater than or equal to the first outlier boundary and less than or equal to the second outlier boundary from the plurality of crop cultivation characteristic temperatures;

[0048] Step 7: Set the distribution range of the cultivation characteristic temperature set to the expected range of the ambient temperature.

[0049] Furthermore, the neighborhood temperature aggregation is a customized algorithm of the embodiment of the present application, and the specific process is as follows:

[0050] Traversing the crop cultivation temperature record time series data set to perform neighborhood temperature aggregation, and obtaining multiple crop cultivation characteristic temperatures of the crop planting time information, including:

[0051] Extracting first crop cultivation temperature record time series data according to the crop cultivation temperature record time series data set;

[0052] Extracting a first time zone crop cultivation temperature record value and a second time zone crop cultivation temperature record value according to the first crop cultivation temperature record time series data, wherein the first time zone and the second time zone are adjacent time zones;

[0053] When the crop cultivation temperature record value deviation between the crop cultivation temperature record value in the first time zone and the crop cultivation temperature record value in the second time zone is less than or equal to the crop cultivation temperature record value deviation threshold, merging the first time zone and the second time zone into a third time zone;

[0054] The crop cultivation temperature record value in the first time zone and the crop cultivation temperature record value in the second time zone are integrated to obtain the crop cultivation temperature record value in the third time zone;

[0055] Circular analysis, when the deviation of the crop cultivation temperature record values ​​in any two adjacent time zones is greater than the crop cultivation temperature record value deviation threshold, output the first crop cultivation characteristic temperature of the crop planting time information, and add it to the multiple crop cultivation characteristic temperatures;

[0056] Wherein, fusing the crop cultivation temperature record value in the first time zone and the crop cultivation temperature record value in the second time zone to obtain the crop cultivation temperature record value in the third time zone includes:

[0057] Calculate the time ratio of the first time zone to the third time zone to obtain a first fusion weight, and calculate the time ratio of the second time zone to the third time zone to obtain a second fusion weight;

[0058] According to the first fusion weight and the second fusion weight, a weighted mean analysis is performed on the crop cultivation temperature record value in the first time zone and the crop cultivation temperature record value in the second time zone to obtain the crop cultivation temperature record value in the third time zone.

[0059] In detail, the neighborhood temperature aggregation method of each crop cultivation temperature record time series data of the crop cultivation temperature record time series data set is the same, and the first crop cultivation temperature record time series data is used as an example:

[0060] According to the first crop cultivation temperature record time series data, the crop cultivation temperature record value of the first time zone and the crop cultivation temperature record value of the second time zone are extracted when the first time zone and the second time zone are adjacent time zones; the absolute value of the temperature deviation between the crop cultivation temperature record value of the first time zone and the crop cultivation temperature record value of the second time zone is calculated and stored as the crop cultivation temperature record value deviation; when the crop cultivation temperature record value deviation is less than or equal to the crop cultivation temperature record value deviation threshold value preset by the user, the first time zone and the second time zone are merged into a third time zone, that is, aggregated into a new time zone; when the crop cultivation temperature record value deviation is greater than the crop cultivation temperature record value deviation threshold value preset by the user, subsequent reference is not performed, and other adjacent time zones are continuously compared. The crop cultivation temperature record value of the first time zone and the crop cultivation temperature record value of the second time zone are merged to obtain the crop cultivation temperature record value of the third time zone; cyclic analysis, when the crop cultivation temperature record value deviation of any two adjacent time zones is greater than the crop cultivation temperature record value deviation threshold value, the first crop cultivation characteristic temperature of the crop planting time information is output and added to the multiple crop cultivation characteristic temperatures.

[0061] Furthermore, the detailed process of integrating the first time zone crop cultivation temperature record value and the second time zone crop cultivation temperature record value is as follows:

[0062] Calculate the time ratio between the first time zone and the third time zone to obtain a first fusion weight, calculate the time ratio between the second time zone and the third time zone to obtain a second fusion weight; perform a weighted mean analysis on the crop cultivation temperature record value in the first time zone and the crop cultivation temperature record value in the second time zone according to the first fusion weight and the second fusion weight to obtain the crop cultivation temperature record value in the third time zone.

[0063] The temperature acquisition unit is used to receive the time series information of the temperature monitored inside the greenhouse through the internal temperature sensor, and receive the time series information of the temperature monitored outside the greenhouse and the reference temperature range outside the greenhouse through the external temperature sensor;

[0064] Specifically, the internal temperature sensor is a sensor deployed inside the greenhouse to collect the temperature inside the greenhouse. Through it, the time series information of the greenhouse temperature monitoring within a specified time period can be uploaded in real time. Preferably, the current time of the specified time period is pushed forward 30 minutes. This is because the temperature value at a single moment may be accidental. Collecting temperature values ​​at multiple moments will improve the accuracy of the monitored temperature. The external temperature sensor is a sensor deployed outside the greenhouse to collect the temperature outside the greenhouse. Through it, the time series information of the greenhouse temperature monitoring outside within a specified time period can be uploaded in real time. Preferably, the current time of the specified time period is pushed forward 30 minutes. This is because the temperature value at a single moment may be accidental. Collecting temperature values ​​at multiple moments will improve the accuracy of the monitored temperature.

[0065] Furthermore, since the main factors affecting the temperature in the greenhouse are the outside temperature and the air supply control parameters, when the air supply control parameters remain unchanged, it is mainly affected by the outside temperature of the greenhouse. Therefore, each setting of the air supply control parameters is based on the monitored outside temperature of the greenhouse. If the outside temperature of the greenhouse changes significantly, the corresponding air supply control parameters will also become invalid. Therefore, the system triggering condition of the embodiment of the present application is that the outside temperature of the greenhouse has changed significantly. Therefore, each time the air supply control parameters are set, the corresponding outside temperature of the greenhouse will be stored, which is the outside reference temperature range, and is used as the trigger evaluation condition for temperature adjustment.

[0066] A control parameter retrieving unit, used to obtain a number of temperature control port air supply control parameters through a temperature control device;

[0067] A temperature curve prediction unit is used to process the time series information of the temperature monitoring inside the greenhouse, the time series information of the temperature monitoring outside the greenhouse and the air supply control parameters of the temperature regulating port through a greenhouse temperature prediction component to obtain a greenhouse temperature prediction curve in a preset time zone when the time series information of the temperature monitoring inside the greenhouse belongs to the expected ambient temperature range and the time series information of the temperature monitoring outside the greenhouse does not belong to the reference temperature range outside the greenhouse;

[0068] Specifically, the temperature regulating outlet air supply control parameters refer to the air supply temperature, air supply direction and air supply speed parameters of each air supply outlet. The proportion of the time series information of the monitored temperature in the greenhouse that does not belong to the time series information of the monitored temperature in the greenhouse is counted to obtain the abnormal probability of the monitored temperature in the greenhouse; when the abnormal probability of the monitored temperature in the greenhouse is less than or equal to the abnormal probability threshold preset by the user, the default value of the abnormal probability threshold is 30%, it is considered that the time series information of the monitored temperature in the greenhouse belongs to the expected range of the ambient temperature; when the abnormal probability of the monitored temperature in the greenhouse is greater than the abnormal probability threshold preset by the user, it is considered that the time series information of the monitored temperature in the greenhouse does not belong to the expected range of the ambient temperature. The method for judging whether the time series information of the monitored temperature outside the greenhouse belongs to the said reference temperature range outside the greenhouse is the same as the method for judging whether the time series information of the monitored temperature inside the greenhouse belongs to the expected range of the ambient temperature.

[0069] Case 1:

[0070] If the time series information of the temperature monitored in the greenhouse does not belong to the expected range of the ambient temperature, then the temperature in the greenhouse does not match the expected range at the current moment, and it is necessary to determine whether the time series information of the temperature monitored outside the greenhouse belongs to the reference temperature range outside the greenhouse:

[0071] If the time series information of the temperature monitored outside the greenhouse belongs to the reference temperature range outside the greenhouse, it means that the ambient temperature outside the greenhouse has not changed, the control parameters have not changed, but the temperature inside the greenhouse is abnormal, which is an uncontrollable abnormality, and an early warning signal is generated and sent to the user end.

[0072] Case 2:

[0073] If the time series information of the temperature monitored in the greenhouse does not belong to the expected range of the ambient temperature, then the temperature in the greenhouse does not match the expected range at the current moment, and it is necessary to determine whether the time series information of the temperature monitored outside the greenhouse belongs to the reference temperature range outside the greenhouse:

[0074] If the time series information of the temperature monitored outside the greenhouse does not belong to the reference temperature range outside the greenhouse, it means that the ambient temperature outside the greenhouse has changed, the control parameters have not changed, but the temperature inside the greenhouse is abnormal. This is a controllable abnormality, so the control parameters are temporarily adjusted, which is a traditional adjustment with hysteresis.

[0075] Case 3:

[0076] If the time series information of the temperature monitored in the greenhouse belongs to the expected range of the ambient temperature, the temperature in the greenhouse at the current moment is consistent, and it is necessary to determine whether the time series information of the temperature monitored outside the greenhouse belongs to the reference temperature range outside the greenhouse:

[0077] If the time series information of the temperature monitored outside the greenhouse belongs to the reference temperature range outside the greenhouse, it means that the ambient temperature outside the greenhouse has not changed, that is, the original control parameters can ensure constant temperature and stop the execution process.

[0078] Case 4:

[0079] If the time series information of the temperature monitored in the greenhouse belongs to the expected range of the ambient temperature, the temperature in the greenhouse at the current moment is consistent, and it is necessary to determine whether the time series information of the temperature monitored outside the greenhouse belongs to the reference temperature range outside the greenhouse:

[0080] If the time series information of the monitored temperature outside the greenhouse does not belong to the reference temperature range outside the greenhouse, it means that the ambient temperature outside the greenhouse has changed. At this time, it is necessary to predict the temperature curve inside the greenhouse in the future time zone based on the current control parameters, so as to perform preliminary temperature adjustment, which is the detailed steps of the embodiment of the present application.

[0081] When the greenhouse monitoring temperature time series information belongs to the expected ambient temperature range, and the outside greenhouse monitoring temperature time series information does not belong to the outside greenhouse reference temperature range, it is the above-mentioned situation four. When situation four is triggered, the greenhouse temperature prediction component is used to process the greenhouse monitoring temperature time series information, the outside greenhouse monitoring temperature time series information and the several temperature control port air supply control parameters to obtain a preset time zone greenhouse temperature prediction curve. The greenhouse temperature prediction component is a machine learning model built for the greenhouse in the embodiment of the present application, which is used to process the greenhouse monitoring temperature time series information, the outside greenhouse monitoring temperature time series information and several temperature control port air supply control parameters, and output a preset time zone greenhouse temperature prediction curve, that is, a temperature prediction curve pushed forward a specified time period from the current moment. Preferably, the preset time zone is pushed forward 3 hours from the current moment.

[0082] like Figure 2 As shown, in order to facilitate the technical personnel in the field to understand the scheme, the construction steps of the greenhouse temperature prediction component are disclosed in detail:

[0083] Obtaining a three-dimensional model of a greenhouse, wherein the three-dimensional model of the greenhouse has temperature control port distribution position labels;

[0084] Performing cube segmentation on the greenhouse three-dimensional model according to a preset grid side length to obtain grid distribution position labels;

[0085] Perform topological simulation according to the three-dimensional model of the greenhouse to construct a greenhouse graph neural network, wherein the temperature control port distribution position label is a first type input node in the greenhouse graph neural network, and the grid distribution position label is a second type input node in the greenhouse graph neural network;

[0086] Based on the first type of input nodes and the second type of input nodes, supervised training is performed on the greenhouse graph neural network to obtain the greenhouse temperature prediction component;

[0087] Among them, the input of the first type of input node is the air supply control parameter of the temperature regulating port, the input of the second type of input node is the time series information of the temperature monitored inside the greenhouse and the time series information of the temperature monitored outside the greenhouse, and the output of the greenhouse graph neural network is the temperature prediction curve inside the greenhouse in the specified time zone.

[0088] In detail, Graph Neural Networks (GNN) is a deep learning model specifically designed for processing graph structured data. Graph neural networks are motivated by the success of convolutional neural networks (CNN), but unlike CNN, GNN can be applied to graph structured data in non-Euclidean spaces. Graph neural networks can be used to predict state data in a certain space. The embodiment of the present application uses graph neural networks to predict the temperature distribution state of the space in the greenhouse. Since the basic changes are relatively small after a greenhouse is built, the graph neural network in the embodiment of the present application is based on the quantitative configuration of one model for one greenhouse to ensure the accuracy of the predicted data.

[0089] The greenhouse three-dimensional model is the outline frame model of the greenhouse. The greenhouse three-dimensional model has the temperature control port distribution position label of each temperature control port spatial position coordinate; the greenhouse three-dimensional model is cube-divided according to the preset grid side length to obtain the grid distribution position label. By default, the preset grid side length is 50cm. The longer the preset grid side length, the higher the processing efficiency of the graph neural network. However, due to the poor grid division accuracy, the analysis of local temperature differences cannot be realized; the shorter the preset grid side length, the lower the processing efficiency of the graph neural network. However, due to the poor grid division accuracy, the analysis of local temperature differences can be realized. When training the model, users can customize this parameter according to their needs.

[0090] Construct a greenhouse graph neural network with the same topological structure as the three-dimensional model of the greenhouse, set the temperature control port distribution position label in the greenhouse graph neural network as the first type of input node, and its input data is the control parameters of the temperature control port, namely wind speed, wind direction and wind temperature, and set the grid distribution position label in the greenhouse graph neural network as the second type of input node, and its input data is the time series information of the temperature monitored inside the greenhouse and the time series information of the temperature monitored outside the greenhouse. Since the model topological structure of the greenhouse graph neural network has been built, the variables affecting the temperature inside the greenhouse have also been set as output nodes, at this time, the labeled data is collected, and the greenhouse graph neural network is supervisedly trained to obtain the greenhouse temperature prediction component. The training method can adopt traditional model training methods, or the training methods disclosed in the embodiments of the present application.

[0091] The training algorithm process disclosed in the embodiment of this application is as follows:

[0092] Further, based on the first type of input nodes and the second type of input nodes, supervised training is performed on the greenhouse graph neural network to obtain the greenhouse temperature prediction component, including:

[0093] Collect greenhouse temperature historical monitoring data and several temperature control port historical air supply control parameters, wherein the greenhouse temperature historical monitoring data includes greenhouse temperature historical monitoring time series data and greenhouse temperature external monitoring time series data;

[0094] Performing neighborhood temperature aggregation on the historical monitoring time series data of the temperature outside the greenhouse to obtain a segmentation sequence of the historical monitoring data of the temperature outside the greenhouse;

[0095] Segment the greenhouse temperature history monitoring time series data according to the greenhouse temperature history monitoring data segmentation sequence to obtain the greenhouse temperature history monitoring time series data segmentation sequence;

[0096] According to the grid distribution position label, the shed temperature history monitoring time series data segmentation sequence is grid distributed to obtain the first grid shed temperature history monitoring time series data segmentation sequence to the Nth grid shed temperature history monitoring time series data segmentation sequence;

[0097] According to the historical air supply control parameters of the several temperature control ports, the segmentation sequence of the historical temperature monitoring data outside the greenhouse, the segmentation sequence of the first grid greenhouse temperature historical monitoring time series data until the segmentation sequence of the Nth grid greenhouse temperature historical monitoring time series data, based on the first type of input nodes and the second type of input nodes, the greenhouse graph neural network is supervised trained to obtain the greenhouse temperature prediction component.

[0098] Specifically, the greenhouse temperature historical monitoring data and several temperature control port historical air supply control parameters are collected. The greenhouse temperature historical monitoring data includes the greenhouse temperature historical monitoring time series data and the greenhouse temperature historical monitoring time series data at multiple locations inside and outside the greenhouse. The greenhouse temperature historical monitoring time series data is subjected to neighborhood temperature aggregation to obtain a segmentation sequence of the greenhouse temperature historical monitoring data. The neighborhood temperature aggregation of the greenhouse temperature historical monitoring time series data here is exactly the same as the aforementioned neighborhood temperature aggregation method. According to the greenhouse temperature historical monitoring data segmentation sequence, the greenhouse temperature historical monitoring time series data corresponding to each other are segmented in time series to obtain a greenhouse temperature historical monitoring time series data segmentation sequence. Furthermore, since each spatial grid also needs to input data, the greenhouse temperature historical monitoring time series data segmentation sequence is grid distributed according to the grid distribution position label to obtain the first grid greenhouse temperature historical monitoring time series data segmentation sequence until the Nth grid greenhouse temperature historical monitoring time series data segmentation sequence. According to the several temperature control port historical air supply control parameters, the greenhouse external temperature historical monitoring data segmentation sequence, the first grid greenhouse internal temperature historical monitoring time series data segmentation sequence until the Nth grid greenhouse internal temperature historical monitoring time series data segmentation sequence, based on the first type of input nodes and the second type of input nodes, supervised training is performed on the greenhouse graph neural network to obtain the greenhouse temperature prediction component. Through this step, the collected data can be automatically processed to facilitate adaptation to the subsequent training process, so as to more efficiently train the greenhouse temperature prediction component.

[0099] After the above-mentioned several temperature control port historical air supply control parameters, the partition sequence of the historical monitoring data of the temperature outside the shed, the first grid shed temperature historical monitoring time series data partition sequence until the Nth grid shed temperature historical monitoring time series data partition sequence are prepared, the detailed training process is as follows:

[0100] Further, according to the historical air supply control parameters of the several temperature control ports, the segmentation sequence of the historical monitoring data of the temperature outside the greenhouse, the segmentation sequence of the historical monitoring time series data of the temperature inside the first grid greenhouse until the segmentation sequence of the historical monitoring time series data of the temperature inside the Nth grid greenhouse, the greenhouse graph neural network is supervisedly trained based on the first type of input nodes and the second type of input nodes to obtain the greenhouse temperature prediction component, including:

[0101] According to the partition sequence of the historical monitoring data of the temperature outside the greenhouse, a first partitioned historical monitoring data of the temperature outside the greenhouse is obtained;

[0102] According to the first grid shed temperature history monitoring time series data segmentation sequence to the Nth grid shed temperature history monitoring time series data segmentation sequence, extract the first grid segmented shed temperature history monitoring time series data to the Nth grid segmented shed temperature history monitoring time series data, wherein the first grid segmented shed temperature history monitoring time series data to the Nth grid segmented shed temperature history monitoring time series data correspond to the first segmented shed external temperature history monitoring data;

[0103] According to the specified time zone length, the first grid-divided greenhouse temperature history monitoring time series data is traversed until the N-th grid-divided greenhouse temperature history monitoring time series data is tail-aligned and split into two, and the first grid-divided greenhouse temperature history monitoring input time series data and the first grid-divided greenhouse temperature history monitoring supervision time series data are obtained, until the N-th grid-divided greenhouse temperature history monitoring input time series data and the N-th grid-divided greenhouse temperature history monitoring supervision time series data;

[0104] Construct a training loss evaluation function:

[0105] ,

[0106] in, Represents the global loss of training, Represents the predicted temperature value of the i-th grid at the k-th moment, represents the supervised temperature value of the kth moment of the i-th grid, a represents a small constant, and Q represents the total number of moments of the supervised data;

[0107] According to the training loss evaluation function, the greenhouse graph neural network is supervised by taking the first grid-divided greenhouse temperature historical monitoring supervision time series data until the N-th grid-divided greenhouse temperature historical monitoring supervision time series data as supervision, taking the first grid-divided greenhouse temperature historical monitoring input time series data until the N-th grid-divided greenhouse temperature historical monitoring input time series data and the first divided greenhouse external temperature historical monitoring data as inputs of the second type of input nodes, taking the several temperature control port historical air supply control parameters as inputs of the first type of input nodes, and performing supervised training on the greenhouse graph neural network to obtain the greenhouse temperature prediction component.

[0108] Specifically, the first segmented external shed temperature history monitoring data refers to the representative temperature data of a certain time zone in the segmented sequence of external shed temperature history monitoring data; the first grid segmented shed temperature history monitoring time series data until the Nth grid segmented shed temperature history monitoring time series data refer to the temperature data corresponding to the time zone of each grid and the first segmented external shed temperature history monitoring data. Further, according to the specified time zone length, that is, the aforementioned temperature curve duration that needs to be predicted, the first grid segmented shed temperature history monitoring time series data until the Nth grid segmented shed temperature history monitoring time series data are traversed to perform tail alignment, and the first grid segmented shed temperature history monitoring time series data until the Nth grid segmented shed temperature history monitoring time series data are divided into two, thereby obtaining the first grid segmented shed temperature history monitoring input time series data and the first grid segmented shed temperature history monitoring supervision time series data, until the Nth grid segmented shed temperature history monitoring input time series data and the Nth grid segmented shed temperature history monitoring supervision time series data.

[0109] Furthermore, the same processing is performed on each time zone of the segmented sequence of historical monitoring data of the temperature outside the greenhouse to obtain a training data set for the greenhouse temperature prediction component.

[0110] Furthermore, we construct a training loss evaluation function: ,in, Represents the global loss of training, Represents the predicted temperature value of the i-th grid at the k-th moment, Characterizes the supervised temperature value at the kth moment of the ith grid, a characterizes a small constant, and Q characterizes the total number of moments of the supervised data. According to the training loss evaluation function, the first grid-divided greenhouse temperature historical monitoring supervision time series data until the Nth grid-divided greenhouse temperature historical monitoring supervision time series data is used as supervision, the first grid-divided greenhouse temperature historical monitoring input time series data until the Nth grid-divided greenhouse temperature historical monitoring input time series data and the first divided greenhouse external temperature historical monitoring data are used as inputs of the second type of input nodes, and the several temperature control port historical air supply control parameters are used as inputs of the first type of input nodes, and the greenhouse graph neural network is supervisedly trained to obtain the greenhouse temperature prediction component. Preferably, when the loss value of the loss evaluation function is trained for a preset number of consecutive times and is less than or equal to the loss threshold, the greenhouse temperature prediction component is deemed to have converged.

[0111] Through the loss evaluation function customized in the embodiment of the present application, the overall global loss of the loss evaluation of each grid output can be analyzed. Through the exponential form, the difference between different model parameters can be expanded, thereby ensuring a greenhouse temperature prediction component with higher accuracy.

[0112] An abnormal time zone extraction unit is used to extract the abnormal prediction time zone in which the temperature prediction curve in the greenhouse in the preset time zone does not belong to the expected range of the ambient temperature;

[0113] Specifically, the greenhouse temperature prediction component can be used to predict the greenhouse temperature prediction curve in the preset time zone, that is, the future greenhouse temperature development state under the current greenhouse temperature and control parameters. The greenhouse temperature prediction curve in the preset time zone does not belong to the expected ambient temperature range and is set as the abnormal prediction time zone.

[0114] A control parameter adjustment unit, for adjusting the air supply control parameters of the temperature regulating ports according to the time series information of the temperature monitored outside the greenhouse and the time series information of the temperature monitored inside the greenhouse through a greenhouse temperature configuration component and the greenhouse temperature prediction component when the time length of the abnormal prediction time zone is greater than or equal to the abnormal time length threshold, to obtain the recommended air supply control parameters of the temperature regulating ports;

[0115] Specifically, when the duration of the abnormal prediction time zone is greater than or equal to the abnormal duration threshold, it means that the abnormal time zone is too long, and the future time zone is unsafe. It is necessary to use the greenhouse temperature configuration component and the greenhouse temperature prediction component to adjust the several temperature control port air supply control parameters according to the external monitoring temperature timing information and the internal monitoring temperature timing information to obtain several recommended temperature control port air supply control parameters.

[0116] Preferably, the steps of adjusting the air supply control parameters of the plurality of temperature regulating ports are specifically as follows:

[0117] Further, when the duration of the abnormal prediction time zone is greater than or equal to the abnormal duration threshold, the greenhouse temperature configuration component and the greenhouse temperature prediction component are used to adjust the air supply control parameters of the plurality of temperature regulating ports according to the time series information of the temperature monitored outside the greenhouse and the time series information of the temperature monitored inside the greenhouse, and obtain the plurality of recommended air supply control parameters of the temperature regulating ports, including:

[0118] Adjust the wind speed and / or wind temperature and / or wind direction of the several temperature regulating port air supply control parameters to obtain several temperature regulating port air supply update control parameters;

[0119] Through the greenhouse temperature prediction component, the greenhouse temperature monitoring time series information, the greenhouse temperature monitoring time series information and the several temperature control outlet air supply update control parameters are processed to obtain a preset time zone update greenhouse temperature prediction curve;

[0120] When the duration of the abnormal prediction time zone of the preset time zone updating the indoor temperature prediction curve is less than the abnormal duration threshold, the several temperature regulating outlet air supply update control parameters are set as the several temperature regulating outlet recommended air supply control parameters.

[0121] In detail, the wind speed and / or wind temperature and / or wind direction of the several temperature regulating outlet air supply control parameters are randomly adjusted according to the step size preset by the user and the interval preset by the user to obtain several temperature regulating outlet air supply update control parameters; through the greenhouse temperature prediction component, the greenhouse monitoring temperature timing information, the greenhouse monitoring temperature timing information and the several temperature regulating outlet air supply update control parameters are processed to obtain the greenhouse temperature prediction curve updated in the preset time zone; when the duration of the abnormal prediction time zone of the greenhouse temperature prediction curve updated in the preset time zone is less than the abnormal duration threshold, the several temperature regulating outlet air supply update control parameters are set as the several temperature regulating outlet recommended air supply control parameters. When the duration of the abnormal prediction time zone of the greenhouse temperature prediction curve updated in the preset time zone is greater than or equal to the abnormal duration threshold, the control parameters need to be adjusted again.

[0122] The temperature control unit in the greenhouse is used to initialize the temperature control device to control the ambient temperature according to the recommended air supply control parameters of the several temperature control ports, and at the same time update the reference temperature range outside the greenhouse according to the timing information of the temperature monitored outside the greenhouse.

[0123] Further, updating the outdoor reference temperature interval according to the outdoor monitoring temperature time series information includes:

[0124] Performing a fluctuation concentration trend analysis on the time series information of the temperature monitored outside the greenhouse to obtain a concentrated temperature range of the temperature monitored outside the greenhouse;

[0125] The outside greenhouse reference temperature interval is updated according to the outside greenhouse monitored temperature concentrated temperature interval.

[0126] Specifically, due to the change in the temperature outside the greenhouse, the original air supply control parameters are no longer applicable to the temperature control inside the greenhouse for a certain period of time in the future. Therefore, after the recommended air supply control parameters for several temperature control ports are generated, the temperature control device is initialized according to the recommended air supply control parameters for the temperature control ports to control the ambient temperature. At the same time, due to the change in the temperature outside the greenhouse, the reference temperature interval outside the greenhouse needs to be updated. Therefore, the fluctuation concentration trend of the time series information of the outside greenhouse monitoring temperature is analyzed to obtain the concentrated temperature interval of the outside greenhouse monitoring temperature, and then the reference temperature interval outside the greenhouse is updated according to the concentrated temperature interval of the outside greenhouse monitoring temperature.

[0127] The detailed process of the fluctuation concentration trend analysis preferably uses conventional algorithms such as outlier analysis to determine the outlier temperature of the time series information of the outside monitoring temperature, and then stores the remaining temperatures to generate the outside monitoring temperature concentration temperature; then analyzes the distribution interval of the outside monitoring temperature concentration temperature and stores it as the outside monitoring temperature concentration temperature interval. By updating the outside reference temperature interval, it is used as the judgment basis for subsequent temperature pre-control, so as to facilitate the realization of continuous temperature pre-control.

[0128] The embodiment of the present invention provides a greenhouse temperature automatic control system based on temperature warning, which has at least the following technical effects:

[0129] By analyzing plant varieties and growth stages through big data, the expected range of environmental temperature preferred by crops is configured; then, the temperature sensors inside and outside the greenhouse are used to collect the temperature monitoring time series information inside and outside the greenhouse. If the greenhouse monitoring temperature time series information belongs to the expected range of environmental temperature, and the greenhouse monitoring temperature time series information does not belong to the greenhouse reference temperature range, the greenhouse temperature prediction component is used to process the greenhouse monitoring temperature time series information, the greenhouse monitoring temperature time series information and the several temperature control port air supply control parameters to obtain the greenhouse temperature prediction curve in the preset time zone; then, according to the greenhouse temperature prediction curve in the preset time zone; extract the The preset time zone greenhouse temperature prediction curve does not belong to the abnormal prediction time zone of the expected ambient temperature interval; when the duration of the abnormal prediction time zone is greater than or equal to the abnormal duration threshold, the greenhouse temperature configuration component and the greenhouse temperature prediction component are used to adjust the several temperature control port air supply control parameters according to the greenhouse external monitoring temperature timing information and the greenhouse internal monitoring temperature timing information, obtain several temperature control port recommended air supply control parameters, initialize the temperature control device according to the several temperature control port recommended air supply control parameters to control the ambient temperature, and update the greenhouse external reference temperature interval according to the greenhouse external monitoring temperature timing information. The greenhouse temperature prediction component is used to predict the change trend of the greenhouse temperature, and then adjust the temperature control port control parameters according to the change trend, reduce the hysteresis of temperature regulation, and achieve the technical effect of improving the predictability of greenhouse temperature control.

[0130] Embodiment 2:

[0131] See also Figure 3 , Figure 3 Schematic diagram of an electronic device provided by an embodiment of the present invention. Figure 3 As shown, an embodiment of the present invention provides an electronic device 500, including a memory 510, a processor 520, and a computer program 511 stored in the memory 510 and executable on the processor 520. When the processor 520 executes the computer program 511, the following steps are implemented:

[0132] A temperature frequency matching unit is used to receive information about planted crop varieties and crop planting duration information to perform temperature frequency matching and obtain an expected range of ambient temperature;

[0133] The temperature acquisition unit is used to receive the time series information of the temperature monitored inside the greenhouse through the internal temperature sensor, and receive the time series information of the temperature monitored outside the greenhouse and the reference temperature range outside the greenhouse through the external temperature sensor;

[0134] A control parameter retrieving unit, used to obtain a number of temperature control port air supply control parameters through a temperature control device;

[0135] A temperature curve prediction unit is used to process the time series information of the temperature monitoring inside the greenhouse, the time series information of the temperature monitoring outside the greenhouse and the air supply control parameters of the temperature regulating port through a greenhouse temperature prediction component to obtain a greenhouse temperature prediction curve in a preset time zone when the time series information of the temperature monitoring inside the greenhouse belongs to the expected ambient temperature range and the time series information of the temperature monitoring outside the greenhouse does not belong to the reference temperature range outside the greenhouse;

[0136] An abnormal time zone extraction unit is used to extract the abnormal prediction time zone in which the temperature prediction curve in the greenhouse in the preset time zone does not belong to the expected range of the ambient temperature;

[0137] A control parameter adjustment unit, for adjusting the air supply control parameters of the temperature regulating ports according to the time series information of the temperature monitored outside the greenhouse and the time series information of the temperature monitored inside the greenhouse through a greenhouse temperature configuration component and the greenhouse temperature prediction component when the time length of the abnormal prediction time zone is greater than or equal to the abnormal time length threshold, to obtain the recommended air supply control parameters of the temperature regulating ports;

[0138] The temperature control unit in the greenhouse is used to initialize the temperature control device to control the ambient temperature according to the recommended air supply control parameters of the several temperature control ports, and at the same time update the reference temperature range outside the greenhouse according to the timing information of the temperature monitored outside the greenhouse.

[0139] Embodiment three:

[0140] See also Figure 4 , Figure 4 A schematic diagram of an embodiment of a computer-readable storage medium provided in an embodiment of the present invention. Figure 4 As shown, this embodiment provides a computer-readable storage medium 600, on which a computer program 611 is stored. When the computer program 611 is executed by a processor, the following steps are implemented:

[0141] A temperature frequency matching unit is used to receive information about planted crop varieties and crop planting duration information to perform temperature frequency matching and obtain an expected range of ambient temperature;

[0142] The temperature acquisition unit is used to receive the time series information of the temperature monitored inside the greenhouse through the internal temperature sensor, and receive the time series information of the temperature monitored outside the greenhouse and the reference temperature range outside the greenhouse through the external temperature sensor;

[0143] A control parameter retrieving unit, used to obtain a number of temperature control port air supply control parameters through a temperature control device;

[0144] A temperature curve prediction unit is used to process the time series information of the temperature monitoring inside the greenhouse, the time series information of the temperature monitoring outside the greenhouse and the air supply control parameters of the temperature regulating port through a greenhouse temperature prediction component to obtain a greenhouse temperature prediction curve in a preset time zone when the time series information of the temperature monitoring inside the greenhouse belongs to the expected ambient temperature range and the time series information of the temperature monitoring outside the greenhouse does not belong to the reference temperature range outside the greenhouse;

[0145] An abnormal time zone extraction unit is used to extract the abnormal prediction time zone in which the temperature prediction curve in the greenhouse in the preset time zone does not belong to the expected range of the ambient temperature;

[0146] A control parameter adjustment unit, for adjusting the air supply control parameters of the temperature regulating ports according to the time series information of the temperature monitored outside the greenhouse and the time series information of the temperature monitored inside the greenhouse through a greenhouse temperature configuration component and the greenhouse temperature prediction component when the time length of the abnormal prediction time zone is greater than or equal to the abnormal time length threshold, to obtain the recommended air supply control parameters of the temperature regulating ports;

[0147] The temperature control unit in the greenhouse is used to initialize the temperature control device to control the ambient temperature according to the recommended air supply control parameters of the several temperature control ports, and at the same time update the reference temperature range outside the greenhouse according to the timing information of the temperature monitored outside the greenhouse.

[0148] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0149] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as systems, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0150] The present invention is described with reference to flowcharts and / or block diagrams of systems, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1A device that provides the functions specified in a block or multiple blocks.

[0151] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0152] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0153] Although preferred embodiments of the present invention have been described, additional changes and modifications may occur to these embodiments once those skilled in the art understand the basic inventive concepts.

[0154] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention belong to the scope of the present invention and its equivalent technologies, the present invention is also intended to include these changes and variations.

Claims

1. A greenhouse temperature automatic control system based on temperature early warning, characterized in that: Deployed in the automatic controller of the temperature control device, the automatic controller of the temperature control device includes a greenhouse temperature prediction component and a greenhouse temperature configuration component, the automatic controller of the temperature control device is communicatively connected with the temperature control device, and the temperature control device has several temperature control ports in the greenhouse, including: A temperature frequency matching unit is used to receive information about planted crop varieties and crop planting duration information to perform temperature frequency matching and obtain an expected range of ambient temperature; The temperature acquisition unit is used to receive the time series information of the temperature monitored inside the greenhouse through the internal temperature sensor, and receive the time series information of the temperature monitored outside the greenhouse and the reference temperature range outside the greenhouse through the external temperature sensor; A control parameter retrieving unit, used to obtain a number of temperature control port air supply control parameters through a temperature control device; A temperature curve prediction unit is used to process the time series information of the temperature monitoring inside the greenhouse, the time series information of the temperature monitoring outside the greenhouse and the air supply control parameters of the temperature regulating port through a greenhouse temperature prediction component to obtain a greenhouse temperature prediction curve in a preset time zone when the time series information of the temperature monitoring inside the greenhouse belongs to the expected ambient temperature range and the time series information of the temperature monitoring outside the greenhouse does not belong to the reference temperature range outside the greenhouse; An abnormal time zone extraction unit is used to extract the abnormal prediction time zone in which the temperature prediction curve in the greenhouse in the preset time zone does not belong to the expected range of the ambient temperature; A control parameter adjustment unit, for adjusting the air supply control parameters of the temperature regulating ports according to the time series information of the temperature monitored outside the greenhouse and the time series information of the temperature monitored inside the greenhouse through a greenhouse temperature configuration component and the greenhouse temperature prediction component when the time length of the abnormal prediction time zone is greater than or equal to the abnormal time length threshold, to obtain the recommended air supply control parameters of the temperature regulating ports; The greenhouse temperature control unit is used to initialize the temperature control device to control the ambient temperature according to the recommended air supply control parameters of the plurality of temperature control ports, and update the outdoor reference temperature range according to the outdoor monitoring temperature time series information; The greenhouse temperature prediction component construction process is as follows: Obtaining a three-dimensional model of a greenhouse, wherein the three-dimensional model of the greenhouse has temperature control port distribution position labels; Performing cube segmentation on the greenhouse three-dimensional model according to a preset grid side length to obtain grid distribution position labels; Perform topological simulation according to the three-dimensional model of the greenhouse to construct a greenhouse graph neural network, wherein the temperature control port distribution position label is a first type input node in the greenhouse graph neural network, and the grid distribution position label is a second type input node in the greenhouse graph neural network; Based on the first type of input nodes and the second type of input nodes, supervised training is performed on the greenhouse graph neural network to obtain the greenhouse temperature prediction component; The input of the first type of input node is the air supply control parameter of the temperature regulating port, the input of the second type of input node is the time series information of the temperature monitored inside the greenhouse and the time series information of the temperature monitored outside the greenhouse, and the output of the greenhouse graph neural network is the temperature prediction curve inside the greenhouse in the specified time zone; The greenhouse graph neural network is supervisedly trained based on the first type input node and the second type input node to obtain the greenhouse temperature prediction component, including: Collect greenhouse temperature historical monitoring data and several temperature control port historical air supply control parameters, wherein the greenhouse temperature historical monitoring data includes greenhouse temperature historical monitoring time series data and greenhouse temperature external monitoring time series data; Performing neighborhood temperature aggregation on the historical monitoring time series data of the temperature outside the greenhouse to obtain a segmentation sequence of the historical monitoring data of the temperature outside the greenhouse; Segment the greenhouse temperature history monitoring time series data according to the greenhouse temperature history monitoring data segmentation sequence to obtain the greenhouse temperature history monitoring time series data segmentation sequence; According to the grid distribution position label, the shed temperature history monitoring time series data segmentation sequence is grid distributed to obtain the first grid shed temperature history monitoring time series data segmentation sequence to the Nth grid shed temperature history monitoring time series data segmentation sequence; According to the several temperature control port historical air supply control parameters, the greenhouse external temperature historical monitoring data segmentation sequence, the first grid greenhouse internal temperature historical monitoring time series data segmentation sequence until the Nth grid greenhouse internal temperature historical monitoring time series data segmentation sequence, based on the first type of input nodes and the second type of input nodes, supervised training is performed on the greenhouse graph neural network to obtain the greenhouse temperature prediction component; Among them, according to the historical air supply control parameters of the several temperature control ports, the segmentation sequence of the historical monitoring data of the temperature outside the greenhouse, the segmentation sequence of the historical monitoring time series data of the temperature inside the first grid greenhouse until the segmentation sequence of the historical monitoring time series data of the temperature inside the Nth grid greenhouse, the greenhouse graph neural network is supervisedly trained based on the first type of input nodes and the second type of input nodes to obtain the greenhouse temperature prediction component, including: According to the partition sequence of the historical monitoring data of the temperature outside the greenhouse, a first partitioned historical monitoring data of the temperature outside the greenhouse is obtained; According to the first grid shed temperature history monitoring time series data segmentation sequence to the Nth grid shed temperature history monitoring time series data segmentation sequence, extract the first grid segmented shed temperature history monitoring time series data to the Nth grid segmented shed temperature history monitoring time series data, wherein the first grid segmented shed temperature history monitoring time series data to the Nth grid segmented shed temperature history monitoring time series data correspond to the first segmented shed external temperature history monitoring data; According to the specified time zone length, the first grid-divided greenhouse temperature history monitoring time series data is traversed until the N-th grid-divided greenhouse temperature history monitoring time series data is tail-aligned and split into two, and the first grid-divided greenhouse temperature history monitoring input time series data and the first grid-divided greenhouse temperature history monitoring supervision time series data are obtained, until the N-th grid-divided greenhouse temperature history monitoring input time series data and the N-th grid-divided greenhouse temperature history monitoring supervision time series data; Construct a training loss evaluation function: , in, Represents the global loss of training, Represents the predicted temperature value of the i-th grid at the k-th moment, represents the supervised temperature value of the kth moment of the i-th grid, a represents a small constant, and Q represents the total number of moments of the supervised data; According to the training loss evaluation function, the greenhouse graph neural network is supervised by taking the first grid-divided greenhouse temperature historical monitoring supervision time series data until the N-th grid-divided greenhouse temperature historical monitoring supervision time series data as supervision, taking the first grid-divided greenhouse temperature historical monitoring input time series data until the N-th grid-divided greenhouse temperature historical monitoring input time series data and the first divided greenhouse external temperature historical monitoring data as inputs of the second type of input nodes, taking the several temperature control port historical air supply control parameters as inputs of the first type of input nodes, and performing supervised training on the greenhouse graph neural network to obtain the greenhouse temperature prediction component.

2. The system according to claim 1, characterized in that Receive crop variety information and crop planting time information to perform temperature frequency matching and obtain the expected range of ambient temperature, including: Input the planting crop variety information into a crop growth cycle calibration table, and output a crop growth stage sequence, wherein the crop growth stage sequence has a crop planting time label sequence; According to the crop planting time label sequence and the crop growth stage sequence, the crop planting time information is calibrated to obtain the crop growth calibration stage; Positive samples are collected according to the planted crop variety information and the crop growth calibration stage to obtain a crop cultivation temperature record time series data set; Traversing the crop cultivation temperature record time series data set to perform neighborhood temperature aggregation to obtain a plurality of crop cultivation characteristic temperatures of the crop planting time information; A central trend analysis is performed on the multiple crop cultivation characteristic temperatures to construct the expected range of the environmental temperature.

3. The system according to claim 2, characterized in that Traversing the crop cultivation temperature record time series data set to perform neighborhood temperature aggregation, and obtaining multiple crop cultivation characteristic temperatures of the crop planting time information, including: Extracting first crop cultivation temperature record time series data according to the crop cultivation temperature record time series data set; Extracting a first time zone crop cultivation temperature record value and a second time zone crop cultivation temperature record value according to the first crop cultivation temperature record time series data, wherein the first time zone and the second time zone are adjacent time zones; When the crop cultivation temperature record value deviation between the crop cultivation temperature record value in the first time zone and the crop cultivation temperature record value in the second time zone is less than or equal to the crop cultivation temperature record value deviation threshold, merging the first time zone and the second time zone into a third time zone; The crop cultivation temperature record value in the first time zone and the crop cultivation temperature record value in the second time zone are integrated to obtain the crop cultivation temperature record value in the third time zone; Circular analysis, when the deviation of the crop cultivation temperature record values ​​in any two adjacent time zones is greater than the crop cultivation temperature record value deviation threshold, output the first crop cultivation characteristic temperature of the crop planting time information, and add it to the multiple crop cultivation characteristic temperatures; Wherein, fusing the crop cultivation temperature record value in the first time zone and the crop cultivation temperature record value in the second time zone to obtain the crop cultivation temperature record value in the third time zone includes: Calculate the time ratio of the first time zone to the third time zone to obtain a first fusion weight, and calculate the time ratio of the second time zone to the third time zone to obtain a second fusion weight; According to the first fusion weight and the second fusion weight, a weighted mean analysis is performed on the crop cultivation temperature record value in the first time zone and the crop cultivation temperature record value in the second time zone to obtain the crop cultivation temperature record value in the third time zone.

4. The system according to claim 1, characterized in that When the duration of the abnormal prediction time zone is greater than or equal to the abnormal duration threshold, the plurality of temperature regulating port air supply control parameters are adjusted according to the external monitoring temperature timing information and the internal monitoring temperature timing information through the greenhouse temperature configuration component and the greenhouse temperature prediction component to obtain a plurality of temperature regulating port recommended air supply control parameters, including: Adjust the wind speed and / or wind temperature and / or wind direction of the several temperature regulating port air supply control parameters to obtain several temperature regulating port air supply update control parameters; Through the greenhouse temperature prediction component, the greenhouse temperature monitoring time series information, the greenhouse temperature monitoring time series information and the several temperature control outlet air supply update control parameters are processed to obtain a preset time zone update greenhouse temperature prediction curve; When the duration of the abnormal prediction time zone of the preset time zone updating the indoor temperature prediction curve is less than the abnormal duration threshold, the several temperature regulating outlet air supply update control parameters are set as the several temperature regulating outlet recommended air supply control parameters.

5. The system according to claim 1, wherein: Updating the outdoor reference temperature interval according to the outdoor monitoring temperature time series information includes: Performing a fluctuation concentration trend analysis on the time series information of the temperature monitored outside the greenhouse to obtain a concentrated temperature range of the temperature monitored outside the greenhouse; The outside greenhouse reference temperature interval is updated according to the outside greenhouse monitored temperature concentrated temperature interval.

6. An electronic device, characterized in that: include: Memory for storing computer software programs; A processor is used to read and execute the computer software program, thereby realizing a greenhouse temperature automatic control system based on temperature warning as described in any one of claims 1 to 5.

7. A non-transitory computer-readable storage medium, characterized in that: The storage medium stores a computer software program, and when the computer software program is executed by the processor, it implements a greenhouse temperature automatic control system based on temperature warning as described in any one of claims 1 to 5.