CN113553767A - Method and system for building greenhouse crop photosynthetic rate prediction model - Google Patents

Method and system for building greenhouse crop photosynthetic rate prediction model Download PDF

Info

Publication number
CN113553767A
CN113553767A CN202110846097.8A CN202110846097A CN113553767A CN 113553767 A CN113553767 A CN 113553767A CN 202110846097 A CN202110846097 A CN 202110846097A CN 113553767 A CN113553767 A CN 113553767A
Authority
CN
China
Prior art keywords
training
photosynthetic rate
model
prediction model
value
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.)
Pending
Application number
CN202110846097.8A
Other languages
Chinese (zh)
Inventor
刘潭
袁青云
谭东明
王永刚
张大鹏
张楠楠
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenyang Agricultural University
Original Assignee
Shenyang Agricultural University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shenyang Agricultural University filed Critical Shenyang Agricultural University
Priority to CN202110846097.8A priority Critical patent/CN113553767A/en
Publication of CN113553767A publication Critical patent/CN113553767A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G06N20/20—Ensemble learning

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Medical Informatics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • Computing Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Mathematical Physics (AREA)
  • Computer Hardware Design (AREA)
  • Geometry (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The application discloses a method and a system for building a greenhouse crop photosynthetic rate prediction model, firstly, obtaining sample data, and carrying out normalization operation on the sample data to obtain a training sample set; then, constructing a photosynthetic rate prediction model based on the SOPSO-LSSVM, and training the photosynthetic rate prediction model by utilizing a training sample set, wherein the training process comprises the following steps: inputting the training sample set into an LSSVM model for training to obtain a predicted value of the light combination rate in the current training batch; obtaining the optimal parameter value of the current training batch through an SOPSO algorithm according to the predicted value, and updating the parameters in the LSSVM model by using the optimal parameter value; and finally, when the training batch reaches a preset condition, obtaining a greenhouse crop photosynthetic rate prediction model. As can be seen, the model can accurately predict CO at different temperatures2The photosynthetic rate under the conditions of concentration and PPFD can provide a good model base for the optimal control of greenhouse environment, and the photosynthetic rate under the conditions of concentration and PPFD can be compared with the current photosynthetic rateCompared with the method, the method has higher prediction precision and robustness.

Description

Method and system for building greenhouse crop photosynthetic rate prediction model
Technical Field
The invention relates to the field of biological information, in particular to a method and a system for constructing a greenhouse crop photosynthetic rate prediction model.
Background
Crop photosynthesis refers to a biochemical process of converting carbon dioxide and water into organic matters under a certain photon flux density condition to realize substance accumulation, which determines the yield and quality of crops. Therefore, creating a good microclimate environment to meet the photosynthetic demand of crops is the key to increasing photosynthetic rate, increasing material accumulation, increasing crop yield and quality.
In the prior art, a Support Vector Machine (SVM) technology is usually adopted to simulate the relationship between these environmental factors and the photosynthetic rate, however, when the number of input variables is increased, the prior art has the disadvantages of high fitting complexity, large error and the like.
Disclosure of Invention
Based on the above, the embodiment of the application provides a method and a system for building a greenhouse crop photosynthetic rate prediction model, which can improve the accuracy of photosynthetic rate prediction.
In a first aspect, a method for constructing a greenhouse crop photosynthetic rate prediction model is provided, and the method comprises the following steps:
acquiring sample data, and carrying out normalization operation on the sample data to obtain a training sample set;
constructing a photosynthetic rate prediction model based on the SOPSO-LSSVM, and training the photosynthetic rate prediction model by using the training sample set, wherein the training process comprises the following steps:
inputting the training sample set into an LSSVM model for training to obtain a predicted value of the light combination rate in the current training batch;
obtaining the optimal parameter value of the current training batch through an SOPSO algorithm according to the predicted value, and updating the parameters in the LSSVM model by using the optimal parameter value;
and when the training batch reaches a preset condition, obtaining a greenhouse crop photosynthetic rate prediction model.
Optionally, the sample data comprises temperature, humidity, photosynthetic photon flux density, carbon dioxide concentration and photosynthetic rate values.
Optionally, the temperature, the humidity, the photosynthetic photon flux density, and the carbon dioxide concentration are acquired through node acquisition of a wireless sensor network, and the photosynthetic rate value is measured by using a portable photosynthetic apparatus within a preset time.
Optionally, the sample data is normalized by a first formula to obtain a training sample set, where the first formula specifically includes:
Figure BDA0003180612930000021
wherein,
Figure BDA0003180612930000022
Xian input vector, X, representing the ith variablei' (j) denotes the j-th sample value of the i-th variable, i 1, 2.. the n denotes the number of variables, and j 1, 2.. the m denotes the number of samples.
Optionally, the constructing a photosynthetic rate prediction model based on the SOPSO-LSSVM includes:
the parameters of the SOPSO algorithm are set and then the velocity and position of the particles are initialized.
Optionally, the training sample set is input into an LSSVM model for training, and a predicted value of the photosynthetic rate is obtained according to a second formula, where the second formula specifically includes:
Figure BDA0003180612930000023
wherein,
Figure BDA0003180612930000024
Figure BDA0003180612930000025
indicating the predicted value of photosynthetic Rate, alphai≧ 0 denotes the lagrange multiplier, b denotes the bias term, I ═ 1,1]T denotes transposition, Ωi＝K(X,Xi) Representing elements in the matrix, σ representing a kernel function constant, γ representing a penalty factor, exp () representing an exponential function with e as the base, Xi,XjDenotes an input vector, i 1, 2., n denotes the number of variables, and j 1, 2., m denotes the number of samples.
Optionally, the obtaining the optimal value of the parameter of the current training batch through the SOPSO algorithm according to the predicted value specifically includes:
and calculating the fitness value of the current training batch according to the predicted value through a third formula, and obtaining the optimal parameter value according to the particle position when the fitness value is minimum.
Optionally, the third formula specifically includes:
Figure BDA0003180612930000031
where fit represents the fitness value, yiRepresents the actual measurement value of the ith sample,
Figure BDA0003180612930000032
represents the predicted value of the ith sample, and n represents the number of predicted samples.
Optionally, the preset condition includes: the maximum number of iterations of the model training or the minimum fitness value obtained in the model training.
In a second aspect, a greenhouse crop photosynthetic rate prediction model construction system is provided, which comprises:
an obtaining module, configured to obtain sample data, and perform normalization operation on the sample data to obtain a training sample set
The training module is used for constructing a photosynthetic rate prediction model based on the SOPSO-LSSVM and training the photosynthetic rate prediction model by utilizing the training sample set, wherein the training process comprises the following steps:
inputting the training sample set into an LSSVM model for training to obtain a predicted value of the light combination rate in the current training batch;
obtaining the optimal parameter value of the current training batch through an SOPSO algorithm according to the predicted value, and updating the parameters in the LSSVM model by using the optimal parameter value;
and the generation module is used for obtaining the greenhouse crop photosynthetic rate prediction model when the training batch reaches the preset condition.
The technical scheme provided by the embodiment of the application provides a method for constructing a greenhouse crop photosynthetic rate prediction model, and the method comprises the following steps of firstly, obtaining sample data, and carrying out normalization operation on the sample data to obtain a training sample set; then, constructing a photosynthetic rate prediction model based on the SOPSO-LSSVM, and training the photosynthetic rate prediction model by utilizing a training sample set, wherein the training process comprises the following steps: inputting the training sample set into an LSSVM model for training to obtain a predicted value of the light combination rate in the current training batch; obtaining the optimal parameter value of the current training batch through an SOPSO algorithm according to the predicted value, and updating the parameters in the LSSVM model by using the optimal parameter value; and finally, when the training batch reaches a preset condition, obtaining a greenhouse crop photosynthetic rate prediction model. As can be seen, the model can accurately predict CO at different temperatures2The tomato photosynthetic rate under the conditions of concentration and PPFD can provide a good model foundation for the optimal control of greenhouse environment.
Drawings
FIG. 1 is a flowchart of a method for constructing a model for predicting photosynthetic rate of greenhouse crops according to an embodiment of the present disclosure;
FIG. 2 is a block diagram of a photosynthetic rate measurement process provided by an embodiment of the present application;
FIG. 3 is a predicted value of sample data based on a BP model using an embodiment;
FIG. 4 is a predicted value based on an LSSVM model using sample data of an embodiment;
FIG. 5 is a predicted value based on the SOPSO-LSSVM model using sample data of an embodiment;
fig. 6 is a block diagram of a greenhouse crop photosynthesis rate prediction model construction system according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
In the description of the present invention, "a plurality" means two or more unless otherwise specified. The terms "first," "second," "third," "fourth," and the like in the description and claims of the present invention and in the above-described drawings (if any) are intended to distinguish between the referenced items. For a scheme with a time sequence flow, the term expression does not need to be understood as describing a specific sequence or a sequence order, and for a scheme of a device structure, the term expression does not have distinction of importance degree, position relation and the like.
Furthermore, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements specifically listed, but may include other steps or elements not expressly listed that are inherent to such process, method, article, or apparatus or that are added to a further optimization scheme based on the present inventive concept.
According to the embodiment of the application, the crop tomato with the largest facility cultivation area in China is selected as a research object, and a photosynthesis rate prediction model under the influence of multi-environment factors is established by using the LSSVM. However, since some parameters in the LSSVM model are usually given by experience, the prediction performance of the model will be affected. In order to quickly and accurately determine the optimal values of the parameters, the method further utilizes a second-order oscillatory particle swarm optimization (SOPSO) to solve the parameter optimization problem of the LSSVM model and determine the optimal values of the LSSVM model. Namely, the SOPSO algorithm is integrated into the LSSVM model, and the tomato photosynthetic rate prediction model based on the SOPSO-LSSVM is established. The research of the application can establish an accurate photosynthetic rate prediction model for regulating and controlling the greenhouse luminous environment, and provides reliable objective function support for obtaining the luminous environment regulation and control dynamic target considering photosynthetic requirements.
Referring to fig. 1, a flowchart of a method for constructing a greenhouse crop photosynthetic rate prediction model provided in an embodiment of the present application is shown, where the method for constructing the greenhouse crop photosynthetic rate prediction model may include the following steps:
step 101, obtaining sample data, and performing normalization operation on the sample data to obtain a training sample set.
The system comprises a wireless sensor network, a portable photosynthetic apparatus and a sample data, wherein the sample data comprises temperature, humidity, photosynthetic photon flux density, carbon dioxide concentration and photosynthetic rate values, the temperature, the humidity, the photosynthetic photon flux density and the carbon dioxide concentration are acquired through node acquisition of the wireless sensor network, and the photosynthetic rate values are acquired through measurement by the portable photosynthetic apparatus within preset time.
In the examples of the present application, temperature, humidity, Photosynthetic Photon Flux Density (PPFD) and CO were selected by analyzing the mechanism of the effect of environmental factors on photosynthesis in a greenhouse2And environmental factors such as concentration and the like which have large influence on the photosynthetic rate are used as the input of the photosynthetic rate soft measurement model, and accordingly, the photosynthetic rate is selected as the output. Therefore, to build a soft measurement model for predicting photosynthetic rate, information on environmental factors and photosynthetic rate needs to be collected in order to obtain sample data. The environmental factor information is obtained by a node of the wireless sensor network. Firstly, each environmental factor sensor node suspended 10cm above a measured blade is used for measuring environmental factors in real time, then a wireless gateway transmits the measured value of each environmental factor to a remote data management center, and the acquisition interval of the environmental factor parameters is 30 min. Through photosynthetic rate test under the nested combination of multiple environmental factors, the output sample data is collected, and the experimental region and the specific method are as follows:
the experimental greenhouse adopted in the embodiment of the application is a sunlight greenhouse located in the northern mountain experimental base of Shenyang agricultural university, and the tested tomato variety is Liaoyueli, which has the characteristics of vigorous growth, no yellow leaves, no premature senility and high yield. Firstly, healthy and plump tomato seeds are selected, soaked and then sowed in a 50-hole plug tray for seedling raising. In the process of seedling culture, a professional substrate with the organic matter content of 50 percent, the pH value of 5.5-6.5 and the humic acid content of 20 percent is used, and the water and fertilizer are sufficient. When the tomato seedlings grow to 3 leaves and 1 heart, the plant height is about 15 cm, and the stem thickness is about 0.4 cm, the seedlings can be transplanted. Until the width of the 4 th leaf is more than 3cm from top to bottom, selecting tomato seedlings with good illumination, irrigation and growth for photosynthesis test, and not using pesticide and hormone medicine in the period.
In order to avoid the influence of the 'noon break' phenomenon of the plant on the test result, the photosynthetic rate test is carried out between 9:00-11:30 and 14:30-17:30, the 4 th blade with vigorous growth and consistent growth is selected, the photosynthetic rate value is measured by using a portable photosynthetic apparatus, and the measurement process is shown in figure 2.
The photosynthetic rate is known to be at temperature (X)1) Humidity (X)2),PPFD(X3) And CO2Concentration (X)4) The corresponding environmental factor gradient settings, measured under nested combination conditions, are shown in table 1.
Table 1 experimental environment variable gradient setup
Figure BDA0003180612930000061
Carrying out normalization operation on the sample data through a first formula to obtain a training sample set, wherein the first formula specifically comprises the following steps:
Figure BDA0003180612930000071
wherein,
Figure BDA0003180612930000072
Xian input vector, X, representing the ith variablei' (j) denotes the j-th sample value of the i-th variable, i 1, 2.. the n denotes the number of variables, and j 1, 2.. the m denotes the number of samples.
And 102, constructing a photosynthetic rate prediction model based on the SOPSO-LSSVM, and training the photosynthetic rate prediction model by utilizing a training sample set.
Wherein, the training process includes: inputting the training sample set into an LSSVM model for training to obtain a predicted value of the light combination rate in the current training batch; and obtaining the optimal parameter value of the current training batch through an SOPSO algorithm according to the predicted value, and updating the parameters in the LSSVM model by using the optimal parameter value.
In the embodiment of the present application, the specific process of the training process includes:
step 1021, set the parameters of the SOPSO algorithm, and then initialize the velocity and position [ γ ] of the particlei,σi]I 1,2, M denotes the total number of particles.
Step 1022, putting the training samples into the LSSVM model for learning to obtain the predicted value of the photosynthetic rate, and calculating the current [ gamma ], (Y)i,σi]Lower fitness value
Suppose training sample set S ═ (X)i,yi),i＝1,2,...,n，Xi∈RnFor the input vector of the ith sample, the input vector comprises temperature, humidity, PPFD and CO2Concentration, yiIs the corresponding output, i.e. the photosynthetic rate. A decision function is constructed using a training sample set S, which can be expressed as:
Figure BDA0003180612930000073
wherein,
Figure BDA0003180612930000074
and representing a function of mapping the input vector to a high-dimensional space, wherein omega epsilon R is a weight coefficient, b is a deviation term, and the constructed structure risk function is as follows:
Figure BDA0003180612930000075
where J is the objective function and gamma is a penalty factor, the value of which determines the penalty on the errorDegree, ζT＝[ζ1,ζ2,...,ζn]Is an allowable error vector. To obtain the ω and b values, based on the principle of minimization of structural risk, when using LSSVM for regression, the following optimization problem needs to be calculated:
Figure BDA0003180612930000081
Figure BDA0003180612930000082
by using the Lagrange function and its dual function, the above-mentioned optimization objective function can be converted into:
Figure BDA0003180612930000083
wherein alpha isiLagrange multipliers are greater than or equal to 0.
Calculate L pairs of ω, b, ζ, respectivelyi，αiSuch that they are 0:
Figure BDA0003180612930000084
Figure BDA0003180612930000085
Figure BDA0003180612930000086
Figure BDA0003180612930000087
through simplification, the following results are obtained:
Figure BDA0003180612930000089
b and a can be obtained by solving the equationiThe value of (c).
Wherein, I ═ 1, 1., 1]，Ωi＝K(X,Xi) Representing an element in the matrix, K (X, X)i) The expression of (a) is:
Figure BDA00031806129300000810
because the Gaussian radial basis function requires fewer preset parameters, it can be used to solve practical problems. Therefore, it is chosen as a kernel function, which can be expressed as:
Figure BDA00031806129300000811
where σ represents a kernel function constant.
Furthermore, an LSSVM-based photosynthetic rate prediction model can be established, and can be expressed as a second formula, where the second formula specifically includes:
Figure BDA0003180612930000091
Figure BDA0003180612930000092
indicating the predicted value of the photosynthetic rate.
Aiming at the problem of parameter selection of the LSSVM model, namely determining the optimal values of a penalty coefficient gamma and a kernel constant sigma in the LSSVM model, the method adopts a second-order oscillatory particle swarm optimization (SOPSO) to solve the problem.
Step 1023, obtaining the optimal parameter value of the current training batch through an SOPSO algorithm according to the predicted value,
in the embodiment of the present application, the update calculation formula of the velocity and position of the particle in space in the SOPSO algorithm is as follows:
vid(k+1)＝ωvid(k)+c1r1(k)(pbestid(k)-(1+ξ1)xid(k)-ξ1xid(k-1))+c2r2(k)(gbestid(k)-(1+ξ2)xid(k)-ξ2xid(k-1))
xid(k+1)＝xid(k)+vid(k+1)
wherein v isid(i 1, 2.. the m, d 1, 2.. the n) is the speed of the particles in d-dimensional space, m is the total number of particles, and k 1, 2.. the k is the 1, 2.. the kmaxIs the number of iterations. x is the number ofidFor the current particle position (solution) in d-dimensional space, ω is the inertial weight, which is set to 0.9 in the embodiment of the present application. pbestidRepresenting the best position (fitness) of each particle so far, gbestidRepresenting the global optimal position. r is1And r2Is at [0,1 ]]Two random numbers in between. c. C1And c2For the learning factor, 0.95 and 1.0 are set in the embodiment of the present application, respectively. In the SOPSO algorithm, ξ1And xi2Are two random functions. In the initial stage of the SOPSO algorithm (k < k)max/4)，ξ1And xi2Are respectively arranged as
Figure BDA0003180612930000093
And
Figure BDA0003180612930000094
the algorithm is oscillation convergence and has strong global searching capability. When the algorithm is executed to the later stage (k ≧ k)max/4)，ξ1And xi2Are respectively arranged as
Figure BDA0003180612930000095
And
Figure BDA0003180612930000096
the asymptotic convergence of the algorithm can be ensured. In addition, a fitness function is constructed by using the predicted root mean square error, and is expressed as a third formula:
the third formula specifically includes:
Figure BDA0003180612930000101
wherein, yiRepresents the actual measurement value of the ith sample,
Figure BDA0003180612930000102
and (3) representing the predicted value of the ith sample, wherein n represents the number of predicted samples, and when fit is minimum, the corresponding particle position is the optimal value of gamma and sigma.
In summary, the particle [ y ] is found according to the third formulai,σi]Local optimum position pbest ofidAnd global optimal position gbestid。
And 103, when the training batch reaches a preset condition, obtaining a greenhouse crop photosynthetic rate prediction model.
In the embodiment of the application, whether the maximum iteration number is reached or the minimum adaptive value is obtained is judged, and if the maximum iteration number is reached or the minimum adaptive value is obtained, the iteration is ended. At the same time, the corresponding particle position represents the optimal value of the parameter. Otherwise, the particle [ y ] is updated according to the third formulai,σi]Recalculating the adaptive values of the updated parameters, and searching the local optimal position and the global optimal position again.
Through the calculation, the optimal values of gamma and sigma are obtained, namely, the photosynthetic rate prediction model based on the SOPSO-LSSVM is obtained through the training sample set S.
In another embodiment, 900 sets of sample data are obtained for the photosynthetic rate prediction model of the SOPSO-LSSVM of the present application, and 720 sets of samples accounting for 80% of the total number of samples are selected as the training sample set S of the model. The other 180 sets were used as test sample set T to validate the model, where the temperature (X)1) Humidity (X)2),PPFD(X3) And CO2Concentration (X)4) Constituting an input vector with the photosynthetic rate (y) as the output. An SOPSO-LSSVM-based photosynthetic rate prediction model is established by utilizing an MATLAB programming platform, firstly, a penalty coefficient gamma and a kernel function constant sigma of the LSSVM model are optimized by utilizing an SOPSO algorithm, and the optimized values are 0.21 and 0 respectively.098. And then substituting the optimal value into an LSSVM model to construct a photosynthetic rate prediction model. After the SOPSO-LSSVM model is constructed, a photosynthetic rate prediction model is respectively established by using the same training set and adopting a BP neural network and a standard LSSVM algorithm for comparative analysis. In order to verify the prediction performance of the model, the same sample is predicted by using the three models, and the prediction result is shown in fig. 3. The network structure of the BP neural network is 5-5-1, namely 5 nodes are arranged on an input layer, 5 nodes are arranged on a hidden layer, and 1 node is arranged on an output layer. Fig. 3 to 5 show the prediction results of 3 models, respectively, wherein the black solid line indicates the measurement result, and the gray dashed line indicates the prediction result.
In order to further evaluate the prediction performance of the photosynthetic rate model based on the SOPSO-LSSVM, the Maximum Relative Error (MRE) and the Root Mean Square Error (RMSE) of the 3 models are given, respectively, and the results are shown in table 2.
Comparison of MRE and RMSE values in the 23 models
Figure BDA0003180612930000111
Wherein MRE is calculated
Figure BDA0003180612930000112
Obtaining;
RMSE by calculation
Figure BDA0003180612930000113
And (4) obtaining.
Therefore, the photosynthetic rate prediction model based on the SOPSO-LSSVM has obvious advantages in prediction precision, and can provide a stable model basis for the optimal control of the greenhouse light environment based on photosynthetic demand.
The application provides a novel regression modeling method, which optimizes partial parameters in an LSSVM model by utilizing an SOPSO algorithm. Firstly, under the condition of multi-environmental factor nested combination, sample data is obtained through a photosynthetic rate experiment. Then, parameters of the LSSVM model are optimized by utilizing the SOPSO algorithm, and the temperature based on the SOPSO-LSSVM algorithm is establishedA model for predicting the photosynthetic rate of the indoor tomato crop. Simulation results show that the model is well represented on a test set, and has higher prediction precision and robustness compared with a commonly used BP neural network model and a standard LSSVM model. The model can accurately predict CO at different temperatures2The tomato photosynthetic rate under the conditions of concentration and PPFD can provide a good model foundation for the optimal control of greenhouse environment.
Referring to fig. 6, a schematic diagram of a greenhouse crop photosynthetic rate prediction model construction system 600 provided by an embodiment of the present application is shown, the method includes:
an obtaining module 601, configured to obtain sample data, and perform normalization operation on the sample data to obtain a training sample set
A training module 602, configured to construct a photosynthetic rate prediction model based on the SOPSO-LSSVM, and train the photosynthetic rate prediction model by using a training sample set, where the training process includes:
inputting the training sample set into an LSSVM model for training to obtain a predicted value of the light combination rate in the current training batch;
obtaining the optimal parameter value of the current training batch through an SOPSO algorithm according to the predicted value, and updating the parameters in the LSSVM model by using the optimal parameter value;
and the generating module 603 is configured to obtain a greenhouse crop photosynthesis rate prediction model when the training batch reaches a preset condition.
For specific limitations of the construction system of the photosynthesis rate prediction model of the greenhouse crop, reference may be made to the above limitations of the construction method of the photosynthesis rate prediction model of the greenhouse crop, and details are not repeated here. All or part of each module in the greenhouse crop photosynthesis rate prediction model construction system can be realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in M forms, such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), synchronous Link (SyMchliMk) DRAM (SLDRAM), RaMbus (RaMbus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
The technical features of the embodiments described above may be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the embodiments described above are not described, but should be considered as being within the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the claims. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims (10)

1. A method for constructing a greenhouse crop photosynthetic rate prediction model is characterized by comprising the following steps of:
acquiring sample data, and carrying out normalization operation on the sample data to obtain a training sample set;
constructing a photosynthetic rate prediction model based on the SOPSO-LSSVM, and training the photosynthetic rate prediction model by using the training sample set, wherein the training process comprises the following steps:
inputting the training sample set into an LSSVM model for training to obtain a predicted value of the light combination rate in the current training batch;
obtaining the optimal parameter value of the current training batch through an SOPSO algorithm according to the predicted value, and updating the parameters in the LSSVM model by using the optimal parameter value;
and when the training batch reaches a preset condition, obtaining a greenhouse crop photosynthetic rate prediction model.
2. The method of claim 1, wherein said sample data comprises temperature, humidity, photosynthetic photon flux density, carbon dioxide concentration, and photosynthetic rate values.
3. The method of claim 2, wherein the temperature, humidity, photosynthetic photon flux density and carbon dioxide concentration are acquired by nodes of a wireless sensor network, and the photosynthetic rate value is measured by using a portable photosynthetic apparatus within a preset time.
4. The method according to claim 1, wherein the normalizing the sample data by a first formula to obtain a training sample set, the first formula specifically comprises:
Figure FDA0003180612920000011
wherein,
Figure FDA0003180612920000012
Xian input vector, X ', representing the ith variable'i(j) The jth sample value representing the ith variable, i 1, 2.
5. The method of claim 1, wherein the constructing the SOPSO-LSSVM-based photosynthetic rate prediction model comprises:
the parameters of the SOPSO algorithm are set and then the velocity and position of the particles are initialized.
6. The method of claim 1, wherein the training sample set is input into an LSSVM model for training, and the predicted value of the photosynthetic rate is obtained according to a second formula, wherein the second formula specifically comprises:
Figure FDA0003180612920000021
wherein,
Figure FDA0003180612920000022
Figure FDA0003180612920000023
indicating the predicted value of photosynthetic Rate, alphai≧ 0 denotes the lagrange multiplier, b denotes the bias term, I ═ 1,1]T denotes transposition, Ωi＝K(X,Xi) Representing elements in the matrix, σ representing a kernel function constant, γ representing a penalty factor, exp () representing an exponential function with e as the base, Xi,XjDenotes an input vector, i 1, 2., n denotes the number of variables, and j 1, 2., m denotes the number of samples.
7. The method according to claim 1, wherein obtaining the optimal value of the parameter of the current training batch through the SOPSO algorithm according to the predicted value specifically comprises:
and calculating the fitness value of the current training batch according to the predicted value through a third formula, and obtaining the optimal parameter value according to the particle position when the fitness value is minimum.
8. The method according to claim 7, wherein the third formula comprises in particular:
Figure FDA0003180612920000024
where fit represents the fitness value, yiRepresents the actual measurement value of the ith sample,
Figure FDA0003180612920000025
represents the predicted value of the ith sample, and n represents the number of predicted samples.
9. The method according to claim 1, wherein the preset condition comprises:
the maximum number of iterations of the model training or the minimum fitness value obtained in the model training.
10. A greenhouse crop photosynthetic rate prediction model construction system is characterized by comprising:
an obtaining module, configured to obtain sample data, and perform normalization operation on the sample data to obtain a training sample set
The training module is used for constructing a photosynthetic rate prediction model based on the SOPSO-LSSVM and training the photosynthetic rate prediction model by utilizing the training sample set, wherein the training process comprises the following steps:
inputting the training sample set into an LSSVM model for training to obtain a predicted value of the light combination rate in the current training batch;
obtaining the optimal parameter value of the current training batch through an SOPSO algorithm according to the predicted value, and updating the parameters in the LSSVM model by using the optimal parameter value;
and the generation module is used for obtaining the greenhouse crop photosynthetic rate prediction model when the training batch reaches the preset condition.
CN202110846097.8A 2021-07-26 2021-07-26 Method and system for building greenhouse crop photosynthetic rate prediction model Pending CN113553767A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110846097.8A CN113553767A (en) 2021-07-26 2021-07-26 Method and system for building greenhouse crop photosynthetic rate prediction model

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110846097.8A CN113553767A (en) 2021-07-26 2021-07-26 Method and system for building greenhouse crop photosynthetic rate prediction model

Publications (1)

Publication Number Publication Date
CN113553767A true CN113553767A (en) 2021-10-26

Family

ID=78104551

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110846097.8A Pending CN113553767A (en) 2021-07-26 2021-07-26 Method and system for building greenhouse crop photosynthetic rate prediction model

Country Status (1)

Country Link
CN (1) CN113553767A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114648214A (en) * 2022-03-14 2022-06-21 江西省农业科学院园艺研究所 Proportion allocation method and system for physiological and biochemical indexes of facility crops
CN115226516A (en) * 2022-07-07 2022-10-25 西北农林科技大学 Facility light and carbon dioxide environment cooperative regulation and control method

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105654203A (en) * 2015-12-31 2016-06-08 西北农林科技大学 Cucumber whole-course photosynthetic rate predicting model based on support vector machine, and establishing method
CN111310348A (en) * 2020-02-25 2020-06-19 江西理工大学 Material constitutive model prediction method based on PSO-LSSVM
CN112308288A (en) * 2020-09-29 2021-02-02 百维金科(上海)信息科技有限公司 Particle swarm optimization LSSVM-based default user probability prediction method
US20210049515A1 (en) * 2019-08-16 2021-02-18 China Institute Of Water Resources And Hydropower Research Prediction method and system of high slope deformation

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105654203A (en) * 2015-12-31 2016-06-08 西北农林科技大学 Cucumber whole-course photosynthetic rate predicting model based on support vector machine, and establishing method
US20210049515A1 (en) * 2019-08-16 2021-02-18 China Institute Of Water Resources And Hydropower Research Prediction method and system of high slope deformation
CN111310348A (en) * 2020-02-25 2020-06-19 江西理工大学 Material constitutive model prediction method based on PSO-LSSVM
CN112308288A (en) * 2020-09-29 2021-02-02 百维金科(上海)信息科技有限公司 Particle swarm optimization LSSVM-based default user probability prediction method

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
TAN LIU 等: "Prediction Model of Photosynthetic Rate Based on SOPSO-LSSVM for Regulation of Greenhouse Light Environment", 《ENGINEERING LETTERS》, pages 1 - 5 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114648214A (en) * 2022-03-14 2022-06-21 江西省农业科学院园艺研究所 Proportion allocation method and system for physiological and biochemical indexes of facility crops
CN114648214B (en) * 2022-03-14 2023-09-05 江西省农业科学院园艺研究所 Proportion allocation method and system for physiological and biochemical indexes of facility crops
CN115226516A (en) * 2022-07-07 2022-10-25 西北农林科技大学 Facility light and carbon dioxide environment cooperative regulation and control method
CN115226516B (en) * 2022-07-07 2023-09-26 西北农林科技大学 Cooperative regulation and control method for facility light and carbon dioxide environment

Similar Documents

Publication Publication Date Title
Navarro-Hellín et al. A decision support system for managing irrigation in agriculture
CN112906298B (en) Blueberry yield prediction method based on machine learning
BR112021001441A2 (en) method and system for estimating effective nitrogen applications in harvest
WO2019033055A1 (en) Development of complex agricultural simulation models from limited datasets
CN101480143B (en) Method for predicating single yield of crops in irrigated area
CN107329511A (en) Based on the vegetable aquaculture luminous environment high efficiency regulatory method and system between suitable root warm area
CN104521699A (en) Field intelligent irrigation on-line control management method
CN113553767A (en) Method and system for building greenhouse crop photosynthetic rate prediction model
CN107145941B (en) Method for dynamically acquiring light demand quantity in real time based on optimal light quality and photon flux density
CN116595333B (en) Soil-climate intelligent rice target yield and nitrogen fertilizer consumption determination method
CN105446142A (en) Greenhouse CO2 gas fertilizer increasing method, device and system
Yu et al. A PSO-XGBoost Model for Estimating Daily Reference Evapotranspiration in the Solar Greenhouse.
Priya et al. An IoT based gradient descent approach for precision crop suggestion using MLP
CN111915062B (en) Greenhouse crop water demand regulation and control method with water utilization rate and photosynthesis rate being coordinated
JP2021056573A (en) Crop growth prediction program, crop growth prediction method, and crop growth prediction device
CN105389452A (en) Cucumber whole-course photosynthetic rate prediction model based on neural network, and establishment method
Buis et al. The STICS crop model and associated software for analysis, parameterization, and evaluation
Wu et al. Simulation of daily maize evapotranspiration at different growth stages using four machine learning models in semi-humid regions of northwest China
Liu et al. Prediction model of photosynthetic rate based on SOPSO-LSSVM for regulation of greenhouse light environment
US20240341249A1 (en) Modeling of soil compaction and structural capacity for field trafficability by agricultural equipment from diagnosis and prediction of soil and weather conditions associated with user-provided feedback
CN105913326B (en) Constrain the Crop growing stage model Cultivar parameter optimization method of sex knowledge and the tactful Genetic Algorithm Fusion of elite individual
Wu et al. Crop yield estimation and irrigation scheduling optimization using a root-weighted soil water availability based water production function
EP3474167A1 (en) System and method for predicting genotype performance
Amiri et al. Performance of WOFOST model for simulating maize growth, leaf area index, biomass, grain yield, yield gap, and soil water under irrigation and rainfed conditions
Marinković et al. Data mining approach for predictive modeling of agricultural yield data

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