CN112527037A - Greenhouse environment regulation and control method and system with environment factor prediction function - Google Patents
Greenhouse environment regulation and control method and system with environment factor prediction function Download PDFInfo
- Publication number
- CN112527037A CN112527037A CN202011573255.9A CN202011573255A CN112527037A CN 112527037 A CN112527037 A CN 112527037A CN 202011573255 A CN202011573255 A CN 202011573255A CN 112527037 A CN112527037 A CN 112527037A
- Authority
- CN
- China
- Prior art keywords
- greenhouse
- data
- wolf
- strategy
- environment
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D27/00—Simultaneous control of variables covered by two or more of main groups G05D1/00 - G05D25/00
- G05D27/02—Simultaneous control of variables covered by two or more of main groups G05D1/00 - G05D25/00 characterised by the use of electric means
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Air Conditioning Control Device (AREA)
Abstract
The application discloses a greenhouse environment regulation and control method and system with an environment factor prediction function, and belongs to the technical field of intelligent agriculture. The regulation and control method comprises the following steps: s1, collecting environmental data in the greenhouse and meteorological data outside the greenhouse in a preset time period; s2, processing the environmental data and the meteorological data to obtain a sample set, and dividing the sample set into a training set and a testing set; s3, determining initial data of the improved wolf optimization algorithm; s4, converting parameters needing to be optimized in the long-term and short-term memory network method into position coordinates of the wolf cluster, and optimizing the position coordinates by using an improved wolf optimization algorithm; s5, obtaining an optimal long-short term memory network method model; s6, collecting real-time data in real time; s7, pre-measuring the air temperature in the temperature chamber and the future value of the photosynthetically active radiation; and S8, regulating and controlling the temperature and the light of the greenhouse. The greenhouse environment regulation and control method can be used for measuring the indoor air temperature and the future value of photosynthetically active radiation in advance and regulating and controlling the greenhouse environment.
Description
Technical Field
The application belongs to the technical field of intelligent agriculture, and particularly relates to a greenhouse environment regulation and control method and system with an environment factor prediction function.
Background
In recent years, facility agriculture is rapidly developed, and an intelligent greenhouse is taken as a high-level representative of the facility agriculture, and can regulate and control various environmental factors such as temperature, light, water, gas and the like by utilizing a comprehensive environment control system, so that annual high yield, standardization and precision of the facility agriculture production are realized. Traditional intelligent greenhouses are mostly regulated and controlled based on the actual environmental conditions of the current greenhouses, and irreversible damage to crops is already caused in many times. How to keep the greenhouse in an optimal environment suitable for crop growth in real time becomes a problem to be solved urgently at present. The temperature and the illumination are used as main growth environmental factors influencing crops in the intelligent greenhouse, and how to accurately measure the indoor air temperature, the photosynthetically active radiation and the like is a key link for supporting intelligent regulation and control of the greenhouse environment.
Content of application
The embodiment of the application aims to provide a greenhouse environment regulation and control method with an environment factor prediction function, a greenhouse environment regulation and control system and electronic equipment, which can solve one of the problems of the background art.
In order to solve the technical problem, the present application is implemented as follows:
in a first aspect, an embodiment of the present application provides a greenhouse environment regulation method with an environmental factor prediction function, including the following steps: s1, collecting environmental data in the greenhouse and meteorological data outside the greenhouse in a preset time period; s2, processing the environmental data and the meteorological data to obtain a sample set, and dividing the sample set into a training set and a testing set; s3, determining initial data of the improved wolf optimization algorithm; s4, converting parameters needing to be optimized in the long and short term memory network method into position coordinates of the wolf cluster, and optimizing the position coordinates by using an improved wolf optimization algorithm to obtain optimized parameters of the long and short term memory network method; s5, selecting a test set, and testing the long-short term memory network method of the optimized parameters to obtain an optimal long-short term memory network method model; s6, collecting environmental data in the greenhouse and meteorological data outside the greenhouse in real time to obtain real-time data; s7, inputting the real-time data into an optimal long-term and short-term memory network method model, and predicting the air temperature in a temperature measuring room and the future value of photosynthetically active radiation; and S8, regulating and controlling the temperature and the light of the greenhouse according to the predicted result.
Optionally, in step S2, performing error data elimination, data supplementation, and data normalization processing on the environmental data and the meteorological data to obtain the sample set.
Optionally, in step S3, the initial data includes the number of gray wolves, the number of iterations, and the dimension.
Optionally, in step S4, the parameters to be optimized include: a forgetting gate weight parameter, a forgetting gate threshold parameter, an input gate weight parameter, an input gate threshold parameter, a candidate gate weight parameter, a candidate gate threshold parameter, an output gate weight parameter, an output gate threshold parameter, and an output layer weight parameter.
Optionally, in step S8, when the current air temperature in the greenhouse is higher than the optimum high value, the equipment control execution module implements a cooling strategy; when the current air temperature in the greenhouse is lower than the optimum low value, the equipment control execution module implements a temperature rise strategy; when t 'is satisfied under the condition that the current air temperature in the greenhouse is within a suitable value range'max-t＜tthred_airtempWhen the equipment is used, the equipment control execution module implements a cooling strategy; when t 'is satisfied under the condition that the current air temperature in the greenhouse is within a suitable value range'min-t＜tthred_airtempWhen the temperature rises, the equipment control execution module implements a temperature rise strategy; when the current light effective radiation in the greenhouse is higher than the optimum high value, the equipment controls the execution module to implement a shading strategy; when the current light effective radiation in the greenhouse is lower than the optimum low value, the equipment controls the execution module to implement a light supplement strategy; strip for greenhouse with currently available light radiation in a suitable range of valuesIf t 'is satisfied'max-t＜tthred_PIRWhen the device is in use, the device control execution module implements shading strategy; when t 'is satisfied under the condition that the current light effective radiation in the greenhouse is in a suitable value range'min-t＜tthred_PIRWhen the device is in use, the device control execution module implements a light supplement strategy; wherein t is the current time point, t'maxPredicted value for the first time higher than optimum high value corresponds to time point, t'minFor the first time below the predicted value of the optimum low value, tthred_airtempCritical time value, t, for a greenhouse air temperature start strategythred_PIRCritical time values for a greenhouse photosynthetically active radiation startup strategy.
In a second aspect, an embodiment of the present application provides a greenhouse environment regulation system, including: the data monitoring module is used for acquiring data to be processed; the cloud platform service module comprises a data storage and processing module, a greenhouse environment intelligent decision-making module and a data storage and processing module, wherein the data storage and processing module is used for solving a greenhouse environment regulation and control strategy according to the data to be processed and calculating an environment factor prediction model based on a long-short term memory network method of an improved grayling optimization algorithm; the greenhouse environment intelligent decision-making module makes an environmental equipment control strategy according to the environmental factor prediction model and the greenhouse environment regulation and control strategy, and the cloud platform service module can receive information sent by the data monitoring module; and the equipment control execution module executes automatic control operation on environmental data in the greenhouse according to the environmental equipment control strategy, and the cloud platform service module can send information to the equipment control execution module.
Optionally, the data to be processed includes meteorological data outside the greenhouse, environmental data inside the greenhouse, and crop body data, and the data monitoring module includes: and the wireless sensor acquires meteorological data outside the greenhouse, environmental data inside the greenhouse and the crop body data.
Optionally, the greenhouse outside weather data comprises outdoor air temperature, outdoor air humidity, outdoor wind speed, and rainfall data; the greenhouse internal environment data comprise indoor air temperature and humidity, illumination intensity, photosynthetic effective radiation and carbon dioxide concentration data; the crop ontology data includes leaf area, stem thickness and fruit diameter data of the crop.
Optionally, the data storage and processing module includes: an allocation unit that allocates the sorted time series sample sets into a training set and a test set; an initialization unit capable of determining initial data that improves a graying optimization algorithm; the position acquisition unit is used for converting the parameters needing to be optimized in the long-short term memory network method into position coordinates of the wolf pack; the optimization parameter unit selects the training set, calculates the individual fitness value of the wolf group by using an improved wolf optimization algorithm to reach the maximum iteration times, and obtains the optimal weight and the threshold parameter value of the long-short term memory network algorithm; the testing unit tests the long-short term memory network method optimized by the optimization parameter unit to obtain an optimal long-short term memory network model; and the prediction unit inputs the data to be processed acquired in real time through the data monitoring module into the optimal long-short term memory network model so as to predict the data to be processed in a preset time period.
Optionally, the optimization parameter unit defines an individual corresponding to the optimal fitness value as an α wolf, a suboptimal definition as a β wolf, and a third best definition as a δ wolf, and the optimization parameter unit includes: the position updating module is used for updating the positions of the rest wolfs according to alpha, beta and delta, and the adopted formula is as follows:
X(t+1)＝X1·Wα+X2·Wβ+X3·Wδ；
wherein D isα，DβAnd DδRespectively representing the distances between alpha, beta and delta and other wolfs, t is the current iteration number, Xα(t)、Xβ(t) and Xδ(t) represents the positions of α, β and δ at the current iteration number, respectively, X (t) is the current gray wolf position, Wα，WβAnd WδRepresenting the weight of alpha, beta and delta, respectively, fα、fβAnd fδRepresenting fitness values for alpha, beta, and delta, respectively, and X (t +1) representing the location of the gray wolf for the next iteration.
In a third aspect, an embodiment of the present application provides an electronic device, which includes a processor, a memory, and a program or instructions stored on the memory and executable on the processor, and when executed by the processor, the program or instructions implement the steps of the method according to the first aspect.
The method and the device can accurately predict the air temperature and the photosynthetically active radiation in the greenhouse based on the artificial intelligence algorithm, and provide decision basis for making regulation and control of the equipment control execution module in advance and keeping the greenhouse in the most suitable growth environment. According to the greenhouse environment regulation and control method and system with the environment factor prediction function, the environment factor information in the greenhouse can be comprehensively acquired based on the Internet of things, and the air temperature and the photosynthetic effective radiation of the greenhouse in a future period of time can be predicted by selecting an improved long-term and short-term memory network algorithm. And adjusting and controlling the equipment control execution module in real time according to the prediction result so as to keep the greenhouse in an optimal environment suitable for crop growth.
Drawings
FIG. 1 is a flow chart of a method for regulating a greenhouse environment with an environmental factor prediction function according to an embodiment of the present application;
FIG. 2 is a schematic diagram of a computing environment factor prediction model based on a long-short term memory network method of an improved grayish optimization algorithm for a greenhouse environment regulation and control method with an environment factor prediction function according to an embodiment of the present application;
FIG. 3 is a flow chart of a greenhouse environment regulation method with environmental factor prediction function according to an embodiment of the present application;
FIG. 4 is a block diagram of a greenhouse environment conditioning system according to an embodiment of the application.
Reference numerals
Environmental data inside the greenhouse and meteorological data outside the greenhouse 10; a data monitoring module 20; a cloud platform service module 30; a data storage and processing module 31; a greenhouse environment intelligent decision module 32; the device control execution module 40.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some, but not all, embodiments of the present application. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The terms first, second and the like in the description and in the claims of the present application are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the application are capable of operation in sequences other than those illustrated or described herein. In addition, "and/or" in the specification and claims means at least one of connected objects, a character "/" generally means that a preceding and succeeding related objects are in an "or" relationship.
The greenhouse environment regulation and control method with the environmental factor prediction function provided by the embodiment of the application is described in detail through specific embodiments and application scenarios thereof with reference to the attached drawings.
The greenhouse environment regulation and control method with the environmental factor prediction function according to the embodiment of the application comprises the following steps:
and S1, collecting environmental data in the greenhouse and meteorological data outside the greenhouse within a preset time period 10.
The meteorological data outside the greenhouse may include outdoor air temperature, outdoor air humidity, outdoor wind speed, and rainfall, among others. Environmental data within a greenhouse may include indoor air temperature and humidity, illumination intensity, photosynthetically active radiation, and carbon dioxide concentration. Specifically, the environmental data inside the greenhouse and the meteorological data outside the greenhouse 10 may be collected using parts or components such as wireless sensors. In addition, crop body data can also be collected through a wireless sensor.
And S2, processing the environmental data and the meteorological data to obtain a sample set, and dividing the sample set into a training set and a testing set.
Optionally, the step of processing the environmental data and the meteorological data in step S2 includes, but is not limited to, error data elimination, data supplementation, and data normalization. In the classification, the processed data can be classified into full-combination data according to the season category, the weather condition category and the time period category, and various types of classified combination data sets can be generated through integration.
S3, determining initial data of the improved Grey wolf optimization algorithm (GWO).
Optionally, the initial data of the gray wolf optimization algorithm in step S3 includes the number of gray wolfs, the number of iterations, and the dimensions.
S4, converting parameters needing to be optimized by a long-short term memory network (LSTM) method into position coordinates of the wolf cluster, and optimizing the position coordinates by using an improved wolf optimization algorithm to obtain optimized parameters of the long-short term memory network method, such as optimal weight and threshold parameter values.
Optionally, the parameters to be optimized for the long-short term memory network method include a forgetting gate weight parameter, a forgetting gate threshold parameter, an input gate parameter weight parameter, an input gate threshold parameter, a candidate gate weight parameter, a candidate gate threshold parameter, an output gate weight parameter, an output gate threshold parameter, and an output layer weight parameter.
Need to explainThe long-short term memory network (LSTM) is more suitable for the time series field by increasing the function of the memory line. It has four gates in the structure design, which are respectively a forgetting gate, an input gate, a candidate gate and an output gate. The forgetting door mainly controls the forgetting of old information and has the calculation formula of
The input gate is mainly used for controlling new information input, and the calculation formula is
The candidate gate calculates the total amount of the current information and the old information according to the formula
Updating the memory cell to ct＝ft×ct-1+it×c′t(ii) a The output gate has the formula of
The calculation formula of the whole hidden layer is ht＝ot×tanh(ct) (ii) a The calculation formula of the output layer is
Wherein the variables in the above formula have the following meanings:
ftrepresenting a forgetting gate; h ist-1Representing the output value of the hidden layer at the last t-1 moment;
representing a weight matrix of an output value of a hidden layer at a t-1 moment on the forgetting gate;
a weight matrix representing an input value at the moment t of forgetting to gate; i.e. itRepresenting the input value at the current time t; bfA bias term representing a forgetting gate; i'tRepresenting an input gate;
a weight matrix representing the output value of the hidden layer at a time t-1 on the input gate;
a weight matrix representing the input values at time t of the input gate; biAn offset term representing an input gate; c'tRepresenting a candidate gate;
a weight matrix representing the output value of the hidden layer at the last t-1 moment of the candidate gate;
a weight matrix representing input values at the time t of the candidate gate; bcA bias term representing a candidate gate; c. CtRepresenting the memory information at the time t; c. Ct-1Representing the memory information at the t-1 moment; otRepresents an output gate;
a weight matrix representing the output value of the hidden layer at a t-1 moment on the output gate;
a weight matrix representing input values at time t of the output gate; boA bias term representing an output gate; h istRepresenting the output value of the hidden layer at the time t; y istRepresenting the output value of the output layer at the time t;
representing the output layer weight matrix.
In step S4, the improved grayish optimization algorithm is applied to 13 parameters (in the long-short term memory network) ((S))
bf、
bi、
bc、
boAnd
) Optimizing, comprising the following two steps:
s41, surrounding the prey. The mathematical model of the behavior of the grayish wolf trap is as follows:
D＝|C·Xp(t)-X(t)|，X(t+1)＝Xp(t)-A·D。
wherein t is the current iteration number; xp(t) is the current prey position; x (t) is the current gray wolf position; x (t +1) is the position of the gray wolf after the next iteration; d is the distance between the gray wolf and the prey; a and C are coefficient vectors, and the updating formula is that A is 2a r1-a，C＝2·r2Where a is the convergence factor, the original convergence factor a decreases linearly from 2 to 0 with the number of iterations.
S42, hunting, wherein when the gray wolf identifies the position of a hunting object, the alpha wolf guides the beta and delta pairs to carry out hunting attack. The first three optimal solutions are selected, and the rest wolfs update own positions according to alpha, beta and delta:
wherein D isα，DβAnd DδRespectively representing the distances between alpha, beta and delta and other wolfs, t is the current iteration number, Xα(t)、Xβ(t) and Xδ(t) represents the positions of α, β, and δ, respectively, at the current iteration number.
Wherein the variables in the above formula have the following meanings:
C1a coefficient vector representing the movement of a gray wolf in the wolf group to alpha; c2A coefficient vector representing the movement of a gray wolf in the wolf group to alpha; c3A coefficient vector representing the movement of a gray wolf in the wolf group to alpha;
X1a vector representing the movement of a gray wolf in the wolf cluster to alpha; x2A vector representing the movement of a gray wolf in the wolf cluster to beta; x3A vector representing the movement of a gray wolf in the wolf cluster to delta;
A1a coefficient vector representing the movement of a gray wolf in the wolf group to alpha; a. the2A coefficient vector representing the movement of a gray wolf in the wolf group to alpha; a. the3Representing the coefficient vector of a gray wolf in the wolf group moving towards alpha.
Original GWO, using X as described above1，X2And X3The average value of the three is used as the position of the next iteration of the gray wolf, and in practice, three gray wolfs of alpha, beta and delta have different grade characteristics in the practical hunting, the invention provides a method for determining the iteration position of the gray wolf based on the weighted fitness value and static weighted mixture, and the formula is as follows:
X(t+1)＝X1·Wα+X2·Wβ+X3·Wδ。
wherein Wα，WβAnd WδRepresenting the weight of alpha, beta and delta, respectively, fα、fβAnd fδRepresenting fitness values for alpha, beta, and delta, respectively, and X (t +1) representing the location of the gray wolf for the next iteration. The method embodies the highest position of the alpha wolf in the gray wolf group, the beta wolf occupies the second position and the lowest position of the delta wolf in the first three wolfs (optimal solution).
And S5, selecting a test set, and testing the long-short term memory network method with optimized parameters to obtain an optimal long-short term memory network method model.
And S6, acquiring the environmental data in the greenhouse and the meteorological data 10 outside the greenhouse in real time to obtain real-time data.
And S7, inputting the real-time data into the optimal long-term and short-term memory network method model, and predicting the air temperature in the temperature measuring room and the future value of the photosynthetically active radiation. Specifically, the greenhouse air temperature value at the future time and the photosynthetically active radiation value in the greenhouse can be predicted for the greenhouse state under different scenes using the classification dataset in step S2.
And S8, regulating and controlling the temperature and light of the greenhouse according to the predicted result, wherein the regulating and controlling mode can comprise heating, cooling, light supplementing and shading. For example, the current environmental regulation threshold corresponding to the current greenhouse may be given according to the current weather condition, the current crop growth period, the current season and the current time period, and the crop state in the greenhouse, the current regulation strategy may be determined by using the environmental factor value predicted in step S7, and the device control execution module 40 may execute the operation according to the regulation strategy. That is, the use of the regulation strategy may start the regulation strategy according to the actual situation based on the predicted value of the future time obtained in step S7. Wherein the future time may be set to 6 h.
Optionally, in step S7, the prediction value refers to the threshold range of the temperature in the designated greenhouse and the threshold range of the photosynthetically active radiation. In step S8, if the temperature in the greenhouse is lower than the threshold value, a temperature-raising strategy is adopted; if the temperature in the greenhouse is higher than the threshold value, a cooling strategy is adopted. If the photosynthetically active radiation in the greenhouse is lower than a threshold value, a light supplement strategy is adopted; if the photosynthetically active radiation in the greenhouse is above a threshold value, a shading strategy is employed.
Specifically, in step S8, the execution operation of the device control execution module 40 includes the following cases:
and S81, predicting the temperature in the future preset time period.
(1) The plant control executive module 40 implements a cooling strategy when the current air temperature in the greenhouse is above an optimum high value.
(2) The plant control enforcement module 40 implements a warming strategy when the current air temperature in the greenhouse is below an optimum low value.
(3) When t 'is satisfied under the condition that the current air temperature in the greenhouse is within a suitable value range'max-t＜tthred_airtempThe device control execution module 40 implements a cooling strategy.
(4) When t 'is satisfied under the condition that the current air temperature in the greenhouse is within a suitable value range'min-t＜tthred_airtempThe device control execution module 40 implements the temperature increase strategy.
S81' prediction of photosynthetically active radiation in a predetermined time period in the future.
(5) When the current effective radiation of light in the greenhouse is higher than the optimum high value, the device control execution module 40 implements a shading strategy;
(6) when the current light effective radiation in the greenhouse is lower than the optimum low value, the equipment control execution module 40 implements a light supplement strategy;
(7) when t 'is satisfied under the condition that the current light effective radiation in the greenhouse is in a suitable value range'max-t＜tthred_PIRThen, the device control execution module 40 implements a shading policy;
(8) when t 'is satisfied under the condition that the current light effective radiation in the greenhouse is in a suitable value range'min-t＜tthred_PIRWhen the device control execution module 40 executes the light supplement strategy;
wherein t is the current time point, t'maxPredicted value for the first time higher than optimum high value corresponds to time point, t'minFor the first time below the predicted value of the optimum low value, tthred_airtempCritical time value, t, for a greenhouse air temperature start strategythred_PIRThe critical time value of the greenhouse photosynthetically active radiation starting strategy needs to be determined according to the actual greenhouse equipment performance.
For example, when the crop is a facility tomato in the region of Jiangsu, the control strategy of the starting apparatus control execution module 40 in the above-mentioned (1) to (8) can be made according to the growing environment and the crop characteristics of the crop. When positive values are spring and fall, tthred_airtemp＝0.5h、tthred_PIR0.2 h; when positive values are in summer and winter, tthred_airtemp＝1h、tthred_PIR＝0.5h。
The embodiment of the present application further provides a greenhouse environment regulation and control system, including: the system comprises a data monitoring module 20, a cloud platform service module 30 and an equipment control execution module 40, wherein the cloud platform service module 30 comprises a data storage and processing module 31 and a greenhouse environment intelligent decision module 32.
Specifically, the data monitoring module 20 is configured to collect data to be processed, where the data to be processed includes meteorological data outside the greenhouse, environmental data inside the greenhouse, and crop body data. The data monitoring module 20 includes: the wireless sensor is used for collecting meteorological data outside the greenhouse, environmental data inside the greenhouse and crop body data. The data storage and processing module 31 solves the greenhouse environment regulation and control strategy according to the data to be processed, and calculates the environmental factor prediction model based on the long-short term memory network method of the improved wolf optimization algorithm.
The greenhouse environment intelligent decision module 32 makes an environment equipment control strategy according to the environment factor prediction model and the greenhouse environment regulation and control strategy, the equipment control execution module 40 executes automatic control operation on environment data in the greenhouse according to the environment equipment control strategy, and the cloud platform service module 30 can receive information sent by the data monitoring module 20 and send the information to the equipment control execution module 40.
The environmental equipment control strategy comprises a weather classification strategy, a growth period strategy and a season time period strategy, and the following aspects are mainly contained:
(1) the weather classification strategies comprise sunny strategies, rainy strategies and rainy and snowy strategies.
(2) The growth period strategy is mainly divided into a seedling period, a flowering period, a fruit setting period and a harvesting period according to the growth rule of crops.
(3) The seasonal period is classified into seasonal categories including spring, summer, fall, and winter, and the seasonal period categories include morning, noon, afternoon, and evening.
(4) The regulation and control modes comprise temperature rise, temperature reduction, light supplement and shading.
Optionally, the meteorological data outside the greenhouse includes outdoor air temperature, outdoor air humidity, outdoor wind speed, and rainfall data; the environmental data in the greenhouse comprise indoor air temperature and humidity, illumination intensity, photosynthetic active radiation and carbon dioxide concentration data; the crop ontology data includes leaf area, stem thickness and fruit diameter data of the crop.
According to an embodiment of the present application, the data storage and processing module 31 comprises: the device comprises an allocation unit, an initialization unit, a position acquisition unit, an optimization parameter unit, a test unit and a prediction unit.
Specifically, the allocation unit allocates the sorted time series sample sets as a training set and a test set. The distribution unit can distribute according to the current weather condition when distributing. For example, in winter, sunny days and morning time periods, corresponding environment factor prediction models can be established in a combined and classified mode according to seasons, weather conditions and time periods of one day, and indoor air temperature and photosynthetic effective radiation can be accurately and dynamically measured. That is, for the accuracy of the model, the classified time series sample sets of the corresponding conditions are assigned as a training set and a test set.
The initialization unit can determine initial data of an improved gray wolf optimization algorithm, and when the initial data are determined, initialization can be performed by adopting a method of randomly generating a population, and the initial data comprise the number of gray wolfs, iteration times and position coordinates. The initialization unit can convert the parameters needing to be optimized by the long-short term memory network method into the position coordinates of the wolf pack. According to the embodiment of the application, 13 parameters needing to be optimized in the long-term and short-term memory network method are converted into the position coordinates of the wolf pack, namely
bf、
bi、
bc、
boAnd
the conversion method is shown in step S41 and step S42, and will not be described herein.
The optimization parameter unit selects a training set, calculates the individual fitness value of the wolf colony by using the improved wolf optimization algorithm to reach the maximum iteration times, and obtains the optimal weight and the threshold parameter value of the long-short term memory network algorithm. When a training set is selected, the improved grey wolf optimization algorithm can be used for calculating the individual fitness value of the wolf colony, the maximum iteration times are reached, and the optimal weight and the threshold parameter value of the long-short term memory network algorithm are obtained.
Wherein, the function of the individual fitness of the wolf group uses the following least square function
It means that the bigger fitness function value represents the closer the wolf pack position is to the prey, where y is the output value of LSTM and y' is the actual value.
And the testing unit tests the long-short term memory network method optimized by the optimization parameter unit to obtain an optimal long-short term memory network model.
The prediction unit inputs the data to be processed, such as weather data outside the greenhouse and environmental data inside the greenhouse, which are acquired in real time by the data monitoring module 20, into the optimal long-short term memory network model to predict the data to be processed within a preset time period. For example, the air temperature and the photosynthetically active radiation in the greenhouse are predicted over a preset period of time.
Optionally, the optimization parameter unit defines an individual corresponding to the optimal fitness value as an α wolf, a suboptimal individual as a β wolf, and a third best as a δ wolf, and includes a position updating module, which is configured to update the position of the remaining gray wolfs according to α, β, and δ, and adopts the following formula:
X(t+1)＝X1·Wα+X2·Wβ+X3·Wδ. The calculation method is as shown in step S42, and is not described herein again.
According to the greenhouse environment regulation and control method and system with the environment factor prediction function, the long-term and short-term memory network parameters are optimized based on the improved grayling optimization algorithm, the defect that the grayling algorithm is easy to fall into local optimization can be effectively overcome, and the adopted LSTM has a good effect on predicting time sequences.
Therefore, according to the greenhouse environment regulation and control method and system with the environment factor prediction function, the environment regulation and control threshold value corresponding to the current greenhouse can be given according to the current weather condition, the current crop growth period, the current season and the current time period by combining the crop state in the greenhouse, the current regulation and control strategy is determined by utilizing the environment factor value predicted by the model, and the greenhouse equipment executes operation according to the regulation and control strategy.
Optionally, an embodiment of the present application further provides an electronic device, which includes a processor, a memory, and a program or an instruction stored in the memory and capable of running on the processor, where the program or the instruction is executed by the processor to implement each process of the above greenhouse environment regulation and control method with an environmental factor prediction function, and can achieve the same technical effect, and in order to avoid repetition, details are not repeated here.
It should be noted that the electronic devices in the embodiments of the present application include the mobile electronic devices and the non-mobile electronic devices described above.
The embodiment of the present application further provides a readable storage medium, where a program or an instruction is stored on the readable storage medium, and when the program or the instruction is executed by a processor, the program or the instruction implements each process of the above greenhouse environment regulation and control method with an environment factor prediction function, and can achieve the same technical effect, and in order to avoid repetition, the detailed description is omitted here.
The processor is the processor in the electronic device described in the above embodiment. The readable storage medium includes a computer readable storage medium, such as a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and the like.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element. Further, it should be noted that the scope of the methods and apparatus of the embodiments of the present application is not limited to performing the functions in the order illustrated or discussed, but may include performing the functions in a substantially simultaneous manner or in a reverse order based on the functions involved, e.g., the methods described may be performed in an order different than that described, and various steps may be added, omitted, or combined. In addition, features described with reference to certain examples may be combined in other examples.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solutions of the present application may be embodied in the form of a software product, which is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal (such as a mobile phone, a computer, a server, an air conditioner, or a network device) to execute the method according to the embodiments of the present application.
While the present embodiments have been described with reference to the accompanying drawings, it is to be understood that the invention is not limited to the precise embodiments described above, which are meant to be illustrative and not restrictive, and that various changes may be made therein by those skilled in the art without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (10)
1. A greenhouse environment regulation and control method with an environmental factor prediction function is characterized by comprising the following steps:
s1, collecting environmental data in the greenhouse and meteorological data outside the greenhouse in a preset time period;
s2, processing the environmental data and the meteorological data to obtain a sample set, and dividing the sample set into a training set and a testing set;
s3, determining initial data of the improved wolf optimization algorithm;
s4, converting parameters needing to be optimized in the long and short term memory network method into position coordinates of the wolf cluster, and optimizing the position coordinates by using an improved wolf optimization algorithm to obtain optimized parameters of the long and short term memory network method;
s5, selecting a test set, and testing the long-short term memory network method of the optimized parameters to obtain an optimal long-short term memory network method model;
s6, collecting environmental data in the greenhouse and meteorological data outside the greenhouse in real time to obtain real-time data;
s7, inputting the real-time data into an optimal long-term and short-term memory network method model, and predicting the air temperature in a temperature measuring room and the future value of photosynthetically active radiation;
and S8, regulating and controlling the temperature and the light of the greenhouse according to the predicted result.
2. The method for regulating and controlling greenhouse environment according to claim 1, wherein in step S2, the environmental data and the meteorological data are processed by error data elimination, data supplementation and data normalization to obtain the sample set.
3. The greenhouse environment regulating method with the environmental factor predicting function as recited in claim 1, wherein in step S3, the initial data includes the number of grayworms, the number of iterations, and the dimension.
4. The method for greenhouse environment regulation with environmental factor prediction function of claim 1, wherein in step S4, the parameters to be optimized include: a forgetting gate weight parameter, a forgetting gate threshold parameter, an input gate weight parameter, an input gate threshold parameter, a candidate gate weight parameter, a candidate gate threshold parameter, an output gate weight parameter, an output gate threshold parameter, and an output layer weight parameter.
5. The greenhouse environment regulation method with environmental factor prediction function of claim 1, wherein in step S8,
when the current air temperature in the greenhouse is higher than the optimum high value, the equipment control execution module implements a cooling strategy;
when the current air temperature in the greenhouse is lower than the optimum low value, the equipment control execution module implements a temperature rise strategy;
when t 'is satisfied under the condition that the current air temperature in the greenhouse is within a suitable value range'max-t＜tthred_airtempWhen the equipment is used, the equipment control execution module implements a cooling strategy;
when t 'is satisfied under the condition that the current air temperature in the greenhouse is within a suitable value range'min-t＜tthred_airtempThe device control execution module implements the temperature rise strategyA little bit;
when the current light effective radiation in the greenhouse is higher than the optimum high value, the equipment controls the execution module to implement a shading strategy;
when the current light effective radiation in the greenhouse is lower than the optimum low value, the equipment controls the execution module to implement a light supplement strategy;
when t 'is satisfied under the condition that the current light effective radiation in the greenhouse is in a suitable value range'max-t＜tthred_PIRWhen the device is in use, the device control execution module implements shading strategy;
when t 'is satisfied under the condition that the current light effective radiation in the greenhouse is in a suitable value range'min-t＜tthred_PIRWhen the device is in use, the device control execution module implements a light supplement strategy;
wherein t is the current time point, t'maxPredicted value for the first time higher than optimum high value corresponds to time point, t'minFor the first time below the predicted value of the optimum low value, tthred_airtempCritical time value, t, for a greenhouse air temperature start strategythred_PIRCritical time values for a greenhouse photosynthetically active radiation startup strategy.
6. A greenhouse environment regulation system, comprising:
the data monitoring module is used for acquiring data to be processed;
the cloud platform service module comprises a data storage and processing module and a greenhouse environment intelligent decision-making module, the data storage and processing module is used for solving a greenhouse environment regulation and control strategy according to the data to be processed and calculating an environment factor prediction model based on a long-term and short-term memory network method of an improved wolf optimization algorithm, the greenhouse environment intelligent decision-making module is used for making an environment equipment control strategy according to the environment factor prediction model and the greenhouse environment regulation and control strategy, and the cloud platform service module can receive information sent by the data monitoring module;
and the equipment control execution module executes automatic control operation on environmental data in the greenhouse according to the environmental equipment control strategy, and the cloud platform service module can send information to the equipment control execution module.
7. The greenhouse environment conditioning system of claim 6, wherein the data to be processed includes meteorological data outside the greenhouse, environmental data inside the greenhouse, and crop ontology data, the data monitoring module comprising:
and the wireless sensor acquires meteorological data outside the greenhouse, environmental data inside the greenhouse and the crop body data.
8. The greenhouse environment conditioning system of claim 7, wherein the greenhouse outdoor meteorological data comprises outdoor air temperature, outdoor air humidity, outdoor wind speed, and rainfall data;
the greenhouse internal environment data comprise indoor air temperature and humidity, illumination intensity, photosynthetic effective radiation and carbon dioxide concentration data;
the crop ontology data includes leaf area, stem thickness and fruit diameter data of the crop.
9. The greenhouse environment regulation system of claim 6, wherein the data storage and processing module comprises:
an allocation unit that allocates the sorted time series sample sets into a training set and a test set;
an initialization unit capable of determining initial data that improves a graying optimization algorithm;
the position acquisition unit is used for converting the parameters needing to be optimized in the long-short term memory network method into position coordinates of the wolf pack;
the optimization parameter unit selects the training set, calculates the individual fitness value of the wolf group by using an improved wolf optimization algorithm to reach the maximum iteration times, and obtains the optimal weight and the threshold parameter value of the long-short term memory network algorithm;
the testing unit tests the long-short term memory network method optimized by the optimization parameter unit to obtain an optimal long-short term memory network model;
and the prediction unit inputs the data to be processed acquired in real time through the data monitoring module into the optimal long-short term memory network model so as to predict the data to be processed in a preset time period.
10. The greenhouse environment regulation system of claim 9, wherein the optimization parameter unit defines an individual corresponding to the optimal fitness value as an α wolf, a suboptimal fitness value as a β wolf, and a third best fitness value as a δ wolf, and the optimization parameter unit comprises:
the position updating module is used for updating the positions of the rest wolfs according to alpha, beta and delta, and the adopted formula is as follows:
X(t+1)＝X1·Wα+X2·Wβ+X3·Wδ；
wherein D isα，DβAnd DδRespectively representing the distances between alpha, beta and delta and other wolfs, t is the current iteration number, Xα(t)、Xβ(t) and Xδ(t) represents the positions of α, β and δ at the current iteration number, respectively, X (t) is the current gray wolf position, Wα，WβAnd WδRespectively representing the weight occupied by alpha, beta and delta,fα、fβand fδRepresenting fitness values for alpha, beta, and delta, respectively, and X (t +1) representing the location of the gray wolf for the next iteration.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011573255.9A CN112527037B (en) | 2020-12-24 | 2020-12-24 | Greenhouse environment regulation and control method and system with environment factor prediction function |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011573255.9A CN112527037B (en) | 2020-12-24 | 2020-12-24 | Greenhouse environment regulation and control method and system with environment factor prediction function |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112527037A true CN112527037A (en) | 2021-03-19 |
| CN112527037B CN112527037B (en) | 2022-02-08 |
Family
ID=74976555
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011573255.9A Active CN112527037B (en) | 2020-12-24 | 2020-12-24 | Greenhouse environment regulation and control method and system with environment factor prediction function |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112527037B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113283165A (en) * | 2021-05-20 | 2021-08-20 | 成都长弓乐水农业科技有限责任公司 | Greenhouse data synchronous simulation system and method based on cloud computing |
| CN113867451A (en) * | 2021-10-29 | 2021-12-31 | 江苏大学 | Intelligent temperature control method for livestock and poultry breeding by combining cloud platform and controller |
| CN115686110A (en) * | 2023-01-07 | 2023-02-03 | 广州市农业科学研究院 | Greenhouse intelligent control method, system, monitoring device, equipment and medium |
| TWI806064B (en) * | 2021-06-08 | 2023-06-21 | 遠東科技大學 | Method, computer program, and computer readable medium for controlling agricultural field environmental parameters in advance |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108388291A (en) * | 2018-01-17 | 2018-08-10 | 中国农业大学 | A kind of greenhouse cluster environment regulation and control method and system |
-
2020
- 2020-12-24 CN CN202011573255.9A patent/CN112527037B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108388291A (en) * | 2018-01-17 | 2018-08-10 | 中国农业大学 | A kind of greenhouse cluster environment regulation and control method and system |
Non-Patent Citations (2)
| Title |
|---|
| 王书芹等: "基于灰狼优化算法的长短期记忆网络在时间序列预测中的应用", 《中国科技论文》 * |
| 裴晓辉: "基于深度学习的智慧大棚监控系统的研究", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113283165A (en) * | 2021-05-20 | 2021-08-20 | 成都长弓乐水农业科技有限责任公司 | Greenhouse data synchronous simulation system and method based on cloud computing |
| TWI806064B (en) * | 2021-06-08 | 2023-06-21 | 遠東科技大學 | Method, computer program, and computer readable medium for controlling agricultural field environmental parameters in advance |
| CN113867451A (en) * | 2021-10-29 | 2021-12-31 | 江苏大学 | Intelligent temperature control method for livestock and poultry breeding by combining cloud platform and controller |
| CN113867451B (en) * | 2021-10-29 | 2022-08-23 | 江苏大学 | Intelligent temperature control method for livestock and poultry breeding by combining cloud platform and controller |
| CN115686110A (en) * | 2023-01-07 | 2023-02-03 | 广州市农业科学研究院 | Greenhouse intelligent control method, system, monitoring device, equipment and medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112527037B (en) | 2022-02-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112527037B (en) | Greenhouse environment regulation and control method and system with environment factor prediction function | |
| WO2022166919A1 (en) | Breeding environment temperature prediction and control system for livestock and poultry house, and adjustment and control method therefor | |
| Yu et al. | Prediction of the temperature in a Chinese solar greenhouse based on LSSVM optimized by improved PSO | |
| CN112288164B (en) | Wind power combined prediction method considering spatial correlation and correcting numerical weather forecast | |
| Gu et al. | Neural network soil moisture model for irrigation scheduling | |
| CN112116144B (en) | Regional power distribution network short-term load prediction method | |
| CN112906298B (en) | Blueberry yield prediction method based on machine learning | |
| CN106842914A (en) | A kind of temperature control energy-saving processing method, apparatus and system | |
| CN114077269B (en) | Greenhouse environment prediction and optimization control method based on data-driven model | |
| CN111580382A (en) | Unit-level heat supply adjusting method and system based on artificial intelligence | |
| CN105446142A (en) | Greenhouse CO2 gas fertilizer increasing method, device and system | |
| WO2023024615A1 (en) | System and method for performing precise irrigation on the basis of plant wilting degree | |
| CN112272074A (en) | Information transmission rate control method and system based on neural network | |
| CN111815039A (en) | Weekly scale wind power probability prediction method and system based on weather classification | |
| CN113962819A (en) | Method for predicting dissolved oxygen in industrial aquaculture based on extreme learning machine | |
| CN113591957B (en) | Wind power output short-term rolling prediction and correction method based on LSTM and Markov chain | |
| CN114092776A (en) | Multi-sensor data fusion method applied to intelligent agriculture | |
| Li et al. | Sustainable energy management of solar greenhouses using open weather data on MACQU platform | |
| Su et al. | Greenhouse climate setpoint optimization: An online decision strategy | |
| CN116109011B (en) | Energy consumption management method and terminal for intelligent park | |
| Körner et al. | An on-line tool for optimising greenhouse crop production | |
| CN116128168A (en) | Weather prediction method based on causal expansion convolution and Autoformer | |
| CN115907366A (en) | Agricultural product growth environment optimal regulation and control method and equipment based on flamingo algorithm | |
| CN115796040A (en) | Facility greenhouse temperature prediction method based on small samples | |
| CN116167508A (en) | Short-term photovoltaic output rapid prediction method and system based on meteorological factor decomposition |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |