CN117035454A - Soil pollution repair model training method, system, electronic equipment and medium - Google Patents
Soil pollution repair model training method, system, electronic equipment and medium Download PDFInfo
- Publication number
- CN117035454A CN117035454A CN202311012036.7A CN202311012036A CN117035454A CN 117035454 A CN117035454 A CN 117035454A CN 202311012036 A CN202311012036 A CN 202311012036A CN 117035454 A CN117035454 A CN 117035454A
- Authority
- CN
- China
- Prior art keywords
- soil
- model
- training
- data
- preset
- 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
- 238000012549 training Methods 0.000 title claims abstract description 158
- 238000000034 method Methods 0.000 title claims abstract description 85
- 238000003900 soil pollution Methods 0.000 title claims abstract description 59
- 230000008439 repair process Effects 0.000 title claims abstract description 40
- 239000002689 soil Substances 0.000 claims abstract description 145
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 27
- 230000002068 genetic effect Effects 0.000 claims abstract description 19
- 238000012360 testing method Methods 0.000 claims abstract description 18
- 238000012545 processing Methods 0.000 claims abstract description 17
- 238000005067 remediation Methods 0.000 claims description 30
- 239000011159 matrix material Substances 0.000 claims description 26
- 238000011160 research Methods 0.000 claims description 24
- 230000006870 function Effects 0.000 claims description 19
- 230000008569 process Effects 0.000 claims description 19
- 238000003062 neural network model Methods 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000003860 storage Methods 0.000 claims description 7
- 239000003795 chemical substances by application Substances 0.000 claims description 6
- 238000005259 measurement Methods 0.000 claims description 6
- 230000035772 mutation Effects 0.000 claims description 5
- 238000012804 iterative process Methods 0.000 claims description 4
- 238000002347 injection Methods 0.000 claims description 3
- 239000007924 injection Substances 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 15
- 238000004891 communication Methods 0.000 description 11
- 238000010586 diagram Methods 0.000 description 10
- 239000000243 solution Substances 0.000 description 10
- 239000003344 environmental pollutant Substances 0.000 description 8
- 231100000719 pollutant Toxicity 0.000 description 8
- 230000008859 change Effects 0.000 description 5
- 238000010219 correlation analysis Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 210000002569 neuron Anatomy 0.000 description 4
- 230000035699 permeability Effects 0.000 description 4
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 4
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000001419 dependent effect Effects 0.000 description 3
- 238000012163 sequencing technique Methods 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- 230000004913 activation Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 239000006185 dispersion Substances 0.000 description 2
- 239000003864 humus Substances 0.000 description 2
- 238000011065 in-situ storage Methods 0.000 description 2
- 238000009533 lab test Methods 0.000 description 2
- 230000003647 oxidation Effects 0.000 description 2
- 238000007254 oxidation reaction Methods 0.000 description 2
- 239000002245 particle Substances 0.000 description 2
- 239000003209 petroleum derivative Substances 0.000 description 2
- 239000000126 substance Substances 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000003213 activating effect Effects 0.000 description 1
- 239000012190 activator Substances 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 239000000356 contaminant Substances 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 239000012530 fluid Substances 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000005416 organic matter Substances 0.000 description 1
- 239000007800 oxidant agent Substances 0.000 description 1
- 230000001590 oxidative effect Effects 0.000 description 1
- 239000011148 porous material Substances 0.000 description 1
- 108090000623 proteins and genes Proteins 0.000 description 1
- 238000009877 rendering Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000004083 survival effect Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- 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
- G06Q10/06315—Needs-based resource requirements planning or 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
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
- G06Q10/103—Workflow collaboration or project management
-
- 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
- G06Q50/10—Services
- G06Q50/26—Government or public services
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Educational Administration (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Quality & Reliability (AREA)
- General Health & Medical Sciences (AREA)
- Operations Research (AREA)
- Data Mining & Analysis (AREA)
- Development Economics (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Primary Health Care (AREA)
- Processing Of Solid Wastes (AREA)
Abstract
A soil pollution repair model training method, a system, electronic equipment and a medium relate to the technical field of big data processing. The method comprises the following steps: acquiring polluted soil data and space distribution information corresponding to the polluted soil; determining pollution characteristic information of each subarea according to the space distribution information and the polluted soil data; according to a preset gray correlation formula, calculating to obtain a main control pollution factor, determining model training parameters through a preset super-parameter genetic algorithm, and constructing an initial training model according to a plurality of main control factors and the model training parameters; inputting a plurality of main control pollution factors as model input features into an initial training model, and training according to training set data and test set data until the error rate between the corresponding output result and a preset standard value is smaller than a preset threshold value; the initial training model is used as a final soil pollution repair model. By implementing the technical scheme provided by the application, the effect of rapidly and accurately predicting the soil pollution restoration result is achieved.
Description
Technical Field
The application relates to the technical field of big data processing, in particular to a soil pollution repair model training method, a system, electronic equipment and a medium.
Background
With the development of big data age, various industries are also increasingly frequent in applying big data processing problems. For example, for soil remediation, soil pollution problems in soft soil are increasingly prominent due to the increasing human activity and industrialization process. To solve this problem and protect human health and environmental sustainability, scientists have actively studied soil pollution remediation technologies for soft soil.
At present, in the technology for restoring the soil pollution of the soft soil, the traditional method mainly analyzes the technology for restoring the soil through laboratory test results, and makes a decision on a scheme for restoring the soil pollution of the soft soil through historical data collected by staff.
However, in practical application, a large amount of repair data in practical projects is not fully utilized, and for the difference of regional soft soil characteristics and pollutant distribution, a large amount of time is required to be consumed by staff when inputting polluted soil information, and the accuracy is low, so that the training method for the soil pollution repair model is urgently improved.
Disclosure of Invention
The application provides a soil pollution repair model training method, a system, electronic equipment and a medium, which have the effect of rapidly and accurately predicting a soil repair result.
In a first aspect, the application provides a soil pollution remediation model training method, comprising the following steps:
acquiring polluted soil data and spatial distribution information corresponding to the polluted soil;
determining pollution characteristic information of each subarea according to the space distribution information and the polluted soil data and a preset space threshold; calculating a plurality of main control pollution factors in the pollution characteristic information according to a preset gray correlation formula, determining model training parameters through a preset super-parameter genetic algorithm, and constructing an initial training model according to the plurality of main control factors and the model training parameters;
acquiring historical soil restoration data, and dividing the historical soil restoration data into training set data and test set data according to a preset proportion;
inputting the multiple main control pollution factors as model input features into the initial training model, and training according to the training set data and the testing set data to obtain corresponding output results;
calculating an error rate between the output result and a preset standard value, and judging whether the error rate is smaller than a preset threshold value or not;
If yes, ending the model training process and taking the initial training model as a soil pollution repair model.
By adopting the technical scheme, the system combines the acquired polluted soil data with the spatial distribution information corresponding to the polluted soil, divides the polluted soil data into a plurality of subareas, calculates and obtains a plurality of main control factors according to a gray correlation formula, calculates model training parameters according to a super-parameter genetic algorithm, then constructs an initial training model, inputs the main control factors into the initial training model to obtain an output result, finishes model training when the error rate between the output result and a preset standard value is smaller than a preset threshold value, takes the initial training model as a soil pollution restoration model, can effectively improve the accuracy of soil restoration effect prediction, and assists staff to carry out restoration scheme decision aiming at the soil information of a specific area.
Optionally, the spatial distribution information is read, and the spatial coordinates corresponding to the contaminated soil data are determined;
establishing a polluted soil space model according to the space coordinates corresponding to the polluted soil data; dividing the polluted area in the polluted soil space model according to a preset space threshold value to obtain a plurality of polluted soil subareas; and according to a preset z-score standardization algorithm, converting the contaminated soil data corresponding to each contaminated soil subarea into the contaminated characteristic information with unified measurement.
By adopting the technical scheme, the system reads the space coordinates corresponding to each polluted soil data in the space distribution information, constructs a polluted soil space model according to the space coordinates, divides the polluted region in the polluted soil space model into subareas with unified size as a preset space threshold value, and converts the polluted soil data of each subarea into the polluted characteristic information with unified measurement format, so that the characteristic information of the polluted region can be converted into a plurality of subareas, and the accuracy of predicting the soil restoration effect of different areas is improved.
Optionally, extracting each research factor in the pollution characteristic information, and establishing a sample matrix corresponding to each research factor; selecting a reference sequence and a comparison sequence in the sample matrix, obtaining each target data sequence matrix through initialization and average calculation, and calculating gray correlation coefficients of each target data sequence matrix; according to the gray correlation coefficients, calculating gray correlation degrees corresponding to the research factors through the preset gray correlation degree formula; judging whether the gray correlation degree corresponding to each research factor is larger than a preset correlation degree threshold value or not; if yes, taking the research factors corresponding to the gray correlation degree as the main control factors.
By adopting the technical scheme, research factors in the pollution characteristic information are extracted according to the gray correlation method, a corresponding sample matrix is established, then a reference sequence and a comparison sequence are selected, initial value and average calculation are carried out to obtain a target data sequence matrix, gray correlation coefficients corresponding to the target data sequence matrix are calculated, colorless correlation corresponding to the research factors is calculated according to the gray correlation coefficients, research factors with gray correlation greater than a preset correlation threshold are screened out and used as main control factors, which characteristics have the most influence on describing and explaining data change can be determined, so that characteristic selection and sequencing are carried out, and the accuracy of soil restoration effect prediction is improved.
Optionally, the gray correlation formula includes:
wherein, gamma i Gray correlation degree, m is the number of samples, delta i And (5) gray correlation coefficients for each target data sequence matrix.
By adopting the technical scheme, the grey correlation degree of each research factor in the sample is calculated, so that the accuracy of soil remediation data prediction can be improved.
Optionally, acquiring each candidate model training parameter, and randomly generating a plurality of candidate model training parameter combinations; combining the candidate model training parameters as a plurality of populations, and calculating the fitness corresponding to each population through a preset fitness function; screening a population corresponding to the fitness being larger than a preset fitness threshold value as a target population; responding to preset crossover operation and mutation operation, and iterating the target population; judging whether the iterative process reaches a preset termination condition or not; if yes, taking the iterated target population as the model training parameter; if not, repeating the iterative process.
According to the technical scheme, candidate model training parameters are divided into a plurality of parameter combinations to be used as a plurality of populations according to the super-parametric genetic algorithm, the plurality of populations are calculated through a preset fitness function to obtain fitness of each population, then populations with fitness larger than a preset fitness threshold are screened out to be used as target populations, cross operation and mutation operation are carried out on the target populations, the calculated target populations are used as model training parameters, optimal model training parameters can be selected for model training, and accuracy of model training is improved.
Optionally, the initial training model is a BP neural network model.
By adopting the technical scheme, the BP neural network model is used as an initial training model for training, so that a plurality of polluted soil data can be processed, the soil pollution repair effect prediction is carried out, and the accuracy of the soil error repair effect is improved.
Optionally, obtaining contaminated soil data of the designated area; inputting the polluted soil data of the designated area into the soil pollution restoration model to obtain a predicted value of soil pollution restoration of the designated area; determining a corresponding optimal soil restoration scheme according to the pollution soil restoration predicted value, wherein the optimal soil restoration scheme comprises restoration agent names, restoration agent concentrations and injection depths; and sending the optimal soil restoration scheme to a user terminal.
By adopting the technical scheme, the polluted soil data of the area set by the staff is acquired, the polluted soil data is input into the soil pollution restoration model, the predicted soil restoration value, namely the restoration effect, is obtained, the corresponding optimal soil restoration scheme is determined, and the determined optimal soil restoration scheme is sent to the user terminal, so that the decision efficiency of the soil restoration scheme can be effectively improved.
In a second aspect of the application, a system for a soil pollution remediation model training method is provided.
The data acquisition module is used for acquiring the polluted soil data and the space distribution information corresponding to the polluted soil;
the data processing module is used for determining pollution characteristic information of each subarea according to the space distribution information and the polluted soil data and a preset space threshold value; calculating a plurality of main control pollution factors in the pollution characteristic information according to a preset gray correlation formula, determining model training parameters through a preset super-parameter genetic algorithm, and constructing an initial training model according to the plurality of main control factors and the model training parameters;
the model training module is used for acquiring historical soil restoration data and dividing the historical soil restoration data into training set data and test set data according to a preset proportion; inputting the plurality of main control pollution factors as model input features into the initial training model, and training according to the training set data and the testing set data to obtain an output result;
The error recognition module is used for calculating an error rate between the output result and a preset standard value and judging whether the error rate is smaller than a preset threshold value or not; if yes, ending the model training process and taking the initial training model as a soil pollution repair model.
In a third aspect of the application, an electronic device is provided.
A system for training a soil pollution remediation model comprises a memory, a processor and a program stored on the memory and capable of running on the processor, wherein the program can be loaded and executed by the processor to realize the training method of the soil pollution remediation model.
In a fourth aspect of the application, a computer readable storage medium is provided.
A computer readable storage medium storing a computer program which, when executed by a processor, causes the processor to implement a soil pollution remediation model training method.
In summary, one or more technical solutions provided in the embodiments of the present application at least have the following technical effects or advantages:
1. according to the method, the contaminated soil data and the spatial distribution information corresponding to the contaminated soil are acquired and combined, the contaminated soil data and the spatial distribution information corresponding to the contaminated soil are divided into a plurality of subareas, each subarea contains the contaminated characteristic information, a plurality of main control factors are calculated according to a gray correlation formula, model training parameters are calculated according to a super-parameter genetic algorithm, an initial training model is built, the main control factors are input into the initial training model to obtain an output result, when the error rate between the output result and a preset standard value is smaller than a preset threshold value, model training is finished, the initial training model is used as a soil contaminated repair model, accuracy of soil repair effect prediction can be effectively improved according to the method, and workers are assisted in making repair scheme decisions aiming at the soil information of a specific area.
2. According to the application, the optimal model training parameters are obtained through the calculation of the super-parameter genetic algorithm, so that the accuracy of model training can be improved.
3. According to the method, the gray correlation degree is calculated and screened according to the characteristic information corresponding to the acquired multiple soil pollution data, so that the main control factors are obtained, and the accuracy of soil remediation effect prediction is improved.
Drawings
Fig. 1 is a schematic flow chart of a training method for a soil pollution remediation model according to an embodiment of the present application.
Fig. 2 is a schematic diagram of a regional division flow of a soil pollution repair model training method according to an embodiment of the present application.
FIG. 3 is a schematic diagram of a main control factor acquisition flow of a soil pollution remediation model training method according to an embodiment of the present application
Fig. 4 is a schematic diagram of a system structure of a training method for a soil pollution remediation model according to an embodiment of the present application.
Fig. 5 is a schematic diagram showing a visual representation of a training method for a soil pollution remediation model according to an embodiment of the present application.
Fig. 6 is a schematic diagram of visualization of model prediction results of a training method for soil pollution remediation model according to an embodiment of the present application.
Fig. 7 is a schematic structural diagram of an electronic device according to the disclosure.
Reference numerals illustrate: 401. a data acquisition module; 402. a data processing module; 403. a model training module; 404. an error recognition module; 700. an electronic device; 701. a processor; 702. a memory; 703. a user interface; 704. a network interface; 705. a communication bus.
Detailed Description
In order that those skilled in the art will better understand the technical solutions in the present specification, the technical solutions in the embodiments of the present specification will be clearly and completely described below with reference to the drawings in the embodiments of the present specification, and it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments.
In describing embodiments of the present application, words such as "for example" or "for example" are used to mean serving as examples, illustrations, or descriptions. Any embodiment or design described herein as "such as" or "for example" in embodiments of the application should not be construed as preferred or advantageous over other embodiments or designs. Rather, the use of words such as "or" for example "is intended to present related concepts in a concrete fashion.
In the description of embodiments of the application, the term "plurality" means two or more. For example, a plurality of systems means two or more systems, and a plurality of screen terminals means two or more screen terminals. Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating an indicated technical feature. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include one or more such feature. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless expressly specified otherwise.
In order to facilitate understanding of the method and system provided by the embodiments of the present application, a description of the background of the embodiments of the present application is provided before the description of the embodiments of the present application.
At present, the traditional method is mainly used for analyzing the soil remediation technology through laboratory test results, and a decision is made on a soil remediation scheme of the soft soil through historical data collected by workers, however, a large amount of remediation data in actual projects is not fully utilized, and for the differences of regional soft soil characteristics and pollutant distribution, workers often need to consume a large amount of time and have low accuracy when inputting contaminated soil information.
The embodiment of the application discloses a soil pollution repair model training method, which is mainly used for solving the problem of inaccurate manual prediction of a polluted soil repair effect by acquiring polluted soil data, selecting a main control factor through a gray correlation method, determining model training parameters through a super-parameter genetic algorithm, inputting the main control factor into a BP neural network model for training, and obtaining the soil pollution repair model.
Those skilled in the art will appreciate that the problems associated with the prior art are solved by the present application, and a detailed description of a technical solution according to an embodiment of the present application is provided below, wherein the detailed description is given with reference to the accompanying drawings.
Referring to fig. 1, a soil pollution repair model training method includes S10 to S50, specifically including the steps of:
s10: and acquiring the polluted soil data and the spatial distribution information corresponding to the polluted soil.
Specifically, the system acquires contaminated soil data in an exemplary site database, and the exemplary site database is used for storing soil information such as soft soil contaminated site soil characteristics, hydrogeologic characteristics, pollutant spatial distribution characteristics and the like in an exemplary area collected by staff. Soil characteristics mainly include particle size distribution, permeability, water content, pH, redox potential, humus content, etc., which affect the migration and conversion of contaminants in the field.
In a preferred embodiment of the present application, raw data collected for a number of reasons typically presents a number of problems, such as incomplete, incorrect, irregular, etc., and after obtaining contaminated soil data and spatially distributed information corresponding to the contaminated soil, the system performs a data cleansing operation. The data cleansing operation includes: filling the missing land information through a preset missing value processing algorithm, for example, in the application, the acquired polluted soil data and the spatial distribution information corresponding to the polluted soil are processed through the existing mean value filling method, the K nearest neighbor filling method and the multiple filling method, and the data is filled to obtain complete data. Then processing the polluted soil data and the corresponding spatial distribution information of the polluted soil through a preset linear correlation analysis algorithm, for example, in the application, selecting the existing Pearson correlation coefficient to perform correlation analysis, calculating according to a product difference method, and taking the dispersion of two variables and respective average values as the basis, wherein the degree of linear correlation between the two variables can be reflected definitely through multiplication of the two dispersions, the coefficient is represented by r, and if r >0, the positive correlation of the two variables is indicated, namely, if one variable value is large, the other variable value is also larger; when r, it indicates that two variables are inversely related, i.e. the larger one variable value is, the smaller the other variable value is; when r=0, it is shown that the two variables are not linearly related, but there may be other ways of correlation, such as a curvilinear way; when r=1, it means that the two variables x and y can be described by a linear equation, and after linear correlation analysis, a worker can primarily understand the association relationship between the characteristics of each contaminated soil.
S20: and determining pollution characteristic information of each subarea according to the space distribution information and the polluted soil data and a preset space threshold value.
Specifically, after the system performs data cleaning on the spatial distribution information and the polluted soil data, the system combines the cleaned spatial distribution information and the cleaned polluted soil data, so that a teaching site is divided as finely as possible, and the polluted soil characteristics of each subarea are accurately obtained.
Referring to fig. 2, specific steps may include S21 to S23:
s21: reading the space distribution information and determining the space coordinates corresponding to the polluted soil data;
and establishing a polluted soil space model according to the space coordinates corresponding to the polluted soil data.
Illustratively, the system generates continuous contaminated soil distribution data, such as nearest neighbor interpolation, from the acquired spatial distribution information and the contaminated soil data by a preset spatial interpolation algorithm, which is a method of deducing values at unknown locations from known discrete point data, for creating continuous spatial data. And then, the space coordinates of the characteristics of each polluted soil in the space distribution information are read, and the system combines the space coordinates and the corresponding characteristics of the polluted soil through preset 3d modeling software, for example, the 3d modeling software is GIS software, so that a 3d soil space model containing the characteristics of the polluted soil is obtained. The staff can more intuitively check the polluted soil data in the demonstration site through the 3d polluted soil space model and carry out subsequent analysis.
S22: and dividing the polluted area in the polluted soil space model according to a preset space threshold value to obtain a plurality of polluted soil subareas.
Illustratively, the preset space threshold is set by a worker, and the system divides the polluted area in the 3d polluted soil space model according to the space threshold index to obtain a plurality of polluted soil subareas with the same space size.
S23: and according to a preset z-score standardization algorithm, converting the polluted soil data corresponding to each polluted soil subarea into pollution characteristic information with unified measurement.
The system converts different amounts of data in the polluted soil data and the space distribution information corresponding to the polluted soil into unified measurement z-score values which can be identified by the system according to the existing z-score standardization algorithm after dividing a plurality of polluted soil subareas with the same space size, namely, different types of polluted soil data are converted into pollution characteristic information of the unified measurement, so that the effect of comparability among all influencing factors is achieved, the original data is standardized, and the influence of different dimensions among all factors is eliminated.
S30: and calculating a plurality of main control pollution factors in the pollution characteristic information according to a preset gray correlation formula.
Specifically, the system obtains main pollution factors in pollution characteristic information according to a preset gray correlation formula. More specifically, the degree of correlation between each pollution factor in the pollutant characteristic information and the total concentration of pollutants is calculated according to a gray correlation algorithm. Then, based on the calculation results, several pollution factors with highest correlation are determined as main (master) pollution factors, i.e., pollution factors that mainly affect the total concentration of the pollutant. For the factor between two systems, the measure of the relevance of the factor changing with time or different objects is called gray relevance, if the trend of the change of the two factors has consistency, the gray relevance is higher than a preset relevance threshold, namely the synchronous change degree is higher; otherwise, the association degree is lower.
Referring to fig. 3, specific steps may include S31 to S33:
s31: and extracting each research factor in the pollution characteristic information, and establishing a sample matrix corresponding to each research factor.
Illustratively, to provide data support for further analysis of the primary pollution factors, the system extracts various aspects affecting pollution levels, such as particle size distribution, permeability, water content, pH, redox potential, humus content, etc., from the complex pollution profile information and then establishes a sample matrix for each pollution factor. Each row of the sample matrix represents data at a point in time, and each column represents a value of a factor, by which the time-varying condition of different factors can be reflected. The sample matrix specifically comprises:
Where m represents the number of samples and n represents the number of variable indices.
S32: and selecting a reference sequence and a comparison sequence in the sample matrix, obtaining each target data sequence matrix through initialization and average calculation, and calculating gray correlation coefficients of each target data sequence matrix.
Illustratively, after obtaining the sample matrix, the system selects a reference data column, i.e., determines the target study factor, and the other sequences are comparison sequences (subsequences). The data sequence is then processed according to existing dimensionless algorithms. In the gray correlation analysis method, two methods of initialization and averaging are often used for carrying out dimensionless treatment on the data sequence so as to eliminate errors caused by differences of data of different sequences in the size range, the metering range and the physical meaning on an evaluation result, and a dimensionless target data sequence matrix is obtained. And then, according to the existing gray correlation coefficient calculation formula, calculating the target data sequence matrix to obtain the gray correlation coefficient corresponding to each research factor.
S33: according to each gray correlation coefficient, calculating gray correlation corresponding to each research factor through a preset gray correlation formula; judging whether the gray correlation degree corresponding to each research factor is larger than a preset correlation degree threshold value or not; if yes, taking the research factors corresponding to the gray correlation degree as main control factors.
After the system calculates the gray correlation coefficient and the target data sequence matrix corresponding to each research factor, the gray correlation degree corresponding to each target data sequence matrix is calculated through a preset gray correlation degree formula. The gray correlation formula specifically comprises:
wherein, gamma i Gray correlation degree, m is the number of samples, delta i And (5) gray correlation coefficients for each target data sequence matrix.
After the gray correlation corresponding to each research factor is calculated through the preset gray correlation formula, the system ranks according to the magnitude of each gray correlation, namely, the influence degree of the subsequence (variable) on the reference sequence (dependent variable) is determined. And comparing each gray correlation degree with a preset correlation degree threshold value to screen out the gray correlation degree larger than the preset correlation degree threshold value, and taking research factors corresponding to the screened gray correlation degree as main control factors. The method has the advantages that the contaminated soil data analysis is carried out through the gray correlation method, the gray correlation method can be used for selecting and sequencing the features with the highest correlation and importance in a plurality of sequence data, and the correlation coefficient is calculated and can be used for determining which features have the highest influence on describing and explaining the data change, so that the feature selection and sequencing are facilitated.
S40: determining model training parameters through a preset super-parameter genetic algorithm, and constructing an initial training model according to a plurality of main control factors and the model training parameters.
Specifically, the super-parametric genetic algorithm is a method for searching an optimal solution by simulating a natural evolution process. It is necessary to implement the encoding work at the beginning. Because the work of imitating gene coding is complex, simplification is often carried out, such as binary coding, after primary population generation, the evolution is carried out generation by generation according to the principles of survival and superior and inferior of the fittest, the approximate solution is better and better, individuals are selected according to the fitness of the individuals in the problem domain in each generation, and combination crossover and mutation are carried out by means of genetic operators of natural genetics, so that the population representing the new solution set is generated. This process will result in the offspring population of the population as if it had evolved naturally being more environmentally friendly than the previous generation, and the optimal individuals in the last population will be decoded and can be used as an approximate optimal solution to the problem.
The step of calculating model training parameters by the super-parameter genetic algorithm specifically comprises the following steps: the system acquires training parameters of each candidate model and randomly generates a plurality of candidate model training parameter combinations, wherein the candidate model training parameters comprise the number of hidden layers, the number of hidden layer neurons, the iteration times, the learning rate, the activation function and the like; combining a plurality of candidate model training parameters as a plurality of populations, and calculating the fitness corresponding to each population through a preset fitness function; then the system screens the population with the corresponding fitness larger than the preset fitness threshold value as the target population, the screening process is a process of simulating natural selection, and excellent individuals are selected for subsequent operation; after the target population is obtained, the iteration operation is performed on the target population in response to the existing crossover operation and mutation operation method, and whether a preset termination condition is reached after each iteration is completed is judged, wherein the termination condition is set by a worker, for example, the termination condition is that the maximum iteration number is reached or the number of the processed target populations reaches a preset maximum threshold value, and the like. And when the current iteration operation is detected to reach the preset termination condition, taking the target population after iteration as an optimal parameter, namely a model training parameter. If the current iteration operation is detected not to reach the preset termination condition, repeating the iteration operation until the termination condition is met. The system constructs an initial training model structure according to the calculated model training parameters and a plurality of main control factors, wherein the initial training model is a BP neural network model, namely, a plurality of main control factors are used as input layers, an hidden layer is determined according to the obtained model training parameters, the hidden layer comprises the number of neurons and the like, and an output layer is a dependent variable corresponding to the input layer, such as a concentration change value of a pollution element and an in-situ injected crack communication area.
S50: acquiring historical soil restoration data, and dividing the historical soil restoration data into training set data and test set data according to a preset proportion; and inputting a plurality of main control pollution factors serving as model input characteristics into an initial training model, and training according to training set data and test set data to obtain corresponding output results.
Specifically, the system may acquire historical soil remediation data prior to model training, including fracturing test data, chemical oxidation test data, and the like. The historical soil remediation data is then divided into training set data and test set data, for example, in a division ratio of 90% data for model training and 10% data for model testing. And inputting a plurality of main control pollution factors into the BP neural network model, starting model training according to model training parameters obtained by a super-parameter genetic algorithm, training set data and test set data, and recording an output result in each iterative operation process by the system.
In an alternative embodiment of the present application, regarding the model training process applied to in-situ injection effect decision, the main control factors obtained by the system through the gray correlation method include 12 influencing factors such as soil type, stress difference, fluid viscosity, displacement, water content, density, specific gravity, pore ratio, depth, compression coefficient, permeability coefficient Kv, permeability coefficient Kh, etc. as independent variables, and the fracture communication area in the obtained historical soil repair data is taken as the dependent variable. And then calculating according to a super-parameter genetic algorithm to obtain model training parameters, wherein the model training parameters comprise a learning rate of 0.1, an activating function selects an existing relu equation, a solver selects an existing sgd function, the iteration number is 150 and the like, and the main control factors are used as model input features to be input into the BP neural network model to obtain an output result, namely a predicted crack communication area.
In another alternative embodiment of the present application, regarding the model training process applied to the chemical oxidation decision, for example, the system takes the total petroleum hydrocarbon pollutant as a research object, obtains experimental data, and calculates 7 indexes of the main control factors including the concentration before repair, the oxidant name, the agent concentration, the activator name, the agent concentration, the water content, the organic matter content, etc. as independent variables through the gray correlation method, and the concentration after repair is taken as the independent variable. And then calculating according to a super-parameter genetic algorithm to obtain model training parameters, wherein the model training parameters comprise parameters of two hidden layers, 10 for the number of neurons of the 1 st hidden layer, 5 for the number of neurons of the 2 nd hidden layer, 0.24 for the learning rate, selecting a relu equation for an activation function, selecting sgd for a solver, and 50 times of iteration, and the like, and inputting the main control factors serving as model input features into the BP neural network model to obtain an output result, namely the concentration of the repaired total petroleum hydrocarbon pollutants.
S60: calculating an error rate between the output result and a preset standard value, and judging whether the error rate is smaller than a preset threshold value or not; if yes, ending the model training process and taking the initial training model as a soil pollution repair model.
Specifically, the system compares the obtained output result with a preset standard value, namely, compares the obtained output result with an actual value to obtain an error value, then calculates an error rate between the output result and the preset standard value according to the error value, constructs a loss function, and then judges whether the loss error of the function is converged to the actual value or not through analyzing the loss function to detect whether the model is trained. And detecting whether the error rate is smaller than a preset threshold value, if the detection result shows that the error rate is smaller than the preset threshold value, ending the model training process, and taking the trained BP neural network model as a soil pollution repair model. The accuracy of the model is evaluated by comparing the prediction result of the model with an actual standard value, and if the accuracy meets the set requirement, the model can be used for soil pollution repair, so that the model can be ensured to effectively predict and guide the soil pollution repair work in actual application.
The following are system embodiments of the present application that may be used to perform method embodiments of the present application. For details not disclosed in the platform embodiments of the present application, reference is made to the method embodiments of the present application.
Referring to fig. 4, a system for training a soil pollution repair model according to an embodiment of the present application includes: a data acquisition module 401, a data processing module 402, a model training module 403, an error identification module 404, wherein: the data acquisition module (401) is used for acquiring the polluted soil data and the space distribution information corresponding to the polluted soil;
The data processing module (402) is used for determining pollution characteristic information of each subarea according to the space distribution information and the pollution soil data and a preset space threshold value; calculating a plurality of main control pollution factors in pollution characteristic information according to a preset gray correlation formula, determining model training parameters through a preset super-parameter genetic algorithm, and constructing an initial training model according to the plurality of main control factors and the model training parameters;
the model training module (403) is used for acquiring historical soil repair data and dividing the historical soil repair data into training set data and test set data according to a preset proportion; inputting a plurality of main control pollution factors as model input features into an initial training model, and training according to training set data and test set data to obtain an output result;
the error recognition module (404) is used for calculating an error rate between the output result and a preset standard value and judging whether the error rate is smaller than a preset threshold value or not; if yes, ending the model training process and taking the initial training model as a soil pollution repair model.
Referring to fig. 5, a schematic diagram of a visual display of a soil pollution repair model training method according to an embodiment of the present application is provided, where the system for training a soil pollution repair model may be used for a user terminal to perform model training to evaluate a model training result, and the user terminal is, for example, a mobile terminal such as a computer.
Referring to fig. 6, a soil pollution repair model training method provided by the embodiment of the application is applied to a user terminal to obtain a visual schematic diagram of a model prediction result.
It should be noted that: in the device provided in the above embodiment, when implementing the functions thereof, only the division of the above functional modules is used as an example, in practical application, the above functional allocation may be implemented by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules, so as to implement all or part of the functions described above. In addition, the embodiments of the apparatus and the method provided in the foregoing embodiments belong to the same concept, and specific implementation processes of the embodiments of the method are detailed in the method embodiments, which are not repeated herein.
The application also discloses electronic equipment. Referring to fig. 7, fig. 7 is a schematic structural diagram of an electronic device according to an embodiment of the present disclosure. The electronic device 700 may include: at least one processor 701, at least one network interface 704, a user interface 703, a memory 702, at least one communication bus 705.
Wherein a communication bus 705 is used to enable connected communication between these components.
The user interface 703 may include a Display screen (Display), a Camera (Camera), and the optional user interface 703 may further include a standard wired interface, and a wireless interface.
The network interface 704 may optionally include a standard wired interface, a wireless interface (e.g., WI-FI interface), among others.
Wherein the processor 701 may include one or more processing cores. The processor 701 connects various portions of the overall server using various interfaces and lines, performs various functions of the server and processes data by executing or executing instructions, programs, code sets, or instruction sets stored in the memory 702, and invoking data stored in the memory 702. Alternatively, the processor 701 may be implemented in hardware in at least one of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), programmable logic array (Programmable Logic Array, PLA). The processor 701 may integrate one or a combination of several of a central processing unit (Central Processing Unit, CPU), an image processor (Graphics Processing Unit, GPU), and a modem, etc. The CPU mainly processes an operating system, a user interface diagram, an application program and the like; the GPU is used for rendering and drawing the content required to be displayed by the display screen; the modem is used to handle wireless communications. It will be appreciated that the modem may not be integrated into the processor 701 and may be implemented by a single chip.
The Memory 702 may include a random access Memory (Random Access Memory, RAM) or a Read-Only Memory (Read-Only Memory). Optionally, the memory 702 includes a non-transitory computer readable medium (non-transitory computer-readable storage medium). Memory 702 may be used to store instructions, programs, code, sets of codes, or instruction sets. The memory 702 may include a stored program area and a stored data area, wherein the stored program area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playing function, an image playing function, etc.), instructions for implementing the various method embodiments described above, etc.; the storage data area may store data or the like involved in the above respective method embodiments. The memory 702 may also optionally be at least one storage device located remotely from the processor 701. Referring to fig. 7, an operating system, a network communication module, a user interface module, and an application program of a soil pollution remediation model training method may be included in a memory 702 as a computer storage medium.
In the electronic device 700 shown in fig. 7, the user interface 703 is mainly used for providing an input interface for a user, and acquiring data input by the user; and processor 701 may be configured to invoke an application program in memory 702 that stores a soil pollution remediation model training method, which when executed by one or more processors 701, causes electronic device 700 to perform the method as described in one or more of the embodiments above. It should be noted that, for simplicity of description, the foregoing method embodiments are all described as a series of acts, but it should be understood by those skilled in the art that the present application is not limited by the order of acts described, as some steps may be performed in other orders or concurrently in accordance with the present application. Further, those skilled in the art will also appreciate that the embodiments described in the specification are all of the preferred embodiments, and that the acts and modules referred to are not necessarily required for the present application.
In the foregoing embodiments, the descriptions of the embodiments are emphasized, and for parts of one embodiment that are not described in detail, reference may be made to related descriptions of other embodiments.
In the several embodiments provided by the present application, it should be understood that the disclosed apparatus may be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative, such as a division of units, merely a division of logic functions, and there may be additional divisions in actual implementation, such as multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some service interface, device or unit indirect coupling or communication connection, electrical or otherwise.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed over a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable memory. Based on this understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in whole or in part in the form of a software product stored in a memory, comprising several instructions for causing a computer device (which may be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method of the various embodiments of the present application. And the aforementioned memory includes: various media capable of storing program codes, such as a U disk, a mobile hard disk, a magnetic disk or an optical disk.
The foregoing is merely exemplary embodiments of the present disclosure and is not intended to limit the scope of the present disclosure. That is, equivalent changes and modifications are contemplated by the teachings of this disclosure, which fall within the scope of the present disclosure. Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure.
This application is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a scope and spirit of the disclosure being indicated by the claims.
Claims (10)
1. The soil pollution repair model training method is characterized by comprising the following steps of:
acquiring polluted soil data and spatial distribution information corresponding to the polluted soil;
determining pollution characteristic information of each subarea according to the space distribution information and the polluted soil data and a preset space threshold; calculating a plurality of main control pollution factors in the pollution characteristic information according to a preset gray correlation formula, determining model training parameters through a preset super-parameter genetic algorithm, and constructing an initial training model according to the plurality of main control factors and the model training parameters;
acquiring historical soil restoration data, and dividing the historical soil restoration data into training set data and test set data according to a preset proportion;
inputting the multiple main control pollution factors as model input features into the initial training model, and training according to the training set data and the testing set data to obtain corresponding output results;
Calculating an error rate between the output result and a preset standard value, and judging whether the error rate is smaller than a preset threshold value or not;
if yes, ending the model training process and taking the initial training model as a soil pollution repair model.
2. The method for training a soil pollution remediation model according to claim 1, wherein determining pollution characteristic information of each sub-region according to the spatial distribution information and the polluted soil data and a preset spatial threshold value includes:
reading the space distribution information and determining space coordinates corresponding to the polluted soil data;
establishing a polluted soil space model according to the space coordinates corresponding to the polluted soil data;
dividing the polluted area in the polluted soil space model according to a preset space threshold value to obtain a plurality of polluted soil subareas;
and according to a preset z-score standardization algorithm, converting the contaminated soil data corresponding to each contaminated soil subarea into the contaminated characteristic information with unified measurement.
3. The method for training a soil pollution remediation model according to claim 1, wherein the calculating the main control pollution factor in the pollution characteristic information according to the preset gray correlation formula includes:
Extracting each research factor in the pollution characteristic information, and establishing a sample matrix corresponding to each research factor;
selecting a reference sequence and a comparison sequence in the sample matrix, obtaining each target data sequence matrix through initialization and average calculation, and calculating gray correlation coefficients of each target data sequence matrix;
according to the gray correlation coefficients, calculating gray correlation degrees corresponding to the research factors through the preset gray correlation degree formula; judging whether the gray correlation degree corresponding to each research factor is larger than a preset correlation degree threshold value or not;
if yes, taking the research factors corresponding to the gray correlation degree as the main control factors.
4. The method of claim 1, wherein the predetermined gray correlation formula comprises:
wherein, gamma i Gray correlation degree, m is the number of samples, delta i And (5) gray correlation coefficients for each target data sequence matrix.
5. The method for training a soil pollution remediation model according to claim 1, wherein determining model training parameters by a preset super-parameter genetic algorithm comprises:
Obtaining each candidate model training parameter, and randomly generating a plurality of candidate model training parameter combinations;
combining the candidate model training parameters as a plurality of populations, and calculating the fitness corresponding to each population through a preset fitness function;
screening a population corresponding to the fitness being larger than a preset fitness threshold value as a target population;
responding to preset crossover operation and mutation operation, and iterating the target population;
judging whether the iterative process reaches a preset termination condition or not;
if yes, taking the iterated target population as the model training parameter;
if not, repeating the iterative process.
6. The method for training a soil pollution remediation model according to claim 1, wherein the initial training model is a BP neural network model.
7. The method for training a soil pollution remediation model according to claim 1, wherein after the model training process is completed, a method for applying a soil pollution remediation model further comprises:
acquiring contaminated soil data of a designated area;
inputting the polluted soil data of the designated area into the soil pollution restoration model to obtain a predicted value of soil pollution restoration of the designated area;
Determining a corresponding optimal soil restoration scheme according to the pollution soil restoration predicted value, wherein the optimal soil restoration scheme comprises restoration agent names, restoration agent concentrations and injection depths;
and sending the optimal soil restoration scheme to a user terminal.
8. A system based on the soil pollution remediation model training method of claims 1-7, the system comprising: the data acquisition module (401) is used for acquiring the polluted soil data and the spatial distribution information corresponding to the polluted soil;
the data processing module (402) is used for determining pollution characteristic information of each subarea according to the space distribution information and the polluted soil data and a preset space threshold value; calculating a plurality of main control pollution factors in the pollution characteristic information according to a preset gray correlation formula, determining model training parameters through a preset super-parameter genetic algorithm, and constructing an initial training model according to the plurality of main control factors and the model training parameters;
the model training module (403) is used for acquiring historical soil repair data and dividing the historical soil repair data into training set data and test set data according to a preset proportion; inputting the plurality of main control pollution factors as model input features into the initial training model, and training according to the training set data and the testing set data to obtain an output result;
An error recognition module (404) for calculating an error rate between the output result and a preset standard value, and judging whether the error rate is smaller than a preset threshold value; if yes, ending the model training process and taking the initial training model as a soil pollution repair model.
9. An electronic device comprising a processor (701), a memory (702), a user interface (703) and a network interface (704), the memory (702) being configured to store instructions, the user interface (703) and the network interface (704) being configured to communicate to other devices, the processor (701) being configured to execute the instructions stored in the memory (702) to cause the electronic device (700) to perform a soil pollution remediation model training method according to any one of claims 1 to 7.
10. A computer readable storage medium storing instructions which, when executed, perform a soil pollution remediation model training method step as claimed in any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311012036.7A CN117035454A (en) | 2023-08-11 | 2023-08-11 | Soil pollution repair model training method, system, electronic equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311012036.7A CN117035454A (en) | 2023-08-11 | 2023-08-11 | Soil pollution repair model training method, system, electronic equipment and medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117035454A true CN117035454A (en) | 2023-11-10 |
Family
ID=88625851
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311012036.7A Pending CN117035454A (en) | 2023-08-11 | 2023-08-11 | Soil pollution repair model training method, system, electronic equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117035454A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117475338A (en) * | 2023-12-28 | 2024-01-30 | 山东舜捷资源综合利用有限公司 | Land reclamation repair process monitoring system based on land management |
-
2023
- 2023-08-11 CN CN202311012036.7A patent/CN117035454A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117475338A (en) * | 2023-12-28 | 2024-01-30 | 山东舜捷资源综合利用有限公司 | Land reclamation repair process monitoring system based on land management |
| CN117475338B (en) * | 2023-12-28 | 2024-03-29 | 山东舜捷资源综合利用有限公司 | Land reclamation repair process monitoring system based on land management |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zurell et al. | A standard protocol for reporting species distribution models | |
| Galante et al. | The challenge of modeling niches and distributions for data‐poor species: a comprehensive approach to model complexity | |
| Syfert et al. | Using species distribution models to inform IUCN Red List assessments | |
| Varela et al. | Using species distribution models in paleobiogeography: a matter of data, predictors and concepts | |
| Mateo-Sánchez et al. | A comparative framework to infer landscape effects on population genetic structure: are habitat suitability models effective in explaining gene flow? | |
| Marinoni | A stochastic spatial decision support system based on PROMETHEE | |
| Massatti et al. | Contrasting support for alternative models of genomic variation based on microhabitat preference: Species‐specific effects of climate change in alpine sedges | |
| Kuehne et al. | Past, present, and future of ecological integrity assessment for fresh waters | |
| Smith et al. | Identifying influences on model uncertainty: an application using a forest carbon budget model | |
| Muduli et al. | CPT-based probabilistic evaluation of seismic soil liquefaction potential using multi-gene genetic programming | |
| CN117035454A (en) | Soil pollution repair model training method, system, electronic equipment and medium | |
| Ballester et al. | A parallel real-coded genetic algorithm for history matching and its application to a real petroleum reservoir | |
| Siade et al. | Using heuristic multi-objective optimization for quantifying predictive uncertainty associated with groundwater flow and reactive transport models | |
| CN112347537B (en) | Calibration method and device for engineering structure numerical model, electronic equipment and medium | |
| Abualdenien et al. | Ensemble-learning approach for the classification of Levels Of Geometry (LOG) of building elements | |
| CN117473840B (en) | Simulation and optimization method and system based on construction process for sheet pile wharf | |
| Wagener et al. | Identification and evaluation of watershed models | |
| Varty et al. | Inference for extreme earthquake magnitudes accounting for a time-varying measurement process | |
| CN117056722A (en) | Prediction method and system for population quantity of planned land parcel | |
| CN115392715A (en) | Power utilization data risk assessment method, device, equipment and storage medium | |
| CN111680572B (en) | Dynamic judgment method and system for power grid operation scene | |
| CN111814309B (en) | Method for analyzing robustness and reliability of unsaturated soil slope stability | |
| Park et al. | Settlement prediction in a vertical drainage-installed soft clay deposit using the genetic algorithm (GA) back-analysis | |
| Gobeyn et al. | A variable length chromosome genetic algorithm approach to identify species distribution models useful for freshwater ecosystem management | |
| CN117095216B (en) | Model training method, system, equipment and medium based on countermeasure generation network |
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 |