[go: up one dir, main page]
More Web Proxy on the site http://driver.im/

CN113159404A - Electric field seed treatment optimal parameter prediction method, system and terminal based on WOA-SVM - Google Patents

Electric field seed treatment optimal parameter prediction method, system and terminal based on WOA-SVM Download PDF

Info

Publication number
CN113159404A
CN113159404A CN202110390319.XA CN202110390319A CN113159404A CN 113159404 A CN113159404 A CN 113159404A CN 202110390319 A CN202110390319 A CN 202110390319A CN 113159404 A CN113159404 A CN 113159404A
Authority
CN
China
Prior art keywords
svm
data
parameter
electric field
model
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
CN202110390319.XA
Other languages
Chinese (zh)
Inventor
闫银发
马郡祥
李法德
宋占华
刘莫尘
田富洋
彭倩
孙相云
高甜甜
肖化超
杨宽宽
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.)
Shandong Agricultural University
Original Assignee
Shandong 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 Shandong Agricultural University filed Critical Shandong Agricultural University
Priority to CN202110390319.XA priority Critical patent/CN113159404A/en
Publication of CN113159404A publication Critical patent/CN113159404A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/02—Agriculture; Fishing; Forestry; Mining

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Economics (AREA)
  • General Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • Data Mining & Analysis (AREA)
  • General Business, Economics & Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Marketing (AREA)
  • Tourism & Hospitality (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Development Economics (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Game Theory and Decision Science (AREA)
  • Educational Administration (AREA)
  • Agronomy & Crop Science (AREA)
  • Animal Husbandry (AREA)
  • Marine Sciences & Fisheries (AREA)
  • Mining & Mineral Resources (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The application discloses a prediction method, a system and a terminal for optimal parameters of electric field seed treatment based on a WOA-SVM, which are used for determining electric field parameter indexes influencing seed vigor; carrying out data elimination and normalization processing on the data obtained by the test; establishing an SVM parameter model, acquiring actual data corresponding to each index, dividing the actual data into training data and testing data, and optimizing the SVM parameter model by using a WOA search algorithm; inputting the SVM parameters obtained after optimization into the SVM parameter model to obtain a trained SVM parameter model; and inputting the trained SVM parameter model into a prediction function of the LIBSVM to obtain a predicted value of the test sample. Starting from the aspect of influencing electric field parameters of seeds, an SVM-based electric field seed treatment key parameter regression prediction model is designed, the problem of poor universality of a traditional electric field seed treatment parameter experimental modeling method is solved, and therefore the prediction result of the electric field seed treatment optimal parameters can be improved.

Description

Electric field seed treatment optimal parameter prediction method, system and terminal based on WOA-SVM
Technical Field
The application relates to the technical field of high-voltage electric field seed treatment, in particular to an electric field seed treatment optimal parameter prediction method, system and terminal based on a WOA-SVM.
Background
The seeds are used as the most basic production data in agricultural production, are important factors influencing the yield and the quality of agricultural products, are generally regarded by people all the time, are treated before germination, are a crucial link in agricultural production, can sterilize and disinfect the seeds, simultaneously enhance the capability of the seeds for resisting diseases and insect pests in the stages of germination and seedling growth and development, and improve the survival rate of the seeds.
However, the conventional method for sterilizing and disinfecting seeds can improve the survival rate of the seeds, but causes environmental pollution. Therefore, the high-voltage electric field seed treatment technology is rapidly popularized and applied in recent years as a seed treatment technology which is rapid, efficient, low in cost, strong in controllability and free of pollution. Different electric field parameter combination conditions have different influences on the seed vitality, and the accurate prediction method is still important at present.
Due to the complexity of the biological electromagnetic effect and physiological mechanism of the crop seeds and the diversity of the electric field parameter conditions, it is difficult to establish a biological effect mathematical model of the electric field treated seeds by a mechanism analysis method, the optimal electric field parameter treatment conditions of the crop seeds are generally obtained by an experimental method modeling, the required sample number set is large, the universality is poor, and the industrial application is difficult to carry out.
Disclosure of Invention
In order to solve the technical problems, the following technical scheme is provided:
in a first aspect, an embodiment of the present application provides a method for predicting optimal parameters of electric field seed treatment based on a WOA-SVM, where the method includes: determining electric field parameter indexes influencing the seed vitality; carrying out data elimination and normalization processing on the data obtained by the test; establishing an SVM parameter model, acquiring actual data corresponding to each index, dividing the actual data into training data and testing data, and optimizing the SVM parameter model by using a WOA search algorithm; inputting the SVM parameters obtained after optimization into the SVM parameter model to obtain a trained SVM parameter model; and inputting the trained SVM parameter model into a prediction function of the LIBSVM to obtain a predicted value of the test sample.
By adopting the implementation mode, starting from the aspect of influencing the electric field parameters of the seeds, the regression prediction model of the electric field seed treatment key parameters based on the SVM is designed, the problem of poor universality of the traditional electric field seed treatment parameter experimental modeling method is solved, and therefore the prediction result of the electric field seed treatment optimal parameters can be improved.
With reference to the first aspect, in a first possible implementation manner of the first aspect, the determining an electric field parameter index affecting seed vigor includes: the power voltage, the pulse frequency and the action time are used as factors influencing the seed vigor, and the comprehensive germination index is used as the seed vigor index.
With reference to the first aspect, in a second possible implementation manner of the first aspect, the performing data elimination on data obtained through the test includes: eliminating abnormal data by utilizing Lauda criterion, and assuming the sample data set as X and the average value as
Figure BDA0003016482960000021
Sample deviation viThe sample standard deviation is sigma calculation formula:
Figure BDA0003016482960000022
Figure BDA0003016482960000023
if formula viIf > 3 σ, i ═ 1, 2.. n holds, then x is considered to be trueiAnd (4) rejecting abnormal data by repeatedly using the Lauda criterion until the abnormal data can be rejected.
With reference to the second possible implementation manner of the first aspect, in a third possible implementation manner of the first aspect, the normalizing the data obtained through the test includes: data normalization is based on the following formula:
Figure BDA0003016482960000024
in the formula: x is the test sample vector, XminIs the minimum value of the test sample vector, XmaxAnd X' is the maximum value of the test sample vector, and is the sample data vector after normalization processing.
With reference to the first aspect, in a fourth possible implementation manner of the first aspect, a mixed kernel function is established in the SVM parameter model, where the kernel function is obtained by linearly stacking a gaussian kernel and a polynomial kernel as basis kernels, and the gaussian kernel and the polynomial kernel are obtained by linearly stacking the basis kernels, and a mathematical expression is as follows:
Figure BDA0003016482960000031
kpolyis a polynomial kernel function, krbfIn the form of a gaussian kernel function,
Figure BDA0003016482960000032
is the weight value of the polynomial kernel.
With reference to the first aspect, in a fifth possible implementation manner of the first aspect, the inputting the SVM parameters obtained after the optimization into the SVM parameter model to obtain the trained SVM parameter model includes:
training a support vector regression model for prediction regression, wherein the training process comprises the following steps: calculating the average deviation and variance of sample data to obtain characteristic data, selecting the first h continuous characteristic data from the characteristic data according to the sample sequence to form a time sequence, and establishing a training data set with the row number of h-m +1 and the column number of m according to a preset mapping dimension m;
training a support vector regression SVM model by using a training data set, optimizing key parameters of the SVM by adopting a whale optimization algorithm, wherein the key parameters of the SVM comprise a penalty factor c of a kernel function, a parameter g of a Gaussian kernel function and a weight coefficient delta, taking SVM model parameters which enable the training data set to predict the highest correlation coefficient R as an optimal parameter combination of the SVM model to obtain the trained support vector regression SVM model, predicting electric field parameter combination conditions by the trained support vector regression SVM model, and the calculation formula of the correlation coefficient R is as follows:
Figure BDA0003016482960000033
xiand xi' true and predicted values, i ═ 1,2, …, n, respectively.
With reference to the fifth possible implementation manner of the first aspect, in a sixth possible implementation manner of the first aspect, the optimizing key parameters of the SVM by using a whale optimization algorithm includes: reading in search conditions, and initializing whale swarm positions and parameters of a whale optimization algorithm; calculating a parameter value of an initial position of the whale colony, and recording a current optimal solution; calculating a convergence factor and a swing factor; updating the space position of the whale colony; calculating the fitness value corresponding to the updated whale colony position vector; comparing the fitness values corresponding to the whale colony position vectors before and after updating, and determining the position of the whale colony of the next generation; judging whether preset conditions are met, and if so, outputting the corresponding position and the corresponding parameter value of the optimal whale colony individual; or, if not, recalculating the convergence factor and the swing factor until the preset condition is met.
With reference to the first aspect, in a seventh possible implementation manner of the first aspect, the method further includes: inverse normalization of the predicted value of the test sample, drawing and comparing analysisTesting the real value and the predicted value of the sample to obtain a prediction result of the electric field parameters, wherein the prediction result comprises the following steps: the data was denormalised based on the following formula: x ═ X' (X)max-Xmin)+XminIn the formula: xminIs the minimum value of the test sample vector; xmaxIs the maximum value of the test sample vector; x 'is the sample data vector after normalization processing, and X' is the output data vector after reverse normalization.
In a second aspect, the embodiment of the present application provides a WOA-SVM-based electric field seed treatment optimal parameter prediction system, where the system includes: the determining module is used for determining electric field parameter indexes influencing the seed vitality; the data processing module is used for carrying out data elimination and normalization processing on the data obtained through the test; the model establishing module is used for establishing an SVM parameter model, acquiring actual data corresponding to each index, dividing the actual data into training data and testing data, and optimizing the SVM parameter model by using a WOA search algorithm; the model training module is used for inputting the SVM parameters obtained after optimization into the SVM parameter model to obtain a trained SVM parameter model; and the predicted value acquisition module is used for inputting the trained SVM parameter model into a prediction function of the LIBSVM to obtain the predicted value of the test sample.
In a third aspect, an embodiment of the present application provides a terminal, including: a processor; a memory for storing processor executable instructions; the processor executes the electric field seed processing optimal parameter prediction method based on the WOA-SVM described in the first aspect or any implementation manner of the first aspect, and predicts the electric field seed processing optimal parameter.
Drawings
FIG. 1 is a schematic flow chart of a method for predicting optimal parameters of electric field seed treatment based on WOA-SVM according to an embodiment of the present disclosure;
FIG. 2 is a schematic diagram illustrating comparison between real values and predicted values of a training set according to an embodiment of the present disclosure;
fig. 3 is a schematic diagram illustrating comparison between results of real values and predicted values of a prediction set provided in an embodiment of the present application;
FIG. 4 is a schematic diagram of an optimal parameter prediction system for electric field seed treatment based on WOA-SVM according to an embodiment of the present application;
fig. 5 is a schematic diagram of a terminal according to an embodiment of the present application.
Detailed Description
The present invention will be described with reference to the accompanying drawings and embodiments.
Fig. 1 is a schematic flow chart of a method for predicting optimal parameters of electric field seed treatment based on a WOA-SVM according to an embodiment of the present application, and referring to fig. 1, the method for predicting optimal parameters of electric field seed treatment based on a WOA-SVM according to the present embodiment includes:
s101, determining electric field parameter indexes influencing the seed vitality.
In the embodiment, the power voltage, the pulse frequency and the action time are used as factors influencing the seed vigor, and the comprehensive germination index is used as the seed vigor index.
And S102, carrying out data elimination and normalization processing on the data obtained through the test.
The data elimination of the data obtained by the test comprises the following steps: eliminating abnormal data by utilizing Lauda criterion, and assuming the sample data set as X and the average value as
Figure BDA0003016482960000051
Sample deviation viThe sample standard deviation is sigma calculation formula:
Figure BDA0003016482960000052
if formula viIf > 3 σ, i ═ 1, 2.. n holds, then x is considered to be trueiAnd (4) rejecting abnormal data by repeatedly using the Lauda criterion until the abnormal data can be rejected.
The normalization processing of the data obtained by the test comprises the following steps: data normalization is based on the following formula:
Figure BDA0003016482960000053
in the formula: x is the test sample vector, XminIs the minimum value of the test sample vector, XmaxIs the maximum value of the test sample vector, and X' is the normalization processAnd the subsequent sample data vector.
S103, establishing an SVM parameter model, collecting actual data corresponding to each index, dividing the actual data into training data and testing data, and optimizing the SVM parameter model by using a WOA search algorithm.
The kernel function in the SVM parameter model is established by selecting a Gaussian kernel and a polynomial kernel as mixed kernel functions after the linear superposition of the basic kernels, and the Gaussian kernel and the polynomial kernel as mixed kernel function support vector machines after the linear superposition of the basic kernels, wherein the mathematical expression is as follows:
Figure BDA0003016482960000061
kpolyis a polynomial kernel function, krbfIn the form of a gaussian kernel function,
Figure BDA0003016482960000062
is the weight value of the polynomial kernel.
And S104, inputting the SVM parameters obtained after optimization into the SVM parameter model to obtain a trained SVM parameter model.
Training a support vector regression model for prediction regression, wherein the training process comprises the following steps: calculating the average deviation and variance of sample data to obtain characteristic data, selecting the first h continuous characteristic data from the characteristic data according to the sample sequence to form a time sequence, and establishing a training data set with the row number of h-m +1 and the column number of m according to a preset mapping dimension m;
training a support vector regression SVM model by using a training data set, optimizing key parameters of the SVM by adopting a whale optimization algorithm, wherein the key parameters of the SVM comprise a penalty factor c of a kernel function, a parameter g of a Gaussian kernel function and a weight coefficient delta, taking SVM model parameters which enable the training data set to predict the highest correlation coefficient R as an optimal parameter combination of the SVM model to obtain the trained support vector regression SVM model, predicting electric field parameter combination conditions by the trained support vector regression SVM model, and the calculation formula of the correlation coefficient R is as follows:
Figure BDA0003016482960000063
xiand xi' real and predicted values are respectively,i＝1,2,…,n。
specifically, optimizing key parameters of the SVM by using a whale optimization algorithm comprises the following steps: reading in search conditions, and initializing whale swarm positions and parameters of a whale optimization algorithm; calculating a parameter value of an initial position of the whale colony, and recording a current optimal solution; calculating a convergence factor and a swing factor; updating the space position of the whale colony; calculating the fitness value corresponding to the updated whale colony position vector; comparing the fitness values corresponding to the whale colony position vectors before and after updating, and determining the position of the whale colony of the next generation; judging whether preset conditions are met, and if so, outputting the corresponding position and the corresponding parameter value of the optimal whale colony individual; or, if not, recalculating the convergence factor and the swing factor until the preset condition is met.
And S105, inputting the trained SVM parameter model into a prediction function of the LIBSVM to obtain a predicted value of the test sample.
The embodiment further includes inverse normalization of the predicted value of the test sample, and drawing, comparing and analyzing the true value and the predicted value of the test sample to obtain the electric field parameter prediction result, including: the data was denormalised based on the following formula: x ═ X' (X)max-Xmin)+XminIn the formula: xminIs the minimum value of the test sample vector; xmaxIs the maximum value of the test sample vector; x 'is the sample data vector after normalization processing, and X' is the output data vector after reverse normalization.
The present embodiment performs electric field treatment on aged cotton seeds under different combination conditions of three electric field parameters of different power supply voltages, pulse frequencies and action times, performs germination experiments, and collects experimental data. The model establishment and electric field parameter prediction process comprises the following steps:
and selecting a part of data as training data, establishing prediction models with different precisions by adopting different kernel functions, and then using the prediction model with higher precision as a method capable of predicting the optimal electric field parameter.
And respectively taking the power supply voltage, the pulse frequency and the action time as input characteristic values, and taking the comprehensive activity index as an output characteristic value. The first 20 sets of data were used for model training and the last 5 sets of data were used to test model accuracy.
And the function mapminmax is used for carrying out normalization processing on the data, so that the influence on the accuracy of the model caused by too large or too small range of the characteristic value is prevented.
Selecting the type of the SVM as epsilon-SVR, selecting a mixed kernel function with a Gaussian kernel and a polynomial kernel as basis kernels and linearly superposed as an inner product kernel function of the model, wherein the formula of the mixed kernel function is as follows:
Figure BDA0003016482960000081
optimizing the parameters by using a whale optimization algorithm to find a group of optimal parameters (c, g and delta) so as to minimize MES (mean square error).
The model effect under the conditions that the parameters c are 169.3455, g are 0.022933 and delta is 0.30 is good, and the best parameters (c, g and delta) are introduced into the libsvm algorithm for training to obtain the trained model.
And inputting the trained model into a prediction function of libsvm to obtain a prediction value of the test sample.
And (4) reverse normalization is carried out on the predicted value of the test sample, and the real value and the predicted value of the test sample are drawn, compared and analyzed.
Figure 2 shows the comparison of the results of the true versus predicted values for the training set.
Fig. 3 shows the comparison of the results of the real values and predicted values of the prediction set.
The accuracy of the model under prediction by using the test set data is shown in table 1, and it can be seen from table 1 that the accuracy of the model under prediction by using the test set data under the mixed kernel function has a correlation coefficient of 81.8874% between the predicted value and the actual value and a mean square error of 0.018350, and the accuracy is acceptable in prediction.
TABLE 1 analytical table of values of prediction test data
Figure BDA0003016482960000082
In the embodiment, the accuracy of the model is checked by using the data of the test set, and the accuracy of the model is checked by using the data of the training set, so that the data prediction result of the training set is better because the data change fluctuation of the training set is not large; the data fluctuation of the prediction set is large, so the prediction result is inferior to that of the test set, but the precision also meets the requirement.
Corresponding to the prediction method for optimal parameters of electric field seed treatment based on the WOA-SVM provided in the foregoing embodiments, the present application also provides an embodiment of a prediction system for optimal parameters of electric field seed treatment based on the WOA-SVM, and referring to fig. 4, the prediction system 20 for optimal parameters of electric field seed treatment based on the WOA-SVM includes: a determination module 201, a data processing module 202, a model building module 203, a model training module 204, and a predicted value acquisition module 205.
The determining module 201 is configured to determine an electric field parameter index affecting seed vigor.
The power voltage, the pulse frequency and the action time are used as factors influencing the seed vigor, and the comprehensive germination index is used as the seed vigor index.
The data processing module 202 is configured to perform data elimination and normalization processing on the data obtained through the test.
Eliminating abnormal data by utilizing Lauda criterion, and assuming the sample data set as X and the average value as
Figure BDA0003016482960000091
Sample deviation viThe sample standard deviation is sigma calculation formula:
Figure BDA0003016482960000092
if formula viIf > 3 σ, i ═ 1, 2.. n holds, then x is considered to be trueiAnd (4) rejecting abnormal data by repeatedly using the Lauda criterion until the abnormal data can be rejected. Data normalization is based on the following formula:
Figure BDA0003016482960000093
in the formula: x is the test sample vector, XminIs the minimum value of the test sample vector, XmaxAnd X' is the maximum value of the test sample vector, and is the sample data vector after normalization processing.
The model establishing module 203 is used for establishing an SVM parameter model, collecting actual data corresponding to each index, dividing the actual data into training data and testing data, and optimizing the SVM parameter model by using a WOA search algorithm.
The kernel function in the SVM parameter model is established by selecting a Gaussian kernel and a polynomial kernel as mixed kernel functions after the linear superposition of the basic kernels, and the Gaussian kernel and the polynomial kernel as mixed kernel function support vector machines after the linear superposition of the basic kernels, wherein the mathematical expression is as follows:
Figure BDA0003016482960000094
kpolyis a polynomial kernel function, krbfIn the form of a gaussian kernel function,
Figure BDA0003016482960000095
is the weight value of the polynomial kernel.
The model training module 204 is configured to input the SVM parameters obtained after the optimization into the SVM parameter model to obtain a trained SVM parameter model.
Training a support vector regression model for prediction regression, wherein the training process comprises the following steps: calculating the average deviation and variance of sample data to obtain characteristic data, selecting the first h continuous characteristic data from the characteristic data according to the sample sequence to form a time sequence, and establishing a training data set with the row number of h-m +1 and the column number of m according to a preset mapping dimension m;
training a support vector regression SVM model by using a training data set, optimizing key parameters of the SVM by adopting a whale optimization algorithm, wherein the key parameters of the SVM comprise a penalty factor c of a kernel function, a parameter g of a Gaussian kernel function and a weight coefficient delta, taking SVM model parameters which enable the training data set to predict the highest correlation coefficient R as an optimal parameter combination of the SVM model to obtain the trained support vector regression SVM model, predicting electric field parameter combination conditions by the trained support vector regression SVM model, and the calculation formula of the correlation coefficient R is as follows:
Figure BDA0003016482960000101
xiand xi' true and predicted values, i ═ 1,2, …, n, respectively.
The method for optimizing the key parameters of the SVM by adopting the whale optimization algorithm comprises the following steps: reading in search conditions, and initializing whale swarm positions and parameters of a whale optimization algorithm; calculating a parameter value of an initial position of the whale colony, and recording a current optimal solution; calculating a convergence factor and a swing factor; updating the space position of the whale colony; calculating the fitness value corresponding to the updated whale colony position vector; comparing the fitness values corresponding to the whale colony position vectors before and after updating, and determining the position of the whale colony of the next generation; judging whether preset conditions are met, and if so, outputting the corresponding position and the corresponding parameter value of the optimal whale colony individual; or, if not, recalculating the convergence factor and the swing factor until the preset condition is met.
The predicted value obtaining module 205 is configured to input the trained SVM parameter model into a prediction function of LIBSVM to obtain a predicted value of the test sample.
This embodiment still includes: the method comprises the following steps of performing inverse normalization on the predicted value of the test sample, drawing, comparing and analyzing the true value and the predicted value of the test sample, and obtaining the electric field parameter prediction result, wherein the electric field parameter prediction result comprises the following steps: the data was denormalised based on the following formula: x ═ X' (X)max-Xmin)+XminIn the formula: xminIs the minimum value of the test sample vector; xmaxIs the maximum value of the test sample vector; x 'is the sample data vector after normalization processing, and X' is the output data vector after reverse normalization.
The embodiment of the present application also provides an embodiment of a terminal, and referring to fig. 5, the terminal 30 includes a processor 301, a memory 302, and a communication interface 303.
In fig. 5, the processor 301, the memory 302, and the communication interface 303 may be connected to each other by a bus; the bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown in FIG. 5, but this is not intended to represent only one bus or type of bus.
The processor 301 generally controls the overall functions of the terminal 30, such as the start-up of the terminal 30 and the determination of electric field parameter indexes affecting the vitality of the seeds after the terminal 30 is started up; carrying out data elimination and normalization processing on the data obtained by the test; establishing an SVM parameter model, acquiring actual data corresponding to each index, dividing the actual data into training data and testing data, and optimizing the SVM parameter model by using a WOA search algorithm; inputting the SVM parameters obtained after optimization into the SVM parameter model to obtain a trained SVM parameter model; and inputting the trained SVM parameter model into a prediction function of the LIBSVM to obtain a predicted value of the test sample.
The processor 301 may be a general-purpose processor such as a Central Processing Unit (CPU), a Network Processor (NP), or a combination of a CPU and an NP. The processor may also be a Microprocessor (MCU). The processor may also include a hardware chip. The hardware chips may be Application Specific Integrated Circuits (ASICs), Programmable Logic Devices (PLDs), or a combination thereof. The PLD may be a Complex Programmable Logic Device (CPLD), a Field Programmable Gate Array (FPGA), or the like.
The memory 302 is configured to store computer-executable instructions to support the operation of the terminal 30 data. The memory 301 may be implemented by any type or combination of volatile or non-volatile memory devices, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks.
After the terminal 30 is started, the processor 301 and the memory 302 are powered on, and the processor 301 reads and executes the computer executable instructions stored in the memory 302 to complete all or part of the steps in the above-mentioned WOA-SVM-based electric field seed processing optimal parameter prediction method embodiment.
The communication interface 303 is used for transmitting data to the terminal 30, and for example, realizes communication with an electric field seed processor. The communication interface 303 includes a wired communication interface, and may also include a wireless communication interface. The wired communication interface comprises a USB interface, a Micro USB interface and an Ethernet interface. The wireless communication interface may be a WLAN interface, a cellular network communication interface, a combination thereof, or the like.
In an exemplary embodiment, the terminal 30 provided by the embodiments of the present application further includes a power supply component that provides power to the various components of the terminal 30. The power components may include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power for the terminal 30.
A communications component configured to facilitate communications between the terminal 30 and other devices in a wired or wireless manner. The terminal 30 may access a wireless network based on a communication standard, such as WiFi, 4G or 5G, or a combination thereof. The communication component receives a broadcast signal or broadcast related information from an external broadcast management system via a broadcast channel. The communication component also includes a Near Field Communication (NFC) module to facilitate short-range communications. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, Ultra Wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.
In an exemplary embodiment, the terminal 30 may be implemented by one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), or other electronic components.
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 identical elements in a process, method, article, or apparatus that comprises the element.
Of course, the above description is not limited to the above examples, and technical features that are not described in this application may be implemented by or using the prior art, and are not described herein again; the above embodiments and drawings are only for illustrating the technical solutions of the present application and not for limiting the present application, and the present application is only described in detail with reference to the preferred embodiments instead, it should be understood by those skilled in the art that changes, modifications, additions or substitutions within the spirit and scope of the present application may be made by those skilled in the art without departing from the spirit of the present application, and the scope of the claims of the present application should also be covered.

Claims (10)

1. An electric field seed treatment optimal parameter prediction method based on WOA-SVM is characterized by comprising the following steps:
determining electric field parameter indexes influencing the seed vitality;
carrying out data elimination and normalization processing on the data obtained by the test;
establishing an SVM parameter model, acquiring actual data corresponding to each index, dividing the actual data into training data and testing data, and optimizing the SVM parameter model by using a WOA search algorithm;
inputting the SVM parameters obtained after optimization into the SVM parameter model to obtain a trained SVM parameter model;
and inputting the trained SVM parameter model into a prediction function of the LIBSVM to obtain a predicted value of the test sample.
2. The method of claim 1, wherein determining an electric field parameter indicative of an effect on seed vigor comprises: the power voltage, the pulse frequency and the action time are used as factors influencing the seed vigor, and the comprehensive germination index is used as the seed vigor index.
3. The method of claim 1, wherein the data culling the trial data comprises: eliminating abnormal data by utilizing Lauda criterion, and assuming the sample data set as X and the average value as
Figure FDA0003016482950000011
Sample deviation viThe sample standard deviation is sigma calculation formula:
Figure FDA0003016482950000012
Figure FDA0003016482950000013
if formula viIf > 3 σ, i ═ 1, 2.. n holds, then x is considered to be trueiAnd (4) rejecting abnormal data by repeatedly using the Lauda criterion until the abnormal data can be rejected.
4. The method of claim 3, wherein normalizing the data from the experiment comprises: data normalization is based on the following formula:
Figure FDA0003016482950000014
in the formula: x is the test sample vector, XminIs the minimum value of the test sample vector, XmaxAnd X' is the maximum value of the test sample vector, and is the sample data vector after normalization processing.
5. The method of claim 1, wherein a mixed kernel function with a gaussian kernel and a polynomial kernel as basis kernels after linear superposition is selected as a kernel function in the SVM parameter model establishment, and a mixed kernel function support vector machine with a gaussian kernel and a polynomial kernel as basis kernels after linear superposition is provided, wherein a mathematical expression is as follows:
Figure FDA0003016482950000021
kpolyis a polynomial kernel function, krbfIn the form of a gaussian kernel function,
Figure FDA0003016482950000022
is the weight value of the polynomial kernel.
6. The method of claim 1, wherein the step of inputting the SVM parameter obtained after the optimization into the SVM parameter model to obtain a trained SVM parameter model comprises:
training a support vector regression model for prediction regression, wherein the training process comprises the following steps: calculating the average deviation and variance of sample data to obtain characteristic data, selecting the first h continuous characteristic data from the characteristic data according to the sample sequence to form a time sequence, and establishing a training data set with the row number of h-m +1 and the column number of m according to a preset mapping dimension m;
training a support vector regression SVM model by using a training data set, optimizing key parameters of the SVM by adopting a whale optimization algorithm, wherein the key parameters of the SVM comprise a penalty factor c of a kernel function, a parameter g of a Gaussian kernel function and a weight coefficient delta, taking SVM model parameters which enable the training data set to predict the highest correlation coefficient R as an optimal parameter combination of the SVM model to obtain the trained support vector regression SVM model, predicting electric field parameter combination conditions by the trained support vector regression SVM model, and the calculation formula of the correlation coefficient R is as follows:
Figure FDA0003016482950000023
xiand xi' true and predicted values, i ═ 1,2, …, n, respectively.
7. The method of claim 6, wherein the optimizing key parameters of the SVM using whale optimization algorithm comprises:
reading in search conditions, and initializing whale swarm positions and parameters of a whale optimization algorithm;
calculating a parameter value of an initial position of the whale colony, and recording a current optimal solution;
calculating a convergence factor and a swing factor;
updating the space position of the whale colony;
calculating the fitness value corresponding to the updated whale colony position vector;
comparing the fitness values corresponding to the whale colony position vectors before and after updating, and determining the position of the whale colony of the next generation;
judging whether preset conditions are met, and if so, outputting the corresponding position and the corresponding parameter value of the optimal whale colony individual; or, if not, recalculating the convergence factor and the swing factor until the preset condition is met.
8. The method of claim 1, further comprising: the method comprises the following steps of performing inverse normalization on the predicted value of the test sample, drawing, comparing and analyzing the true value and the predicted value of the test sample, and obtaining the electric field parameter prediction result, wherein the electric field parameter prediction result comprises the following steps: the data was denormalised based on the following formula: x ═ X' (X)max-Xmin)+XminIn the formula: xminIs the minimum value of the test sample vector; xmaxIs the maximum value of the test sample vector; x 'is the sample data vector after normalization processing, and X' is the output data vector after reverse normalization.
9. An electric field seed treatment optimal parameter prediction system based on WOA-SVM, characterized in that the system comprises:
the determining module is used for determining electric field parameter indexes influencing the seed vitality;
the data processing module is used for carrying out data elimination and normalization processing on the data obtained through the test;
the model establishing module is used for establishing an SVM parameter model, acquiring actual data corresponding to each index, dividing the actual data into training data and testing data, and optimizing the SVM parameter model by using a WOA search algorithm;
the model training module is used for inputting the SVM parameters obtained after optimization into the SVM parameter model to obtain a trained SVM parameter model;
and the predicted value acquisition module is used for inputting the trained SVM parameter model into a prediction function of the LIBSVM to obtain the predicted value of the test sample.
10. A terminal, comprising:
a processor;
a memory for storing processor executable instructions;
the processor executes the electric field seed treatment optimal parameter prediction method based on WOA-SVM of any one of claims 1-8 to predict the electric field seed treatment optimal parameter.
CN202110390319.XA 2021-04-12 2021-04-12 Electric field seed treatment optimal parameter prediction method, system and terminal based on WOA-SVM Pending CN113159404A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110390319.XA CN113159404A (en) 2021-04-12 2021-04-12 Electric field seed treatment optimal parameter prediction method, system and terminal based on WOA-SVM

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110390319.XA CN113159404A (en) 2021-04-12 2021-04-12 Electric field seed treatment optimal parameter prediction method, system and terminal based on WOA-SVM

Publications (1)

Publication Number Publication Date
CN113159404A true CN113159404A (en) 2021-07-23

Family

ID=76889997

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110390319.XA Pending CN113159404A (en) 2021-04-12 2021-04-12 Electric field seed treatment optimal parameter prediction method, system and terminal based on WOA-SVM

Country Status (1)

Country Link
CN (1) CN113159404A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113807598A (en) * 2021-09-26 2021-12-17 辽宁石油化工大学 SVM heavy oil pipeline corrosion rate prediction method based on PSO-WOA hybrid optimization
CN115356647A (en) * 2022-08-29 2022-11-18 湖北工程学院 Method for early predicting service life of lithium ion battery

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102539597A (en) * 2011-05-24 2012-07-04 四川省中医药科学院 Method for quickly identifying notopterygium incisum seed and notopterygium franchetii seed
CN108303134A (en) * 2017-12-22 2018-07-20 中国农业大学 The detecting system of environmental parameter in a kind of Seed Germination
CN108615555A (en) * 2018-04-26 2018-10-02 山东师范大学 Colorectal cancer prediction technique and device based on marker gene and mixed kernel function SVM
CN110516831A (en) * 2019-06-18 2019-11-29 国网（北京）节能设计研究院有限公司 A kind of short-term load forecasting method based on MWOA algorithm optimization SVM
CN110771304A (en) * 2019-11-04 2020-02-11 哈尔滨金田农业科技有限公司 Automatic control device for electrostatic field polarization crop seeds

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102539597A (en) * 2011-05-24 2012-07-04 四川省中医药科学院 Method for quickly identifying notopterygium incisum seed and notopterygium franchetii seed
CN108303134A (en) * 2017-12-22 2018-07-20 中国农业大学 The detecting system of environmental parameter in a kind of Seed Germination
CN108615555A (en) * 2018-04-26 2018-10-02 山东师范大学 Colorectal cancer prediction technique and device based on marker gene and mixed kernel function SVM
CN110516831A (en) * 2019-06-18 2019-11-29 国网（北京）节能设计研究院有限公司 A kind of short-term load forecasting method based on MWOA algorithm optimization SVM
CN110771304A (en) * 2019-11-04 2020-02-11 哈尔滨金田农业科技有限公司 Automatic control device for electrostatic field polarization crop seeds

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
赵丹丹: ""基于特征选择的结直肠癌预测模型研究"", 《中国优秀博硕士学位论文全文数据库(硕士) 信息科技辑》 *
闫银发 等: ""极低频高压脉冲电场提高陈年棉种活力的参数优化"", 《农业工程学报》 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113807598A (en) * 2021-09-26 2021-12-17 辽宁石油化工大学 SVM heavy oil pipeline corrosion rate prediction method based on PSO-WOA hybrid optimization
CN115356647A (en) * 2022-08-29 2022-11-18 湖北工程学院 Method for early predicting service life of lithium ion battery

Similar Documents

Publication Publication Date Title
Wu et al. Efficient hyperparameter optimization through model-based reinforcement learning
CN113159404A (en) Electric field seed treatment optimal parameter prediction method, system and terminal based on WOA-SVM
Müssel et al. Multi-objective parameter selection for classifiers
Demertzis et al. Machine learning use in predicting interior spruce wood density utilizing progeny test information
US11485999B2 (en) Metagenomics for microbiomes
Filatovas et al. A preference-based multi-objective evolutionary algorithm R-NSGA-II with stochastic local search
CN115689008A (en) CNN-BilSTM short-term photovoltaic power prediction method and system based on ensemble empirical mode decomposition
JPWO2019229931A1 (en) Information processing equipment, control methods, and programs
Lin et al. Dynamic risk assessment of food safety based on an improved hidden Markov model integrating cuckoo search algorithm: A sterilized milk study
Jia Methods for handling missing non-normal data in structural equation modeling
CN113408374B (en) Yield estimation method, device, equipment and storage medium based on artificial intelligence
CN114298265A (en) Radio frequency circuit diagnosis model training and diagnosis method, device, medium and terminal
Chrisman et al. Incorporating biological knowledge into evaluation of causal regulatory hypotheses
CN116484193A (en) Crop yield prediction method, system, equipment and medium
Yılmaz et al. A CNN-based hybrid approach to classification of raisin grains
Hanafi et al. Performance Evaluation of Rice Seed Classification System Based on The CNN with VGG-GoogleNet Architecture
CN107067017A (en) The depth of burn forecasting system of near infrared spectrum based on CAGA and SVM
CN115130620A (en) Power consumption mode identification model generation method and device for power equipment
Antonelli et al. A new approach to handle high dimensional and large datasets in multi-objective evolutionary fuzzy systems
Tsakiridis et al. A fuzzy rule-based system utilizing differential evolution with an application in vis-NIR soil spectroscopy
Dhebar et al. Evaluating Nonlinear Decision Trees for Binary Classification Tasks with Other Existing Methods
Dong et al. A PCA-SMO Based Hybrid Classification Model for Predictions in Precision Agriculture
Kent et al. Tissue discrimination from impedance spectroscopy as a multi-objective optimisation problem with weighted naive bayes classification
Wang An incremental learning strategy for support vector regression
Dwivedy et al. Soil Testing using IOT and Machine Learning Classifiers

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
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20210723