CN113837474A - Regional soil heavy metal pollution index prediction method and device - Google Patents
Regional soil heavy metal pollution index prediction method and device Download PDFInfo
- Publication number
- CN113837474A CN113837474A CN202111135431.5A CN202111135431A CN113837474A CN 113837474 A CN113837474 A CN 113837474A CN 202111135431 A CN202111135431 A CN 202111135431A CN 113837474 A CN113837474 A CN 113837474A
- Authority
- CN
- China
- Prior art keywords
- heavy metal
- soil
- pollution
- support vector
- vector regression
- 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
Links
- 229910001385 heavy metal Inorganic materials 0.000 title claims abstract description 107
- 239000002689 soil Substances 0.000 title claims abstract description 88
- 238000000034 method Methods 0.000 title claims abstract description 53
- 230000002068 genetic effect Effects 0.000 claims description 55
- 238000004422 calculation algorithm Methods 0.000 claims description 45
- 238000007637 random forest analysis Methods 0.000 claims description 15
- 230000035772 mutation Effects 0.000 claims description 9
- 238000013145 classification model Methods 0.000 claims description 6
- 238000010353 genetic engineering Methods 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 abstract description 3
- 230000006870 function Effects 0.000 description 90
- 238000011109 contamination Methods 0.000 description 12
- 238000010586 diagram Methods 0.000 description 8
- 108090000623 proteins and genes Proteins 0.000 description 6
- 210000000349 chromosome Anatomy 0.000 description 4
- 238000005070 sampling Methods 0.000 description 3
- 238000000605 extraction Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000003912 environmental pollution Methods 0.000 description 1
- 210000003936 merozoite Anatomy 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000010076 replication Effects 0.000 description 1
- 238000010845 search algorithm Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000003900 soil pollution Methods 0.000 description 1
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
- 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
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06311—Scheduling, planning or task assignment for a person or group
-
- 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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
- G06Q50/265—Personal security, identity or safety
Abstract
The disclosure relates to a regional soil heavy metal pollution index prediction method and a regional soil heavy metal pollution index prediction device. The regional soil heavy metal pollution index prediction method comprises the following steps: acquiring heavy metal content data of a soil sample; determining the ordering of the contribution degrees of various heavy metal elements of the soil sample under different pollution states based on the heavy metal content data; and predicting the Mero comprehensive pollution index in the soil by taking the heavy metal elements with the predetermined number of contribution degrees ranked in the top as input parameters of a support vector regression model. By adopting the method, the appearance conditions of various heavy metal elements in the soil can be obtained through calculation, so that different treatment schemes can be formulated according to the contribution degrees of different heavy metal elements; in addition, the method can predict the heavy metal pollution condition of the soil through small sample data, so that the influence of data loss or insufficient data on pollution prediction is effectively avoided.
Description
Technical Field
The invention relates to the field of environmental pollution prediction, in particular to a method and a device for predicting a heavy metal pollution index of regional soil.
Background
The time period of the heavy metal element contaminated soil is long, the feasibility of repeatedly sampling and measuring the content of heavy metal in the soil and evaluating the heavy metal contaminated state of the soil is low, and therefore a relevant prediction model needs to be established to predict the comprehensive Mello pollution index in the soil heavy metal.
Disclosure of Invention
In order to solve the problems, the invention provides a method and a device for predicting a heavy metal pollution index of regional soil.
According to the present invention, there is provided a regional soil heavy metal pollution index prediction method, which may include: acquiring heavy metal content data of a soil sample; determining the ordering of the contribution degrees of various heavy metal elements of the soil sample under different pollution states based on the heavy metal content data; and predicting the Mero comprehensive pollution index in the soil by taking the heavy metal elements with the predetermined number of contribution degrees ranked in the top as input parameters of a support vector regression model.
According to one embodiment of the present invention, the contamination status may include a safe status, a still safe status, a light contamination status, a moderate contamination status, and a heavy contamination status.
According to an embodiment of the present invention, determining the ranking of the contribution degrees of the heavy metal elements of the soil sample under different pollution states may include: and taking the heavy metal content data and the pollution state as a random forest algorithm training sample set, and constructing a random forest algorithm pollution classification model to sort the contribution degrees of various heavy metal elements in different pollution states.
According to an embodiment of the present invention, the method for predicting the heavy metal pollution index of regional soil may further include: and optimizing the penalty function and the kernel function parameter of the support vector regression model by using a genetic algorithm.
According to one embodiment of the present invention, optimizing penalty functions and kernel function parameters of a support vector regression model using a genetic algorithm may include: setting the value ranges of the penalty function and the kernel function parameter; encoding a set of randomly generated support vector regression parameters; determining the fitness function value of each individual in the population; carrying out genetic operation on the current population until the current population meets the convergence condition; determining the obtained penalty function and kernel function parameters as the optimal penalty function and kernel function parameters of the support vector regression model so as to establish a genetic algorithm-support vector regression model; and determining whether the genetic algorithm-support vector regression model reaches the preset precision, if so, outputting the genetic algorithm-support vector regression model, and otherwise, resetting the value ranges of the penalty function and the kernel function parameters and reestablishing the genetic algorithm-support vector regression model.
According to one embodiment of the present invention, the magnitude of the fitness function value is determined by calculating an error function value, which is inversely proportional to the magnitude of the fitness function value, and the individual with the large fitness function value is inherited to the next generation.
According to one embodiment of the invention, the genetic manipulation may comprise at least one of selection, crossover and mutation.
According to the present invention, there is provided a regional soil heavy metal pollution index prediction apparatus, which may include: the data acquisition unit is used for acquiring the heavy metal content data of the soil sample; the sorting unit is used for determining the sorting of the contribution degrees of various heavy metal elements of the soil sample in different pollution states based on the heavy metal content data; and the prediction unit is used for predicting the comprehensive pollution index of the Mello in the soil by taking the heavy metal elements with the predetermined number of contribution degrees ranked in the top as input parameters of the support vector regression model.
According to one embodiment of the present invention, the contamination status may include a safe status, a still safe status, a light contamination status, a moderate contamination status, and a heavy contamination status.
According to an embodiment of the invention, the sorting unit may be configured to: and taking the heavy metal content data and the pollution state as a random forest algorithm training sample set, and constructing a random forest algorithm pollution classification model to sort the contribution degrees of various heavy metal elements in different pollution states.
According to an embodiment of the present invention, the regional soil heavy metal pollution index prediction device may further include: and the optimizing unit is used for optimizing the penalty function and the kernel function parameter of the support vector regression model by utilizing a genetic algorithm.
According to an embodiment of the invention, the optimizing unit may be configured to: setting the value ranges of the penalty function and the kernel function parameter; encoding a set of randomly generated support vector regression parameters; determining fitness function values of each individual in the population; carrying out genetic operation on the current population until the current population meets the convergence condition; determining the obtained penalty function and kernel function parameters as the optimal penalty function and kernel function parameters of the support vector regression model so as to establish a genetic algorithm-support vector regression model; and determining whether the genetic algorithm-support vector regression model reaches the preset precision, outputting the genetic algorithm-support vector regression model if the genetic algorithm-support vector regression model reaches the preset precision, and resetting the value ranges of the penalty function and the kernel function parameter and reestablishing the genetic algorithm-support vector regression model if the genetic algorithm-support vector regression model does not reach the preset precision.
According to an embodiment of the present invention, the optimizing unit may determine a magnitude of the fitness function value by calculating an error function value, and may pass the individual having the large fitness function value to the next generation, wherein the magnitude of the error function value is inversely proportional to the magnitude of the fitness function value.
According to one embodiment of the invention, the genetic manipulation may comprise at least one of selection, crossover and mutation.
According to the present invention, there is provided a computer-readable storage medium having stored thereon a computer program for executing the regional soil heavy metal pollution index prediction method according to any one of the foregoing embodiments.
According to the invention, a computing device is provided, which comprises a storage component and a processor, wherein the storage component stores computer-executable instructions, and when the computer-executable instructions are executed by the processor, the regional soil heavy metal pollution index prediction method of any one of the preceding embodiments is executed.
By adopting the method, the expression conditions of various heavy metal elements in the soil can be obtained through calculation, so that different treatment schemes can be formulated according to the contribution degrees of different heavy metal elements; in addition, the method can predict the heavy metal pollution condition of the soil through small sample data, so that the influence of data loss or insufficient data on pollution prediction is effectively avoided.
Drawings
The above and/or other objects and advantages of the present invention will become more apparent by describing embodiments below with reference to the accompanying drawings, in which:
FIG. 1 is a flow chart of a regional soil heavy metal pollution index prediction method according to an embodiment of the invention;
FIG. 2 is a block diagram of a random forest-genetic algorithm-support vector regression model according to an embodiment of the invention;
FIG. 3 is a flow diagram for optimizing penalty function and kernel function parameters of a support vector regression model using a genetic algorithm, according to an embodiment of the present invention;
fig. 4 is a block diagram illustrating a structure of a regional soil heavy metal pollution index prediction apparatus according to an embodiment of the present invention; and
FIG. 5 shows a schematic structural diagram of a computing device according to an embodiment of the invention.
Detailed Description
As required, detailed embodiments of the present invention are disclosed herein; however, it is to be understood that the disclosed embodiments are merely exemplary of the invention that may be embodied in various and alternative forms. The figures are not necessarily to scale; some features may be exaggerated or minimized to show details of particular components. Therefore, specific structural and functional details disclosed herein are not to be interpreted as limiting, but merely as a representative basis for teaching one skilled in the art to variously employ the present invention.
The time period of the heavy metal element contaminated soil is long, the feasibility of repeatedly sampling and measuring the content of the heavy metal in the soil and evaluating the heavy metal contaminated state of the soil is low, so that a relevant prediction model needs to be established to calculate the contribution degree of various heavy metal elements to the soil contamination and predict the comprehensive Mello pollution index in the soil heavy metal. The inner-merlo index method is one of the common methods for calculating the comprehensive pollution index, and the method firstly calculates the fractional indexes of each factor, then calculates the average value of each fractional index, and calculates the comprehensive pollution index by using the maximum fractional index and the average value of each fractional index.
Acquiring soil heavy metal content data in different pollution states, and defining the content parameter of each heavy metal in the soil of the area A as Z.
In formula (1), PiIndicating the pollution index of heavy metal element i in the soil, CiRepresents the measured value of the content of heavy metal element i in soil, SiAnd (4) standard data of the content of the heavy metal element i in the soil are shown.
In formula (2), PnIndicating the comprehensive pollution index of soil heavy metal elements of a sampling point, (P)i)maxRepresents the maximum value of the single pollution indexes of the heavy metal elements. P calculated from the formula (2)nThe value of (A) can be divided into a safe state G1, a safe state G2, a light pollution state G3, a moderate pollution state G4 and a heavy pollution state G5.
The influence degrees of various heavy metals in the soil on the comprehensive soil pollution index to be predicted are different, so that the importance scores (or contribution degrees) of different heavy metal elements in different pollution states need to be sequenced, and the heavy metal with high importance score is selected for subsequent regression prediction. Specifically, by using a random forest algorithm in machine learning, feature extraction and importance score sorting are firstly carried out on various heavy metal elements in soil, and then several elements with high importance scores are selected as the input of a support vector regression algorithm, so that the purpose of estimating the comprehensive pollution index of the meropheria in the soil is achieved. For example, the support vector regression model may use Radial Basis Functions (RBFs), and genetic algorithms may be used to optimize the parameters of the support vector regression algorithm. The genetic algorithm is a search algorithm constructed by an artificial mode, and is a mathematical mode simulation of a biological evolution process to a certain extent. The learning purpose is to learn the rules hidden behind the data, and the ability of a trained network to provide suitable output for data outside a learning set with the same rules is called generalization ability, and the generalization ability can represent the prediction ability of the algorithm model for unknown data, and the generalization of the support vector regression model can be improved by optimizing the parameters of the support vector regression algorithm by using the genetic algorithm as described above.
The method and the device for predicting the heavy metal pollution index of regional soil according to the embodiment of the invention are described below with reference to the accompanying drawings.
First, a regional soil heavy metal pollution index prediction method according to an embodiment of the present invention will be described with reference to the accompanying drawings, and fig. 1 is a flowchart of the regional soil heavy metal pollution index prediction method according to an embodiment of the present invention.
As shown in fig. 1, in step 101, data of heavy metal content of a soil sample is obtained. In step 102, based on the heavy metal content data of the soil sample, the contribution degree sequence of various heavy metal elements of the soil sample in different pollution states is determined. The contamination status may include a safe status, a not yet safe status, a lightly contaminated status, a moderately contaminated status, and a heavily contaminated status. In an example, determining the ranking of the contribution degrees of the heavy metal elements of the soil sample under different pollution states may include: and taking the heavy metal content data and the pollution state as a random forest algorithm training sample set, and constructing a random forest algorithm pollution classification model to sort the contribution degrees of various heavy metal elements in different pollution states.
In step 103, a genetic algorithm may be used to optimize the penalty function and kernel function parameters of the support vector regression model. Specifically, the optimizing the penalty function and the kernel function parameter of the support vector regression model by using the genetic algorithm may include: setting the value ranges of the penalty function and the kernel function parameter; encoding a set of randomly generated support vector regression parameters; determining the fitness function value of each individual in the population; carrying out genetic operation on the current population until the current population meets the convergence condition; determining the obtained penalty function and kernel function parameters as the optimal penalty function and kernel function parameters of the support vector regression model so as to establish a genetic algorithm-support vector regression model; and determining whether the genetic algorithm-support vector regression model reaches the preset precision, outputting the genetic algorithm-support vector regression model if the genetic algorithm-support vector regression model reaches the preset precision, and resetting the value ranges of the penalty function and the kernel function parameter and reestablishing the genetic algorithm-support vector regression model if the genetic algorithm-support vector regression model does not reach the preset precision. The generalization of the support vector regression model can be improved by optimizing the parameters of the support vector regression algorithm by using the genetic algorithm.
In the genetic algorithm, the degree of goodness and badness of each individual is evaluated according to the fitness of the individual, so that the genetic chance of the individual is determined. For example, the objective function value may be a non-negative value, and the maximum value of the function is used as the optimization goal, so the objective function value can be directly used as the fitness of the individual. In another example, the magnitude of the fitness function value may be determined by calculating an error function value, the individual with the large fitness function value being inherited to the next generation, wherein the magnitude of the error function value is inversely proportional to the magnitude of the fitness function value. A population is an initially given set of multiple solutions, which is a subset of the problem solution space. Individuals are individual elements in a population and are typically represented by a data structure that describes their basic genetic structure, e.g., a string of 0, 1 and length/. The chromosome is a code string obtained by encoding an individual. Each position in the chromosome is called a gene, and a useful information segment composed of several genes is called a genome. Fitness function is a function that is used to measure the fitness of individuals in a population.
Genetic algorithms are evolutionary operations performed on populations and require some initial population data representing initial search points to be prepared for the population. For example, the size of the population size may be taken to be 4, i.e., the population consists of 4 individuals, each of which may be generated by a random method (e.g., randomly generating 011101, 101011, 011100, and 111001). And the genetic manipulation may include at least one of selection, crossover, and mutation. The selection operation (or called replication operation) is to transmit the individuals with higher fitness in the current population to the next generation population according to some rule or model. In general, individuals with higher fitness will have more opportunity to be passed on to next generation populations. Crossover operations are the main operations in genetic algorithms to create new individuals, exchanging parts of chromosomes between two individuals with a certain probability. In an example, a single-point crossing method can be adopted, and the specific operation process is as follows: firstly, randomly pairing groups; secondly, randomly setting the position of a cross point; finally, the partial genes between the paired chromosomes are exchanged with each other. Mutation is an operation of changing the gene value of an individual or some loci with a small probability, and is also an operation method for generating new individuals. In an example, a basic bit mutation method can be used to perform mutation operation, and the specific operation process is as follows: firstly, determining the gene variation position of each individual; then, the original gene value of the variation point is inverted according to a certain probability.
In step 104, the merlo comprehensive pollution index in the soil can be predicted by taking a predetermined number of heavy metal elements with contribution degrees ranked in the top as input parameters of a support vector regression model. Through the operation, the comprehensive pollution index of the merozoite in the soil can be accurately predicted by using the small sample data.
In addition, it should be noted that the process shown in fig. 1 is merely an example, and those skilled in the art will appreciate that certain steps may be added or omitted as needed (e.g., step 103 may be omitted) without departing from the scope and spirit of the present disclosure.
Fig. 2 is a block diagram of a random forest-genetic algorithm-support vector regression model according to an embodiment of the present invention.
As shown in fig. 2, at 201, the heavy metal element data in each sample soil is studied. The contamination status may include a safe status, a still safe status, a lightly contaminated status, a moderately contaminated status, and a heavily contaminated status. At 202, parameters Z and pollution states of various heavy metal elements are used as a Random Forest (RF) training sample set to construct an RF pollution classification model under the current pollution state (203). By using a random forest algorithm in machine learning, firstly, feature extraction and importance grading sequencing are carried out on various heavy metal elements in soil, and then, several elements with high importance grades are selected as the input of a support vector regression algorithm, so that the purpose of estimating the comprehensive pollution index of the plum in the soil is achieved. For example, the support vector regression model may employ a Radial Basis Function (RBF), and the parameters of the support vector regression algorithm may be optimized (204) using a genetic algorithm (GA algorithm). At 205, a random forest-genetic algorithm-support vector regression model (RF-GA-SVR model) is trained to obtain a RF-GA-SVR model that satisfies a predetermined accuracy. Further, at 206, the Metro composite pollution index in the soil can be accurately predicted by using the RF-GA-SVR model satisfying the predetermined precision.
FIG. 3 is a flow diagram for optimizing penalty function and kernel function parameters of a support vector regression model using a genetic algorithm, according to an embodiment of the invention.
In step 301, a genetic algorithm is used to perform an optimization process on the penalty function and the kernel function parameter of the support vector regression model. In step 302, data is collected and processed. At step 303, the data is randomly ordered, and a training set and a test set are set. In step 304, the population number, the maximum iteration number, the penalty factor, and the function parameter value range are initialized. At step 305, a coding population is generated. In the example, after each training, a set of penalty functions and kernel function parameters is generated, and the generated penalty functions and kernel function parameters are recorded. At step 306, fitness of the individual is calculated. For example, the error of each neural network on the training set can be calculated and used as the fitness function. At step 307, it is determined whether the fitness of the individual meets a termination condition. In an example, whether the fitness of the individual meets the termination condition may be determined by setting a maximum number of evolutionary iterations, for example, the maximum number of evolutionary iterations may be set to 100, and if the fitness function tends to be stable after the model iterates for 100 times, it may be determined that the fitness of the individual meets the convergence condition. If the fitness of the individual is determined not to meet the termination condition, the individual is subjected to genetic manipulation (e.g., selection, crossover or mutation) and the fitness of the individual is recalculated, and the above operations are repeated until the fitness of the individual meets the termination condition. After determining that the fitness of the individual satisfies the termination condition in step 307, step 309 is performed to output an optimal penalty factor and a kernel function parameter. At step 310, a training set is trained. At step 311, it is determined whether the model has reached a predetermined accuracy. If it is determined in step 311 that the model has reached the predetermined accuracy, the model is output in step 312 and the process ends in step 313. If it is determined in step 311 that the model does not reach the predetermined accuracy, the process proceeds to step 304, the population number, the maximum iteration number, the penalty factor and the function parameter value range are initialized again, and steps 305 to 310 are performed again. In an example, if the model iterates many times and does not converge, it is not reasonable to set the parameter value range, and at this time, the parameter value range needs to be reset.
The parameters of the support vector regression prediction model are optimized by using a genetic algorithm, so that the subjectivity of artificially selecting the parameters can be reduced, and the prediction precision of the model is improved.
Fig. 4 is a block diagram illustrating a structure of a regional soil heavy metal pollution index prediction apparatus according to an embodiment of the present invention.
As shown in fig. 4, the regional soil heavy metal pollution index prediction apparatus according to an embodiment of the present invention may include: a data obtaining unit 401, configured to obtain heavy metal content data of a soil sample; the sorting unit 402 is configured to determine, based on the heavy metal content data, a sorting of contribution degrees of various heavy metal elements of the soil sample in different pollution states; and a prediction unit 403 for predicting the merlo comprehensive pollution index in the soil by using a predetermined number of heavy metal elements with the contribution degrees ranked in the top as input parameters of the support vector regression model.
The contamination status may include a safe status, a not yet safe status, a lightly contaminated status, a moderately contaminated status, and a heavily contaminated status.
The sorting unit 402 may determine the sorting of the contribution degrees of the heavy metal elements of the soil sample under different pollution states by: and taking the heavy metal content data and the pollution state as a random forest algorithm training sample set, and constructing a random forest algorithm pollution classification model to sort the contribution degrees of various heavy metal elements in different pollution states.
The regional soil heavy metal pollution index prediction device may further include: and the optimizing unit 404 is configured to optimize the penalty function and the kernel function parameter of the support vector regression model by using a genetic algorithm.
In an example, the optimizing unit 404 may optimize the penalty function and kernel function parameters of the support vector regression model using a genetic algorithm by: setting a value taking range of a penalty function and a kernel function parameter; encoding a set of randomly generated support vector regression parameters; determining fitness function values of all individuals in the population; carrying out genetic operation on the current population until the current population meets the convergence condition; determining the obtained penalty function and kernel function parameters as the optimal penalty function and kernel function parameters of the support vector regression model so as to establish a genetic algorithm-support vector regression model; and determining whether the genetic algorithm-support vector regression model reaches the preset precision, outputting the genetic algorithm-support vector regression model if the genetic algorithm-support vector regression model reaches the preset precision, and resetting the value ranges of the penalty function and the kernel function parameter and reestablishing the genetic algorithm-support vector regression model if the genetic algorithm-support vector regression model does not reach the preset precision.
For example, the optimizing unit 404 may determine the magnitude of the fitness function value by calculating an error function value, which is inversely proportional to the magnitude of the fitness function value, and pass the individual having the large fitness function value to the next generation. In an example, the genetic manipulation can include at least one of selection, crossover, and mutation.
The specific operations shown above in conjunction with fig. 1 to 3 may be respectively performed by corresponding units in the apparatus shown in fig. 4, and details of the specific operations will not be described herein.
FIG. 5 shows a schematic structural diagram of a computing device according to an embodiment of the invention.
As shown in fig. 5, a computing device 500 provided according to an embodiment of the present invention may include a processor 501 and a storage component 502, where the storage component 502 stores therein computer-executable instructions, and when the computer-executable instructions are executed by the processor 501, the regional soil heavy metal pollution index prediction method according to any one of the foregoing embodiments is performed.
In addition, according to an embodiment of the present invention, there is also provided a computer-readable storage medium having a computer program stored thereon for executing the regional soil heavy metal pollution index prediction method according to any one of the foregoing embodiments.
By adopting the method, the expression conditions of various heavy metal elements in the soil can be obtained through calculation, so that different treatment schemes can be formulated according to the contribution degrees of different heavy metal elements; in addition, the method can predict the heavy metal pollution condition of the soil through small sample data, so that the influence of data loss or insufficient data on pollution prediction is effectively avoided.
The processes, methods or algorithms disclosed herein may be delivered to or implemented by a processing device, controller or computer, which may include any existing programmable or dedicated electronic control unit. Similarly, the processes, methods or algorithms may be stored as data and instructions executable by a controller or computer in a variety of forms including, but not limited to, information permanently stored on non-writable storage media such as ROM devices and information variably stored on writable storage media such as floppy disks, magnetic tapes, CDs, RAM devices and other magnetic and optical media. The processes, methods, or algorithms may also be implemented in software executable objects. Alternatively, the processes, methods or algorithms may be implemented in whole or in part using suitable hardware components (such as ASICs, FPGAs, state machines, controllers or other hardware components or devices), or a combination of hardware, software and firmware components.
While exemplary embodiments are described above, it is not intended that these embodiments describe all possible forms of the invention. Rather, the words used in the specification are words of description rather than limitation, and it is understood that various changes may be made without departing from the spirit and scope of the invention. Furthermore, features of various implementing embodiments may be combined to form further embodiments of the invention.
Claims (10)
1. A regional soil heavy metal pollution index prediction method comprises the following steps:
acquiring heavy metal content data of a soil sample;
determining the ordering of the contribution degrees of various heavy metal elements of the soil sample under different pollution states based on the heavy metal content data; and
and predicting the Meiro comprehensive pollution index in the soil by taking the heavy metal elements with the predetermined number of contribution degrees ranked in the top as input parameters of a support vector regression model.
2. The regional soil heavy metal pollution index prediction method of claim 1, wherein the pollution state comprises a safe state, a still safe state, a light pollution state, a moderate pollution state and a heavy pollution state.
3. The regional soil heavy metal pollution index prediction method of claim 1, wherein determining the ranking of the contribution degrees of the heavy metal elements of the soil sample under different pollution states comprises: and taking the heavy metal content data and the pollution state as a random forest algorithm training sample set, and constructing a random forest algorithm pollution classification model to sort the contribution degrees of various heavy metal elements in different pollution states.
4. The regional soil heavy metal pollution index prediction method of claim 1, further comprising: and optimizing the penalty function and the kernel function parameter of the support vector regression model by using a genetic algorithm.
5. The regional soil heavy metal pollution index prediction method of claim 4, wherein the optimizing the penalty function and the kernel function parameters of the support vector regression model using the genetic algorithm comprises:
setting the value ranges of the penalty function and the kernel function parameter;
encoding a set of randomly generated support vector regression parameters;
determining fitness function values of all individuals in the population;
carrying out genetic operation on the current population until the current population meets the convergence condition;
determining the obtained penalty function and kernel function parameters as the optimal penalty function and kernel function parameters of the support vector regression model so as to establish a genetic algorithm-support vector regression model;
and determining whether the genetic algorithm-support vector regression model reaches the preset precision, outputting the genetic algorithm-support vector regression model if the genetic algorithm-support vector regression model reaches the preset precision, or resetting the value ranges of the penalty function and the kernel function parameter and reestablishing the genetic algorithm-support vector regression model.
6. The regional soil heavy metal pollution index prediction method of claim 5, wherein the magnitude of the fitness function value is determined by calculating an error function value, and the individual having the large fitness function value is inherited to a next generation, wherein the magnitude of the error function value is inversely proportional to the magnitude of the fitness function value.
7. The regional soil heavy metal pollution index prediction method of claim 5, wherein the genetic manipulation comprises at least one of selection, crossover, and mutation.
8. An area soil heavy metal pollution index prediction device comprises:
the data acquisition unit is used for acquiring heavy metal content data of the soil sample;
the sorting unit is used for determining the sorting of the contribution degrees of various heavy metal elements of the soil sample in different pollution states based on the heavy metal content data; and
and the prediction unit is used for predicting the Mello comprehensive pollution index in the soil by taking the heavy metal elements with the predetermined number of contribution degrees ranked in the top as input parameters of the support vector regression model.
9. A computer-readable storage medium having stored thereon a computer program for executing the regional soil heavy metal pollution index prediction method according to any one of claims 1 to 7.
10. A computing device comprising a storage component and a processor, wherein the storage component stores computer-executable instructions which, when executed by the processor, perform the regional soil heavy metal pollution index prediction method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111135431.5A CN113837474A (en) | 2021-09-27 | 2021-09-27 | Regional soil heavy metal pollution index prediction method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111135431.5A CN113837474A (en) | 2021-09-27 | 2021-09-27 | Regional soil heavy metal pollution index prediction method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113837474A true CN113837474A (en) | 2021-12-24 |
Family
ID=78970800
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111135431.5A Pending CN113837474A (en) | 2021-09-27 | 2021-09-27 | Regional soil heavy metal pollution index prediction method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113837474A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114509556A (en) * | 2022-01-10 | 2022-05-17 | 北京科技大学 | Method for predicting concentration of heavy metal pollutants in site |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102005135A (en) * | 2010-12-09 | 2011-04-06 | 上海海事大学 | Genetic algorithm-based support vector regression shipping traffic flow prediction method |
| CN104198512A (en) * | 2014-08-18 | 2014-12-10 | 北京农业质量标准与检测技术研究中心 | Support vector machine-based X-ray fluorescence spectrum analysis method and support vector machine-based X-ray fluorescence spectrum analysis device |
| CN107545103A (en) * | 2017-08-19 | 2018-01-05 | 安徽省环境科学研究院 | Coal field heavy metal content in soil spatial model method for building up |
| CN108446715A (en) * | 2018-02-07 | 2018-08-24 | 广东省生态环境技术研究所 | A kind of heavy metal pollution of soil Source Apportionment, system and device |
| CN108805447A (en) * | 2018-06-11 | 2018-11-13 | 深圳汇创联合自动化控制有限公司 | A kind of farmland soil heavy metals contamination analysis early warning system |
| CN110264094A (en) * | 2019-06-24 | 2019-09-20 | 武汉轻工大学 | Evaluation method, equipment, storage medium and the device of heavy metal pollution of soil |
| CN110969345A (en) * | 2019-11-20 | 2020-04-07 | 北京淖尔科技有限公司 | Risk assessment method based on soil heavy metal pollution path analysis |
| CN112614552A (en) * | 2021-01-04 | 2021-04-06 | 武汉轻工大学 | BP neural network-based soil heavy metal content prediction method and system |
| CN113435707A (en) * | 2021-06-03 | 2021-09-24 | 大连钜智信息科技有限公司 | Soil testing and formulated fertilization method based on deep learning and weighted multi-factor evaluation |
-
2021
- 2021-09-27 CN CN202111135431.5A patent/CN113837474A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102005135A (en) * | 2010-12-09 | 2011-04-06 | 上海海事大学 | Genetic algorithm-based support vector regression shipping traffic flow prediction method |
| CN104198512A (en) * | 2014-08-18 | 2014-12-10 | 北京农业质量标准与检测技术研究中心 | Support vector machine-based X-ray fluorescence spectrum analysis method and support vector machine-based X-ray fluorescence spectrum analysis device |
| CN107545103A (en) * | 2017-08-19 | 2018-01-05 | 安徽省环境科学研究院 | Coal field heavy metal content in soil spatial model method for building up |
| CN108446715A (en) * | 2018-02-07 | 2018-08-24 | 广东省生态环境技术研究所 | A kind of heavy metal pollution of soil Source Apportionment, system and device |
| CN108805447A (en) * | 2018-06-11 | 2018-11-13 | 深圳汇创联合自动化控制有限公司 | A kind of farmland soil heavy metals contamination analysis early warning system |
| CN110264094A (en) * | 2019-06-24 | 2019-09-20 | 武汉轻工大学 | Evaluation method, equipment, storage medium and the device of heavy metal pollution of soil |
| CN110969345A (en) * | 2019-11-20 | 2020-04-07 | 北京淖尔科技有限公司 | Risk assessment method based on soil heavy metal pollution path analysis |
| CN112614552A (en) * | 2021-01-04 | 2021-04-06 | 武汉轻工大学 | BP neural network-based soil heavy metal content prediction method and system |
| CN113435707A (en) * | 2021-06-03 | 2021-09-24 | 大连钜智信息科技有限公司 | Soil testing and formulated fertilization method based on deep learning and weighted multi-factor evaluation |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114509556A (en) * | 2022-01-10 | 2022-05-17 | 北京科技大学 | Method for predicting concentration of heavy metal pollutants in site |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Dharma et al. | Prediction of Indonesian inflation rate using regression model based on genetic algorithms | |
| CN110084271B (en) | Method and device for identifying picture category | |
| Köksalan et al. | An interactive territory defining evolutionary algorithm: iTDEA | |
| JP5733229B2 (en) | Classifier creation device, classifier creation method, and computer program | |
| EP2026255A2 (en) | Apparatus and method for processing information, recording medium and computer program | |
| CN116383912B (en) | Micro motor structure optimization method and system for improving control precision | |
| CN108491686B (en) | Bidirectional XGboost-based gene regulation and control network construction method | |
| CN112215259A (en) | Gene selection method and apparatus | |
| CN113837474A (en) | Regional soil heavy metal pollution index prediction method and device | |
| CN114004153A (en) | Penetration depth prediction method based on multi-source data fusion | |
| CN112256209B (en) | Parameter configuration optimization method and optimization system of cloud storage system | |
| CN110298506A (en) | A kind of urban construction horizontal forecast system | |
| TWI781461B (en) | Information processing device, information processing method and program | |
| CN116993548A (en) | Incremental learning-based education training institution credit assessment method and system for LightGBM-SVM | |
| CN115201394B (en) | Multi-component transformer oil chromatography online monitoring method and related device | |
| CN111178580A (en) | Supermarket site selection method based on improved BP neural network | |
| Malik et al. | XGBoost: A deep dive into boosting | |
| CN115908909A (en) | Evolutionary neural architecture searching method and system based on Bayes convolutional neural network | |
| US20230214668A1 (en) | Hyperparameter adjustment device, non-transitory recording medium in which hyperparameter adjustment program is recorded, and hyperparameter adjustment program | |
| CN112132259B (en) | Neural network model input parameter dimension reduction method and computer readable storage medium | |
| CN113642226A (en) | Training method of fair machine learning model based on multi-objective evolutionary algorithm | |
| CN111026661B (en) | Comprehensive testing method and system for software usability | |
| CN110162704B (en) | Multi-scale key user extraction method based on multi-factor genetic algorithm | |
| KR102103578B1 (en) | Method for Selecting Asset Portfolio | |
| Wahyono et al. | Optimization of Random Forest with Genetic Algorithm for Determination of Assessment |
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 |
Application publication date: 20211224 |
|
| RJ01 | Rejection of invention patent application after publication |