CN116662893A - Water quality prediction method for optimizing SVM (support vector machine) based on improved goblet sea squirt algorithm - Google Patents
Water quality prediction method for optimizing SVM (support vector machine) based on improved goblet sea squirt algorithm Download PDFInfo
- Publication number
- CN116662893A CN116662893A CN202310479905.0A CN202310479905A CN116662893A CN 116662893 A CN116662893 A CN 116662893A CN 202310479905 A CN202310479905 A CN 202310479905A CN 116662893 A CN116662893 A CN 116662893A
- Authority
- CN
- China
- Prior art keywords
- water quality
- leader
- algorithm
- sea squirt
- svm
- 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
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims abstract description 74
- 241000251555 Tunicata Species 0.000 title claims abstract description 41
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000012706 support-vector machine Methods 0.000 title claims description 54
- 230000000739 chaotic effect Effects 0.000 claims abstract description 40
- 241000798368 Ecteinascidia Species 0.000 claims abstract description 9
- 238000012545 processing Methods 0.000 claims abstract description 7
- 238000013507 mapping Methods 0.000 claims description 25
- 238000004364 calculation method Methods 0.000 claims description 10
- 238000007781 pre-processing Methods 0.000 claims description 6
- 238000005728 strengthening Methods 0.000 claims description 3
- 230000008859 change Effects 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 35
- 238000012360 testing method Methods 0.000 description 6
- 238000005457 optimization Methods 0.000 description 5
- 238000010606 normalization Methods 0.000 description 3
- 239000010865 sewage Substances 0.000 description 3
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 2
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 229910052760 oxygen Inorganic materials 0.000 description 2
- 239000001301 oxygen Substances 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 241000251557 Ascidiacea Species 0.000 description 1
- OAICVXFJPJFONN-UHFFFAOYSA-N Phosphorus Chemical compound [P] OAICVXFJPJFONN-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000003631 expected effect Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 229910052757 nitrogen Inorganic materials 0.000 description 1
- 229910052698 phosphorus Inorganic materials 0.000 description 1
- 239000011574 phosphorus Substances 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000003911 water pollution Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/18—Water
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- 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
- G06Q50/06—Energy or water supply
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A20/00—Water conservation; Efficient water supply; Efficient water use
- Y02A20/152—Water filtration
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Economics (AREA)
- Chemical & Material Sciences (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Analytical Chemistry (AREA)
- Medicinal Chemistry (AREA)
- Food Science & Technology (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Biochemistry (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Software Systems (AREA)
- Pathology (AREA)
- Immunology (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
A water quality prediction method based on an improved goblet-sea squirt algorithm optimizes a water quality prediction model of the SVM by utilizing a sensor to collect data of water quality in real time, uploading the data to a background through a WIFI module for data processing and adopting the improved goblet-sea squirt algorithm to optimize the water quality prediction model of the SVM; introducing Fuch chaotic map into the Zun ecteinascidia algorithm, optimizing and improving the algorithm, initializing individual positions of Zun ecteinascidia population by using the improved Fuch-Tent chaotic map, and generating a new chaotic sequence so as to avoid the situation of local optimal value; constructing an SVM model by using the optimizing result with higher precision, thereby obtaining an optimized SVM model; the real-time data of each parameter of the water quality is preprocessed and input into an optimized SVM prediction model, so that the water quality can be predicted for a period of time in the future; compared with the original algorithm, the improved goblet sea squirt algorithm reduces the probability of trapping in extreme values, has better optimizing capability, and finally outputs better prediction results. The invention can predict the water quality parameter change in advance and realize global search.
Description
Technical Field
The invention relates to the technical field of water quality prediction, in particular to a water quality prediction method for optimizing SVM based on an improved goblet-sea squirt algorithm.
Background
Along with the gradual enhancement of environmental awareness, water environment management is already becoming an important measure for solving the problems of water resource shortage and water pollution aggravation, and water quality prediction and early warning are one of important research contents of water environment problems.
The water quality prediction is an important ring in the sewage treatment process, and the accuracy of the prediction effect has extremely important significance for sewage treatment. The sewage treatment process has the characteristics of strong coupling and high nonlinearity, and the expected effect is difficult to achieve by a common water quality prediction method. With the rise of machine learning, more and more prediction methods such as SVM (support vector machine) are applied to water quality prediction, and whether the penalty coefficient c and the kernel function are reliable or not seriously affects the prediction performance of the SVM model, but how to quickly and accurately select the penalty coefficient c and the kernel function of the SVM is always a problem.
With the popularity of heuristic algorithms, the goblet-sea squirt algorithm is an optimization algorithm which simulates the life habits of the goblet-sea squirts gathered into chains, the algorithm divides a group into a leader (leader) and a follower (follower), the leader (leader) searches with global optimization as a center, global searching capability is provided for the algorithm, and the follower (follower) follows the previous individual and local searching capability is provided for the algorithm. By virtue of the stability, the method has the characteristics of simple model, high realization speed and the like, and is applied to various fields. The goblet sea squirt algorithm can well solve the problems of low convergence speed and low precision of SVM (support vector machine). However, the algorithm structure of the goblet sea squirt algorithm is not complex, so that the situation of local extremum and low convergence accuracy easily occurs under the condition of facing the problem of complex calculation.
Disclosure of Invention
Aiming at the technical problems, the technical scheme provides a water quality prediction method for optimizing SVM based on an improved goblet-sea squirt algorithm, which introduces Fuch chaotic mapping into the goblet-sea squirt algorithm and optimizes and improves the goblet-sea squirt algorithm, and utilizes the improved Fuch-Tent chaotic mapping to initialize individual positions of the goblet-sea squirt population, so as to generate a new chaotic sequence, thereby avoiding the subsequent occurrence of local optimal values; the problems can be effectively solved.
The invention is realized by the following technical scheme:
a water quality prediction method based on an improved sea squirt algorithm optimizes SVM (support vector machine), utilizes a sensor to collect data of water quality in real time, and uploads the data to a background for data processing through a WIFI module; the real-time data of each parameter of the water quality is preprocessed and input into an optimized SVM prediction model, so that the water quality can be predicted for a period of time in the future; the specific operation steps comprise:
s1: acquiring a water quality sample in real time through a sensor, acquiring water quality data, and uploading the water quality data to a background through a WIFI module;
s2: after receiving the water quality data, the background performs standardized pretreatment on the water quality data in a conventional mode;
s3: taking the Gaussian radial basis function as a kernel function, and obtaining a classification decision function based on the Gaussian radial basis function and an objective function; constructing a support vector machine model according to the classification decision function and the kernel function;
s4: obtaining a penalty factor c and a kernel function with higher precision according to an improved goblet sea squirt algorithm, and optimizing and constructing a water quality prediction model of a Support Vector Machine (SVM) to obtain an optimized SVM model; the improved goblet sea squirt swarm algorithm includes:
step a: introducing Fuch chaotic mapping into a Zun ecteinascidia algorithm, optimizing and improving the algorithm, initializing individual positions of Zun ecteinascidia population by using an improved Fuch-Tent chaotic mapping formula, and generating a new chaotic sequence so as to avoid the situation of local optimal values;
step b: calculating a value of fitness;
step c: dividing the goblet sea squirt group into a leader (leader) and a follower (follower), updating the position of the leader (leader) first, and then updating the position of the follower (follower);
step d: judging whether an ending condition is reached, and if so, obtaining an optimal value; if not, the step c is carried out again.
Further, the normalization preprocessing in step S2 adopts mean variance normalization processing, and the calculation formula of the normalization preprocessing is as follows:
in the above expression, newX represents a normalized value, X represents water quality data, mean (X) represents an average value corresponding to a feature value, and std (X) represents a variance corresponding to the feature value.
The method is more focused on the distribution condition of the samples in the data set, and the influence of a small number of abnormal points on the average value and the standard deviation is small under a certain sample number, so that the standardized result does not have great deviation.
Further, the formula of the gaussian radial basis function in step S3 is:
wherein x represents any point in the kernel function; y represents a kernel function center point; x and y are water quality sample characteristic quantities; σ represents the kernel parameter.
Further, the objective function formula and the constraint condition formula described in step S3 are:
wherein ω represents the normal to the hyperplaneQuantity delta i For the relaxation variable, C is a penalty factor, i represents the ith sample; x is X i Representing a sample feature quantity; y is i Representing a sample class;representing a nonlinear mapping function; b represents the deviation amount; omega Τ Representing the normal vector of the transposed hyperplane; and when the constraint condition is met, obtaining the classification decision function.
Further, the Fuch-Tent chaotic mapping formula in the step a is as follows:
in the above, k n Not equal to 0, n ε Z+; alpha is 0.5.
The Fuch-Tent chaotic map well combines the advantages of the Fuch map and the Tent map, and the improved Fuch-Tent map has the advantages of insensitivity to an initial value, few control parameters, low complexity, traversing equalization, quicker convergence and the like, can generate chaos under the condition that the initial value is not 0, has smaller iteration times compared with the traditional limited folding chaotic map, and can better realize chaos optimizing.
Further, in the step c, the calculation formula for updating the leader (leader) position, that is, the iterative formula of the leader (leader) is:
in the above-mentioned method, the step of,representing upper and lower bounds, respectivelyBoundary (L)>Representing the location of the leader (leader),representing a pre-update food source, c 2 Is [0,1]The uniform random number of the (2) is used for strengthening the randomness of the update of the moving position of the leader and improving the searching capability; c 1 Represents convergence factor, and the value range is +.>
The calculation formula for updating the position of the follower, i.e. the iterative formula of the follower (follower) is:
wherein the follower's position is its own positionAnd the position X of the preceding individual i d At the midpoint of (2).
Further, in the step S4, a penalty factor c and a kernel function with higher accuracy are obtained according to the improved ascidian algorithm, and a water quality prediction model of the SVM is optimized and constructed, and the specific steps include:
step 1: initializing upper, lower, leader (leader) location, food source, c in algorithm 1 And c 2 Randomly initializing the value range of each parameter of the SVM according to the scale of the sea squirt and the maximum iteration number;
step 2, initializing population positions by using Fuch-Tent chaotic mapping, and leading a leader (leader)Setting as current position, calculating fitness value, taking out the +_f of the most suitable goblet sea squirt>As a food source +.>
Step 3, re-determining the position of the leader (leader) of the goblet sea squirt according to the iterative formula of the leader (leader) and determining a new fitness value, and determining the position of the follower (follower) according to the iterative formula of the follower (follower);
step 4, comparing the updated fitness value with the global optimum value of the goblet sea squirt, and updating the fitness value when the conditions are met;
and 5, judging whether the condition meets the requirement, if so, outputting an optimal value, otherwise, returning to the step 1.
Advantageous effects
Compared with the prior art, the water quality prediction method for optimizing the SVM based on the improved goblet sea squirt algorithm has the following beneficial effects:
(1) According to the method, fuch chaotic mapping is introduced into the goblet sea squirt algorithm and optimized and improved, and the improved Fuch-Tent chaotic mapping is utilized to initialize individual positions of the goblet sea squirt population, so that a new chaotic sequence is generated, and the situation that a local optimal value appears later is avoided. And (3) obtaining various water quality data through the sensor by utilizing the prediction capability of an SVM (support vector machine) and the powerful optimizing capability of the Zun sea squirt algorithm, and searching through the improved Zun sea squirt algorithm to obtain a corresponding optimal penalty factor c and a kernel function. And the SVM model is constructed by using the obtained optimal parameters, so that the prediction accuracy of the SVM model is improved. The water quality parameter change can be predicted in advance, global search is realized, SVM parameter search efficiency is improved, and other samples can be rapidly and accurately subjected to fitting prediction on the basis of small sample learning; the problem that the SVM (support vector machine) is low in convergence speed and low in precision and a local extremum is involved in optimization of the SVM by the goblet-sea squirt algorithm is solved. The water quality condition of a period of time in the future can be predicted, the frequency of manually inspecting the water quality condition can be reduced, and the labor cost is reduced.
(2) The invention utilizes an improved goblet-sea squirt algorithm which is improved by introducing Fuch chaotic map and improving the Fuch map. The improved Fuch-Tent chaotic map well combines the advantages of Fuch mapping and Tent mapping, has the advantages of insensitivity to initial values, few control parameters, low complexity, quicker traversing equalization and convergence and the like, can generate chaos under the condition that the initial value is not 0, has smaller iteration times compared with the traditional limited folding chaotic map, and can better realize chaotic optimization. And then a better prediction result is finally output.
Drawings
FIG. 1 is a schematic flow chart of the present invention.
FIG. 2 is a flowchart of the improved ecteinascidial algorithm of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. The described embodiments are only some, but not all, embodiments of the invention. Various modifications and improvements of the technical scheme of the invention, which are made by those skilled in the art, are included in the protection scope of the invention without departing from the design concept of the invention.
Example 1:
as shown in fig. 1, a water quality prediction method based on an improved goblet-sea squirt algorithm optimizes a water quality prediction model of the SVM by utilizing a sensor to collect data of water quality in real time, uploading the data to a background through a WIFI module for data processing and adopting the improved goblet-sea squirt algorithm to optimize the water quality prediction model of the SVM; introducing Fuch chaotic map into the Zun ecteinascidia algorithm, optimizing and improving the algorithm, initializing individual positions of Zun ecteinascidia population by using the improved Fuch-Tent chaotic map, and generating a new chaotic sequence so as to avoid the situation of local optimal value; constructing an SVM model by using the optimizing result with higher precision to obtain an optimized SVM model; the real-time data of each parameter of the water quality is preprocessed and input into an optimized SVM prediction model, so that the water quality can be predicted for a period of time in the future; compared with the original algorithm, the improved goblet sea squirt algorithm reduces the probability of trapping in extreme values, has better optimizing capability, and finally outputs better prediction results. The specific operation steps comprise:
s1: the water quality sample is acquired in real time through the sensor, water quality data is acquired, and the water quality data is uploaded to the background through the WIFI module.
The water quality data includes: and 5d oxygen demand, chemical oxygen demand, pH value, total phosphorus, total nitrogen and other relevant water quality data of the inlet water and the outlet water.
S2: after receiving the water quality data, the background performs standardized pretreatment on the water quality data in a conventional mode.
The standardized preprocessing adopts mean variance standardized processing, and the calculation formula of the standardized preprocessing is as follows:
in the above expression, newX represents a normalized value, X represents water quality data, mean (X) represents an average value corresponding to a feature value, and std (X) represents a variance corresponding to the feature value.
The method is more focused on the distribution condition of the samples in the data set, and the influence of a small number of abnormal points on the average value and the standard deviation is small under a certain sample number, so that the standardized result does not have great deviation.
S3: taking the Gaussian radial basis function as a kernel function, and obtaining a classification decision function based on the Gaussian radial basis function and an objective function; and constructing a support vector machine model according to the classification decision function and the kernel function.
The formula of the gaussian radial basis function is:
wherein x represents any point in the kernel function; y represents a kernel function center point; x and y are water quality sample characteristic quantities; σ represents the kernel parameter.
The objective function formula and the constraint condition formula are as follows:
wherein ω represents the normal vector of the hyperplane, δ i For the relaxation variable, C is a penalty factor, i represents the ith sample; x is X i Representing a sample feature quantity; y is i Representing a sample class;representing a nonlinear mapping function; b represents the deviation amount; omega Τ Representing the normal vector of the transposed hyperplane; and when the constraint condition is met, obtaining the classification decision function.
S4: obtaining a penalty factor c and a kernel function with higher precision according to an improved goblet sea squirt algorithm, and optimizing and constructing a water quality prediction model of a Support Vector Machine (SVM) to obtain an optimized SVM model; the method comprises the following specific steps of:
step 1: initializing upper, lower, leader (leader) location, food source, c in algorithm 1 And c 2 Randomly initializing the value range of each parameter of the SVM according to the scale of the sea squirt and the maximum iteration number;
step 2, initializing population positions by using Fuch-Tent chaotic mapping, and leading a leader (leader)Setting as current position, calculating fitness value, taking out the +_f of the most suitable goblet sea squirt>As a food source +.>
Step 3, re-determining the position of the leader (leader) of the goblet sea squirt according to the iterative formula of the leader (leader) and determining a new fitness value, and determining the position of the follower (follower) according to the iterative formula of the follower (follower);
step 4, comparing the updated fitness value with the global optimum value of the goblet sea squirt, and updating the fitness value when the conditions are met;
and 5, judging whether the condition meets the requirement, if so, outputting an optimal value, otherwise, returning to the step 1.
And 6, outputting an optimal value until the condition meets the requirement, and finally outputting a better prediction result.
Wherein the improved ecteinascidia group algorithm comprises:
step a: introducing Fuch chaotic mapping into the Zun ecteinascidial algorithm, optimizing and improving the Zun ecteinascidial algorithm, initializing individual positions of Zun ecteinascidial population by using an improved Fuch-Tent chaotic mapping formula, and generating a new chaotic sequence so as to avoid the situation of local optimal values.
The Fuch-Tent chaotic mapping formula of the initialized population is as follows:
in the above, k n Not equal to 0, n ε Z+; alpha is 0.5.
The Fuch-Tent chaotic map well combines the advantages of the Fuch map and the Tent map, and the improved Fuch-Tent map has the advantages of insensitivity to an initial value, few control parameters, low complexity, traversing equalization, quicker convergence and the like, can generate chaos under the condition that the initial value is not 0, has smaller iteration times compared with the traditional limited folding chaotic map, and can better realize chaos optimizing.
Step b: a fitness value is calculated.
Step c: the sea squirt group is divided into a leader (leader) and a follower (follower), and the position of the leader (leader) is updated first, and then the position of the follower (follower) is updated.
Updating the calculation formula of the leader (leader), namely the iterative formula of the leader (leader) is as follows:
in the above-mentioned method, the step of,representing an upper bound and a lower bound, respectively, +.>Representing the location of the leader (leader),representing a pre-update food source, c 2 Is [0,1]The uniform random number of the (2) is used for strengthening the randomness of the update of the moving position of the leader and improving the searching capability; c 1 Represents convergence factor, and the value range is +.>
The calculation formula for updating the position of the follower, i.e. the iterative formula of the follower (follower) is:
wherein the follower's position is its own positionAnd the position of the preceding individual->At the midpoint of (2).
Step d: judging whether an ending condition is reached, and if so, obtaining an optimal value; if not, the step c is carried out again.
In order to verify the superiority of the scheme, the inventor selects test functions to be respectively input into the goblet-sea squirt algorithm which is not optimized by the Fuch-Tent chaotic mapping method and the goblet-sea squirt algorithm which is optimized by the Fuch-Tent chaotic mapping method and is provided by the scheme for testing, and the test scheme is as follows:
step one: selecting test functionThe number is-1.28≤x i Less than or equal to 1.28; step two: determining the population number as n=30, the iteration number as t=1000, and the ranges of the upper and lower bounds are: (lb, ub) = (-1.28,1.28);
step three: inputting the test function into a goblet-sea squirt algorithm which is not optimized by a Fuch-Tent chaotic mapping method, and recording each generation of output individual values;
step four: inputting the test function into a goblet sea squirt algorithm optimized by a Fuch-Tent chaotic mapping method, and recording an output individual value;
step five: comparing the individual values of the two times, and the results are shown in table 1;
TABLE 1
And (3) obtaining a conclusion of the merits and merits: as can be seen from Table 1, the goblet-sea squirt algorithm after Fuch-Tent chaotic optimization has better optimizing capability than the original algorithm.
The foregoing is merely exemplary embodiments of the present invention, but the scope of the present invention is not limited thereto, and any changes, substitutions and modifications within the technical scope of the present invention are all within the scope of the present invention.
Claims (7)
1. A water quality prediction method for optimizing SVM based on improved goblet sea squirt algorithm is characterized in that: the sensor is utilized to collect data of water quality in real time, and the data is uploaded to the background through the WIFI module for data processing; the real-time data of each parameter of the water quality is preprocessed and input into an optimized SVM prediction model, so that the water quality can be predicted for a period of time in the future; the specific operation steps comprise:
s1: acquiring a water quality sample in real time through a sensor, acquiring water quality data, and uploading the water quality data to a background through a WIFI module;
s2: after receiving the water quality data, the background performs standardized pretreatment on the water quality data in a conventional mode;
s3: taking the Gaussian radial basis function as a kernel function, and obtaining a classification decision function based on the Gaussian radial basis function and an objective function; constructing a support vector machine model according to the classification decision function and the kernel function;
s4: obtaining a penalty factor c and a kernel function with higher precision according to an improved goblet sea squirt algorithm, and optimizing and constructing a water quality prediction model of a Support Vector Machine (SVM) to obtain an optimized SVM model; the improved goblet sea squirt swarm algorithm includes:
step a: introducing Fuch chaotic mapping into a Zun ecteinascidia algorithm, optimizing and improving the algorithm, initializing individual positions of Zun ecteinascidia population by using an improved Fuch-Tent chaotic mapping formula, and generating a new chaotic sequence so as to avoid the situation of local optimal values;
step b: calculating a value of fitness;
step c: dividing the goblet sea squirt group into a leader (leader) and a follower (follower), updating the position of the leader (leader) first, and then updating the position of the follower (follower);
step d: judging whether an ending condition is reached, and if so, obtaining an optimal value; if not, the step c is carried out again.
2. The improved method for predicting the water quality of the SVM based on the Zun sea squirt algorithm according to claim 1, wherein the method comprises the following steps: the standardized preprocessing in the step S2 adopts mean variance standardized processing, and the calculation formula of the standardized preprocessing is as follows:
in the above expression, newX represents a normalized value, X represents water quality data, mean (X) represents an average value corresponding to a feature value, and std (X) represents a variance corresponding to the feature value.
3. The improved method for predicting the water quality of the SVM based on the Zun sea squirt algorithm according to claim 1, wherein the method comprises the following steps: the formula of the gaussian radial basis function in step S3 is:
wherein x represents any point in the kernel function; y represents a kernel function center point; x and y are water quality sample characteristic quantities; σ represents the kernel parameter.
4. The improved method for predicting the water quality of the SVM based on the Zun sea squirt algorithm according to claim 1, wherein the method comprises the following steps: the objective function formula and the constraint condition formula described in step S3 are:
wherein ω represents the normal vector of the hyperplane, δ i For the relaxation variable, C is a penalty factor, i represents the ith sample; x is X i Representing a sample feature quantity; y is i Representing a sample class;representing a nonlinear mapping function; b represents the deviation amount; omega Τ Representing the normal vector of the transposed hyperplane; and when the constraint condition is met, obtaining the classification decision function.
5. The improved method for predicting the water quality of the SVM based on the Zun sea squirt algorithm according to claim 1, wherein the method comprises the following steps: the Fuch-Tent chaotic mapping formula in the step a is as follows:
in the above, k n Not equal to 0, n ε Z+; alpha is 0.5.
6. The improved method for predicting the water quality of the SVM based on the Zun sea squirt algorithm according to claim 1 or 5, wherein the method comprises the following steps: in the step c, the calculation formula for updating the leader (leader) position, that is, the iterative formula of the leader (leader) is:
in the above-mentioned method, the step of,representing an upper bound and a lower bound, respectively, +.>Indicating the position of the leader (leader,)>Representing a pre-update food source, c 2 Is [0,1]The uniform random number of the (2) is used for strengthening the randomness of the update of the moving position of the leader and improving the searching capability; c 1 Represents convergence factor, and the value range is +.>
The calculation formula for updating the position of the follower, i.e. the iterative formula of the follower (follower) is:
wherein the follower's position is its own positionAnd the position of the preceding individual->At the midpoint of (2).
7. The improved method for predicting the water quality of the SVM based on the Zun sea squirt algorithm according to claim 4, wherein the method comprises the following steps: the step S4 is to obtain a punishment factor c and a kernel function with higher precision according to an improved sea squirt algorithm, and optimize and construct a water quality prediction model of the SVM, and the specific steps comprise:
step 1: initializing upper, lower, leader (leader) location, food source, c in algorithm 1 And c 2 Randomly initializing the value range of each parameter of the SVM according to the scale of the sea squirt and the maximum iteration number;
step 2, initializing population positions by using Fuch-Tent chaotic mapping, and leading a leader (leader)Setting as current position, calculating fitness value, taking out the +_f of the most suitable goblet sea squirt>As a food source +.>
Step 3, re-determining the position of the leader (leader) of the goblet sea squirt according to the iterative formula of the leader (leader) and determining a new fitness value, and determining the position of the follower (follower) according to the iterative formula of the follower (follower);
step 4, comparing the updated fitness value with the global optimum value of the goblet sea squirt, and updating the fitness value when the conditions are met;
and 5, judging whether the condition meets the requirement, if so, outputting an optimal value, otherwise, returning to the step 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310479905.0A CN116662893A (en) | 2023-04-28 | 2023-04-28 | Water quality prediction method for optimizing SVM (support vector machine) based on improved goblet sea squirt algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310479905.0A CN116662893A (en) | 2023-04-28 | 2023-04-28 | Water quality prediction method for optimizing SVM (support vector machine) based on improved goblet sea squirt algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116662893A true CN116662893A (en) | 2023-08-29 |
Family
ID=87716169
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310479905.0A Pending CN116662893A (en) | 2023-04-28 | 2023-04-28 | Water quality prediction method for optimizing SVM (support vector machine) based on improved goblet sea squirt algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116662893A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117471905A (en) * | 2023-09-21 | 2024-01-30 | 昆明理工大学 | Control method for improved pump storage unit regulating system of goblet sea squirt group |
-
2023
- 2023-04-28 CN CN202310479905.0A patent/CN116662893A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117471905A (en) * | 2023-09-21 | 2024-01-30 | 昆明理工大学 | Control method for improved pump storage unit regulating system of goblet sea squirt group |
| CN117471905B (en) * | 2023-09-21 | 2024-04-12 | 昆明理工大学 | Control method for improved pump storage unit regulating system of goblet sea squirt group |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111860982A (en) | Wind power plant short-term wind power prediction method based on VMD-FCM-GRU | |
| CN109886464B (en) | Low-information-loss short-term wind speed prediction method based on optimized singular value decomposition generated feature set | |
| CN116842459B (en) | Electric energy metering fault diagnosis method and diagnosis terminal based on small sample learning | |
| Yuan et al. | Evoq: Mixed precision quantization of dnns via sensitivity guided evolutionary search | |
| CN111599409B (en) | circRNA recognition method based on MapReduce parallelism | |
| CN112086144B (en) | Molecule generation method, device, electronic equipment and storage medium | |
| CN110020712A (en) | A kind of optimization population BP neural network forecast method and system based on cluster | |
| CN113125960A (en) | Vehicle-mounted lithium ion battery charge state prediction method based on random forest model | |
| CN115329908A (en) | Power transformer fault diagnosis method based on deep learning | |
| CN112765902B (en) | Soft measurement modeling method for COD concentration in rural domestic sewage treatment process based on TentFWA-GD RBF neural network | |
| CN116306226B (en) | Fuel cell performance degradation prediction method | |
| CN113539386A (en) | CLMVO-ELM-based dissolved oxygen concentration prediction method, device, equipment and storage medium | |
| CN116662893A (en) | Water quality prediction method for optimizing SVM (support vector machine) based on improved goblet sea squirt algorithm | |
| CN118151020B (en) | Method and system for detecting safety performance of battery | |
| CN114564410A (en) | Software defect prediction method based on class level source code similarity | |
| CN118011211B (en) | Lithium battery state of charge and energy state estimation method | |
| CN113408622A (en) | Non-invasive load identification method and system considering characteristic quantity information expression difference | |
| CN115394381B (en) | High-entropy alloy hardness prediction method and device based on machine learning and two-step data expansion | |
| CN111310974A (en) | Short-term water demand prediction method based on GA-ELM | |
| CN117290673A (en) | Ship energy consumption high-precision prediction system based on multi-model fusion | |
| Xu et al. | Change point detection of multimode processes considering both mode transitions and parameter changes | |
| CN113139624A (en) | Network user classification method based on machine learning | |
| CN118070685B (en) | Optimization method and system for stealth power station design | |
| CN118378178B (en) | Transformer fault identification method and system based on residual map convolution neural network | |
| CN118469045B (en) | Permafrost upper limit prediction method, device, equipment and storage medium |
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 |