PID (Proportion Integration Differentiation)-based support vector machine regression monitoring model for concentration of VOCs (Volatile Organic Compounds) in water-soil-gas

Abstract

The invention discloses a PID (Proportion Integration Differentiation)-based support vector machine regression monitoring model for VOCs (Volatile Organic Compounds) concentration in water-soil-gas, which comprises the following steps of: performing characteristic parameter extraction on a PID signal according to acquired PID signal data; performing principal component extraction on the characteristic parameters by adopting a principal component analysis method; normalizing a training set and a test set in the process of performing VOCs concentration quantitative analysis by using support vector machine regression; selecting a Gaussian radial basis kernel function as a kernel function of SVM regression, and seeking an optimal parameter delta and an optimal penalty factor C of the radial basis kernel function through a genetic algorithm; and carrying out model construction on PID signals under the condition of different concentrations of VOCs by applying a support vector machine regression method on the optimal parameter delta and the optimal penalty factor C. According to the method, the precision of analyzing the concentration of VOCs is improved, so that the prediction effect on the concentration of VOCs is improved, and the monitoring precision of VOCs is improved.

Description

Technical Field

[0001] The present invention relates to the field of VOCs detection, and in particular to a PID-based support vector machine regression monitoring model for VOCs concentration in water, soil and air. Background Art

[0002] Volatile organic compounds (VOCs) are a class of organic chemicals that are easily volatile at room temperature and pressure. They are widely present in various fields such as industrial production, transportation, and daily life. Due to their toxicity, carcinogenicity, and environmental persistence, VOCs pollution has become a global environmental problem. With the acceleration of my country's industrialization process, VOCs emissions have continued to increase, causing serious complex pollution to water bodies, soil, and the atmospheric environment.

[0003] VOCs in mining engineering exist throughout the entire life cycle of mining, ore processing and tailings disposal. VOCs emissions are multi-source and complex. They are produced through diesel equipment exhaust, blasting reactions, chemical volatilization and natural release of ores during mining, transportation, mineral processing and smelting. For example, diesel-powered equipment (such as mining machines and transport trucks) releases benzene series (BTEX), polycyclic aromatic hydrocarbons (PAHs) and aldehydes (such as formaldehyde and acrolein) during high-temperature combustion; organic solvents (such as kerosene and pine oil) volatilize in mineral processing processes such as flotation and leaching, and the secondary release of VOCs adsorbed by dust during ore crushing; the oxidative decomposition of sulfide minerals and the anaerobic degradation of organic matter in tailings ponds produce characteristic VOCs such as mercaptans and sulfides; chlorinated hydrocarbons and ester compounds volatilized from equipment lubricants, hydraulic oil leakage and anti-corrosion coatings. The types and concentrations of VOCs are significantly affected by the type of ore, mining method (open-pit / underground) and environmental conditions (temperature, humidity), and exist in multiple media such as water, soil and air. VOCs, as precursors of ozone (O3) and secondary organic aerosols (SOA), are important causes of haze and photochemical smog. Long-term exposure to VOCs can lead to respiratory diseases, nervous system damage and cancer risks (such as benzene is classified as a Class I carcinogen by the International Agency for Research on Cancer). VOCs can also enter the soil and water through dry and wet deposition, inhibit microbial activity and accumulate along the food chain, leading to the degradation of ecological functions in mining areas. Therefore, it is crucial to strengthen the monitoring of VOCs in mines, clarify the sources and take timely measures to control them. This has important theoretical value and practical significance for achieving green mine construction and sustainable development goals.

[0004] The traditional support vector machine regression method was used to construct a model of PID signals under different VOC concentration conditions. It was found that the VOC concentrations predicted by the model had a large error compared to the actual VOC concentrations. The reasons for the large error in the model were analyzed. The first reason was the uncertainty of the penalty factor C and the kernel function parameter δ, which made it difficult to manually determine the optimal values ​​of these two parameters. The second reason was that the original data used to train the model was redundant, and useful information could not be well identified. In turn, useless information would affect the accuracy of the model construction. As a result, the traditional support vector machine regression method was unable to accurately analyze the accuracy of VOC concentrations. Therefore, a new support vector machine regression monitoring model was urgently needed to improve the accuracy of VOC concentration analysis. Summary of the Invention

[0005] In response to the problems existing in the above-mentioned prior art, the present invention provides a PID-based support vector machine regression monitoring model for VOCs concentration in water, soil and air, which effectively improves the accuracy of support vector machine regression quantitative analysis of VOCs concentration.

[0006] To achieve the above objectives, the present invention adopts a technical solution: a PID-based support vector machine regression monitoring model for VOCs concentration in water, soil and air, comprising:

[0007] Based on the collected PID signal data, the characteristic parameters of the PID signal are extracted from the time domain and frequency domain;

[0008] The principal component analysis method is used to extract the principal components of the characteristic parameters;

[0009] Normalize the training set and test set in the process of quantitative analysis of VOCs concentration using support vector machine regression;

[0010] The Gaussian radial basis kernel function is selected as the kernel function of SVM regression, and the optimal parameters and penalty factor C of the radial basis kernel function are sought through genetic algorithm.

[0011] The support vector machine regression method is used with the optimal parameters and the optimal penalty factor C to construct a model of the PID signal under different VOCs concentrations.

[0012] Furthermore, in the process of quantitative analysis of VOCs concentration by support vector machine regression, in order to prevent the prediction model from overfitting, the data set needs to be reasonably divided. For small sample data, the data is divided into a training set (model construction) and a test set (performance evaluation) in a ratio of 8:2.

[0013] Furthermore, the amplitude analysis method in the time domain analysis method is used to obtain the time domain characteristics of the PID signal. The time domain characteristics of the signal are represented by 12 characteristic parameters, including the mean value, standard deviation, skewness, kurtosis, maximum value, minimum value, peak-to-peak value, root mean square, crest factor, form factor, impulse factor, and margin factor of the signal generated by PID at different VOCs concentrations.

[0014] The signal frequency domain analysis method is used to obtain the frequency domain characteristics of the PID signal, and four types of frequency domain characteristics are selected: average frequency, center of gravity frequency, mean square frequency, and frequency standard deviation.

[0015] Furthermore, the time domain characteristics and frequency domain characteristics of the PID response signal to VOCs were used as the original data and principal component analysis was used to extract the principal components. The nine features were: mean value, mean frequency, center of gravity frequency, root mean square frequency, standard deviation of frequency, standard deviation, skewness, and kurtosis.

[0016] Furthermore, the search space and global search capabilities of the genetic algorithm are utilized to automatically select the optimal penalty factor C and the optimal kernel function parameters during SVM regression, thereby establishing a genetic algorithm-optimized SVM regression model and determining the VOCs concentration through PID signal generation.

[0017] Furthermore, in the genetic algorithm, the larger the chromosome fitness value, the better the individual represented by the chromosome. After initialization, the optimal target and the optimal chromosome are found, and then iteration begins, operator selection, calculation of the current target fitness, calculation of the fitness ratio of each chromosome, and cyclic generation of non-zero random numbers for the population. Starting from the first chromosome, the fitness ratio is accumulated. When the accumulated value is greater than the random number, the last accumulated individual is selected as the individual we choose.

[0018] Select offspring of the same size as the original population, calculate the optimized variable dimension and population size, then determine the crossover probability, generate a random number and compare it with the crossover probability to see if there is a crossover. If so, randomly select two different chromosomes to connect, otherwise the crossover is meaningless;

[0019] We export the two chromosomes we selected, loop through each dimension of the chromosome, simulate binary crossover with the crossover operator, and then limit the boundaries of the individuals after the crossover, replacing the values ​​exceeding the maximum value with the maximum value and those below the minimum value with the minimum value;

[0020] Copy the individuals after crossover into the population, then loop through the population, generate a random number and compare it with the crossover mutation probability. If there is a mutation, randomly select an individual to prepare for the mutation.

[0021] Loop through each gene of the chromosome, perform polynomial mutation on the selected chromosome, then copy the mutated individuals into the population, and recalculate the targets of the offspring after crossover mutation to obtain the optimal and worst targets;

[0022] Then compare the current optimal goal with the global optimal goal. Replacing the worst with the historical global optimal goal can increase the probability of the population iterating out better individuals. Record the average goal and optimal goal of this generation. After the iteration, derive the value of the optimal penalty factor C and the optimal kernel function parameter δ.

[0023] Furthermore, the monitoring PID signal of VOCs obtained in a humidity environment needs to be denoised by adaptive threshold weighted wavelet packet.

[0024] The beneficial effects of the present invention are:

[0025] (1) The VOCs concentration is judged by the time domain and frequency domain characteristics of the PID signal. The feature extraction method combining the time domain and frequency domain can effectively improve the accuracy of the support vector machine regression quantitative analysis of VOCs concentration;

[0026] (2) By utilizing the search space and global search capabilities of the genetic algorithm, the penalty factor C and the kernel function parameter δ are automatically selected during SVM regression, thereby establishing a genetic algorithm-optimized SVM regression model, and determining the VOCs concentration through PID-generated signals, thereby further improving the accuracy of VOCs concentration analysis;

[0027] (3) The invention improves the accuracy of VOCs concentration analysis, thereby improving the prediction effect of VOCs concentration and improving the monitoring accuracy of VOCs. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 The effect diagram of the VOCs concentration quantitative analysis model established by traditional support vector machine regression;

[0029] Figure 2 This is the effect diagram of the support vector machine regression VOCs concentration quantitative analysis model based on signal time domain and frequency domain feature extraction;

[0030] Figure 3 This is the effect diagram of the support vector machine regression VOCs concentration quantitative analysis model based on principal component analysis of signal time domain and frequency domain features;

[0031] Figure 4 This is a graph showing the trend of prediction error changing with evolutionary generations;

[0032] Figure 5 This is the effect diagram of the VOCs concentration quantitative analysis model based on signal time domain and frequency domain feature extraction, principal component analysis, and genetic algorithm optimized support vector machine regression;

[0033] Figure 6 The accuracy graph of the PCA-GA-SVM model under different training samples. DETAILED DESCRIPTION

[0034] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. However, it should be understood that the specific embodiments described herein are only used to illustrate the present invention and are not intended to limit the scope of the present invention.

[0035] Unless otherwise defined, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention pertains. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0036] Support Vector Machine Regression Algorithm

[0037] Based on the characteristic that support vector machine regression can realize the construction of regression model for small sample data, support vector machine is used to construct a regression model for PID response VOCs concentration data; when using support vector machine regression for quantitative analysis of VOCs concentration, first, 84 groups of different VOCs concentration data are randomly arranged, 67 groups of data are used as training sets, and 17 groups of data are used as test sets. Each group of data has 6001 values, of which the first 6000 values ​​are the response values ​​of PID to VOCs concentration, and the 6001th value is the actual concentration of VOCs.

[0038] First, the training set and test set must be normalized to eliminate the adverse effects caused by singular sample data. For SVM, if data normalization is not used to make the features dimensionless, the distribution of features will be elliptical, which will affect the model prediction accuracy during model training and may even cause the model training to enter an infinite loop and fail to converge.

[0039] It is necessary to normalize the data first, and uniformly map the sample data to [0,1] to achieve uniform data parameter scale. The calculation formula is shown in (1);

[0040]

[0041] Where X max is the maximum value of the sample data; X min is the minimum value of the sample data.

[0042] Then set the support vector machine regression parameters and select the Gaussian radial basis kernel function as the kernel function of SVM regression; try the effects of different kernel function parameters δ and penalty factor C on the SVM regression accuracy. Since there is no prior knowledge, first set the penalty factor C to 0.1 and the kernel function parameter δ to 1 temporarily. At this time, the R 2 When R is 0.21 and C is 0.2 2 When R is 0.13 and C is 0.3 2 is 0.015, the model accuracy is not high, and C is set to 1 again. At this time, R 2 When R is 0.30 and C is 2 2 When R is 0.30 and C is 3 2 When R is 0.30 and C is 4 2 When C is 5, R 2 It is 0.29. Since it is difficult to find a mathematical relationship between the value of C and the model regression accuracy, the penalty factor C of SVM regression is temporarily set to 4. Keeping C unchanged, the kernel function parameter δ is obtained to be 0.8 using the same method.

[0043] Finally, the traditional support vector machine regression method was used to construct a model for PID signals under 84 groups of VOCs with different concentrations when the penalty factor C was 4 and the kernel function parameter δ was 0.8. The test set results of the model are shown in Figure 1 As shown in the figure, the model has a mean square error of 69458.6 and an R² of 0.38, indicating a large discrepancy between the model's predicted VOCs concentration and the actual VOCs concentration. Analysis of the reasons for this large error reveals that the first reason is the uncertainty of the penalty factor C and the kernel function parameter δ, making it difficult to manually determine the optimal values ​​for these two parameters. The second reason is that the raw data used to train the model is redundant, meaning useful information cannot be well identified, while useless information can affect the accuracy of the model. Therefore, the model will be optimized from these two aspects.

[0044] PID signal feature extraction

[0045] The sampling interval of PID monitoring for VOCs is 0.02s / time. Each of the 84 groups of VOCs data with different concentrations consists of 6000 sampling points. Figure 1 It can be seen that each set of data consisting of 6000 sampling points contains a large amount of redundant information, which causes the SVM regression process to take a long time and the SVM regression model has low accuracy; therefore, it is very necessary to extract the characteristic parameters of the data generated by PID.

[0046] The collected PID signal data contains a wealth of useful information reflecting VOC concentrations. These features are often expressed as time-domain waveforms. However, in practice, using time-domain waveforms cannot accurately represent the VOC concentrations reflected by PID. If noise interferes with the PID signal acquisition process, the accuracy of determining VOC concentrations using time-domain characteristic parameters will be compromised. Therefore, to more comprehensively extract the characteristic parameters of the PID response to VOCs, it is necessary to extract characteristic parameters from multiple angles of the PID signal; extracting characteristic parameters from both the time and frequency domains, and determining VOC concentrations based on the PID signal's time-domain and frequency-domain characteristics. Experimental results show that the combined time-domain and frequency-domain feature extraction method can effectively improve the accuracy of support vector machine regression quantitative analysis of VOC concentrations.

[0047] (1) PID signal time domain feature extraction

[0048] The signal time domain characteristic parameter analysis method is a way to judge the VOCs concentration by analyzing the signal waveform information generated by PID; the time domain characteristic analysis method is the earliest feature analysis method, it is very mature, and its application range is wide and simple and clear, so the time domain analysis method is widely used in signal feature extraction; in order to realize the extraction of the characteristic quantity of the PID response signal with different VOCs concentrations, the amplitude analysis method in the time domain analysis method is used to obtain the time domain characteristic quantity in the PID signal.

[0049] The time domain characteristics of the signal are represented by 12 characteristic parameters, including the mean value, standard deviation, skewness, kurtosis, maximum value, minimum value, peak-to-peak value, root mean square, amplitude factor, waveform factor, impact factor, and margin factor of the signal generated by PID at different VOCs concentrations.

[0050] The time domain characteristics of the PID signal include: mean value, standard deviation, skewness, kurtosis, maximum value, minimum value, peak-to-peak value. The calculation method of the characteristic parameters is shown in the table below; iis the amplitude of the sampling data point of the signal, N is the number of sampling data points for each sample, 6000; the mean is the average of the signal; the standard deviation is also the mean square error, which is the arithmetic square root of the variance. The standard deviation reflects the degree of dispersion of the PID signal; skewness: used to measure the asymmetry of the probability distribution of random variables, used to describe the distribution of the PID generated signal; skewness is a measure of the direction and degree of skewness of the statistical data distribution, and is a digital feature of the degree of asymmetry of the statistical data distribution; if the skewness value is between -0.5 and 0.5, then the data is quite symmetrical, that is, normally distributed. If the skewness value is between -1 and -0.5 (negative skewness) or 0.5 and 1 (positive skewness), the data is tilted. If the skewness is less than -1 (negative skewness) or greater than 1 (positive skewness), the data is highly tilted; the maximum and minimum values ​​are the maximum and minimum amplitudes of the one-dimensional signal respectively; kurtosis: kurtosis is a numerical statistic that reflects the distribution characteristics of random variables. It is used to describe the flatness of the time domain waveform. It is a fourth-order cumulative amount and reflects the degree of data dispersion; peak-to-peak value: the difference between the maximum and minimum values, which reflects the fluctuation range of the signal and describes the peak degree of the waveform; 7 time domains.

[0051] Table 1 Calculation methods for mean, standard deviation, skewness, kurtosis, maximum, minimum, and peak-to-peak values

[0052]

[0053] The average value, standard deviation, skewness, kurtosis, maximum value, minimum value, and peak-to-peak value of PID response to 84 groups of VOCs signals with different concentrations are shown in the following table:

[0054] Table 2 Average, standard deviation, skewness, kurtosis, maximum, minimum, and peak-to-peak values ​​of PID response signals for 84 groups of VOCs with different concentrations

[0055]

[0056]

[0057]

[0058]

[0059]

[0060] The root mean square, amplitude factor, waveform factor, impact factor, and margin factor of PID responses to 84 groups of VOCs signals with different concentrations are shown in Tables 3, 4, 5, and 6:

[0061] Table 3 RMS, amplitude factor, form factor, impact factor, and margin factor of PID response signals for different concentrations of VOCs in groups 1-17

[0062]

[0063] Table 4 RMS, amplitude factor, form factor, and impact factor of PID response signals for different concentrations of VOCs in groups 18-81

[0064]

[0065]

[0066] Table 5 Margin factors of PID response signals to different concentrations of VOCs in groups 18-81

[0067]

[0068] Table 6 PID response signals for different concentrations of VOCs in groups 82-84: RMS, amplitude factor, form factor, impact factor, and margin factor

[0069]

[0070] (2) PID signal frequency domain feature extraction

[0071] The signal frequency domain analysis method and its application in identifying and analyzing PID response signals under different VOCs concentrations were studied; the frequency domain analysis method realizes the conversion from time domain to frequency domain through Fourier transform, which can intuitively describe the size of different frequency components of the signal, thereby better characterizing the parameters and characteristics of the signal; four frequency domain feature quantities, namely average frequency, center of gravity frequency (FC), mean square frequency (MSF), and standard deviation of frequency (VF), were selected to analyze PID signals.

[0072] The center of gravity frequency (FC) is the center of gravity of the area under the power spectrum curve in the PID signal. The horizontal axis is the frequency, so it is called the center of gravity frequency. The specific formula is as follows:

[0073]

[0074] MSF: used to reflect the energy intensity of the signal spectrum.

[0075]

[0076] VF: Used to reflect the degree of deviation between the PID signal frequency amplitude and its center of gravity.

[0077]

[0078] The four frequency domain characteristic parameters of PID responses to 84 groups of signals with different VOCs concentrations are shown in Table 7;

[0079] Table 7 Average frequency, center of gravity frequency, mean square frequency, and frequency variance of PID response signals for 84 groups of VOCs with different concentrations

[0080]

[0081]

[0082]

[0083] (3) SVM regression based on PID signal feature extraction

[0084] In the SVM regression method based on the extraction of time domain and frequency domain characteristic parameters of PID signals, first, 84 different groups of PIDs are randomly arranged to generate 12 time domain features and 4 frequency domain features that reflect VOCs concentration signals. 67 groups of signal time domain and frequency domain characteristic data are used as training sets, and 17 groups of signal time domain and frequency domain characteristic data are used as test sets. Each group of data has 17 values, of which the first 16 values ​​are the characteristic parameters of the PID response signal to VOCs concentration, and the 17th value is the true concentration of VOCs. The training set and test set are then normalized, and then the support vector machine parameters are set, and the Gaussian radial basis kernel function is selected as the kernel function of the SVM regression. The radial basis kernel function parameter is still set to 0.8, and the penalty factor C of the SVM regression is set to 4. The final test set results are as follows: Figure 2 As shown. The mean square error of the model is 1122.25, R 2 is 0.9899, ​​and the model accuracy is Figure 2 The regression effect of this model has certain errors, so the model will be optimized later.

[0085] SVM regression based on principal component analysis of PID signal characteristics

[0086] 1) Principal component analysis of PID signal time domain and frequency domain characteristics

[0087] Principal components analysis (PCA) is an effective and widely used dimensionality reduction algorithm. The decomposed principal components are mutually orthogonal, which can effectively eliminate redundant and overlapping parts of the original data. Generally speaking, a few important principal components can cover most of the information of the PID response VOCs signal. The calculation process of the PCA algorithm is as follows:

[0088] (1) For a sample set with dimension d and O i =(O1,O2,.....,O n ) is processed by zero mean:

[0089]

[0090] (2) Find the covariance matrix ∑ of O;

[0091] (3) The singular value decomposition method is used to obtain the eigenvalues ​​and eigenvectors of the covariance matrix Σ;

[0092] (4) Take the eigenvectors corresponding to the first v eigenvalues ​​to form a new matrix. In this case, v should be less than n.

[0093] (5) Obtain a new low-dimensional sample set;

[0094] (6) Calculate the principal component contribution rate and cumulative contribution rate.

[0095] The time domain characteristics and frequency domain characteristics of the PID response signal to VOCs are used as the original data to extract the principal components using the principal component analysis method;

[0096] The contribution rates of the mean, mean frequency, center of gravity frequency, frequency root mean square, frequency standard deviation, standard deviation, skewness, kurtosis, and maximum value are 0.4495, 0.2439, 0.1082, 0.0787, 0.0557, 0.0300, 0.0152, 0.0080, and 0.0065, respectively. The cumulative contribution rate of the nine features is 0.9958, so these nine features are used to replace the 16 features of the signal.

[0097] 2) SVM regression after principal component analysis of PID signal characteristics

[0098] In the SVM regression method based on principal component analysis of PID signal time domain and frequency domain extraction, 12 time domain features and 4 frequency domain features reflecting VOCs concentration signals generated by 84 different PIDs were subjected to principal component analysis. Feature data of 67 groups of signals after principal component analysis were used as training sets, and feature data of 17 groups of signals after principal component analysis were used as test sets. Each set of data had 10 values, of which the first 9 values ​​were feature parameters after principal component analysis of PID response signals to VOCs concentrations, and the 10th value was the true concentration of VOCs. The training set and test set were then normalized, and then the support vector machine parameters were set, and the Gaussian radial basis kernel function was selected as the kernel function of the SVM regression. The radial basis kernel function parameter was set to 0.8, and the penalty factor C of the SVM regression was set to 4; the final test set results are as follows: Figure 3 As shown; the mean square error of the model is 372, R 2 The regression effect of the model is still a certain error, so the model will be optimized in the future.

[0099] Currently, there is no theoretical guidance for selecting support vector machine parameters. Therefore, when constructing a traditional SVM regression model, the optimal SVM parameters are usually selected through repeated experiments, and the parameters of the regression model with the highest accuracy obtained through experiments are used as the optimal parameters of the SVM regression model. For example, the existing data set is divided into a training set and a test set, and then the parameters of the SVM regression model are changed and recorded, and finally the parameters are confirmed by the optimal regression model. However, when selecting parameters using this method, the parameter values ​​are easily affected by people's subjective experience and it takes a lot of time. By utilizing the search space and global search capabilities of the genetic algorithm, the values ​​of the penalty factor C and the kernel function parameter δ during SVM regression are automatically selected, thereby establishing a genetic algorithm-optimized SVM regression model, and determining the VOCs concentration by generating a signal through PID.

[0100] 1) The impact of support vector machine parameters on its performance

[0101] When using a support vector machine (SVM) to build a regression model for quantitative analysis of PID signals to determine VOCs, its modeling accuracy is influenced by the kernel function parameter δ and the penalty factor parameter C. The value of the kernel function parameter δ affects the SVM's ability to predict the 17 test data sets using the learning machine derived from the 67 data sets in the training set, thereby affecting the generalizability of the SVM regression. The penalty factor parameter C balances the smoothness of the regression curve and the empirical risk in the feature space, thus affecting the robustness of the regression model. Therefore, when building a regression model, it is necessary to comprehensively consider the values ​​of both parameters δ and C to achieve optimal VOC concentration prediction.

[0102] 2) Principal Component Analysis-Genetic Algorithm-SVM Regression

[0103] When applying genetic algorithms to solve problems, there are two ways to encode chromosomes: binary encoding and floating-point encoding. Floating-point chromosome encoding is suitable for solving problems with a large value range, while binary encoding is suitable for solving problems with a smaller value range. Since this paper uses genetic algorithms to optimize SVM regression to process small sample data, binary encoding is adopted. By studying the setting values ​​of parameters in support vector machine regression at home and abroad, the minimum value of the penalty factor and kernel function parameter is determined to be 0.001, the maximum value of the penalty factor is set to 100, and the maximum value of the kernel function parameter is set to 10.

[0104] Genetic algorithms operate on a population basis. An initial population is required for iteration, and subsequent iterations are based on this population. Based on domestic and international research on genetic algorithm optimization of support vector machine parameters, the population size is set to 20. Experimental results indicate that the iteration curve converges when the number of iterations is less than 50, so the number of iterations is set to 50. Population information is defined as a structure, and then the population cycle begins. The encoding forms a chromosome, and for each gene in the chromosome, a random number between its minimum and maximum values ​​is generated. This is the chromosome. The penalty factor is represented by the first number on the chromosome, and the kernel function parameter is represented by the second number on the chromosome.

[0105] Genetic algorithms distinguish between superior and inferior individuals by evaluating the fitness function value of each chromosome. In genetic algorithms, the larger the chromosome's fitness value, the better the individual represented by that chromosome. After initialization, the optimal target and chromosome are found. Then, iteration begins, selecting an operator, calculating the current target's fitness, calculating the fitness contribution of each chromosome, and cyclically generating non-zero random numbers for the population. Starting with the first chromosome, the fitness contribution is accumulated. When the accumulated fitness contribution is greater than the random number, the last individual accumulated is selected as the selected individual. A progeny of the same size as the original population is selected. The optimization variable dimension and population size are calculated. The crossover probability is then determined. A random number is generated and compared to the crossover probability to determine if crossover occurs. If so, two consecutive chromosomes are randomly selected; otherwise, crossover is meaningless. The two selected chromosomes are derived and each dimension of the chromosome is iterated over. The crossover operator simulates binary crossover. The bounds of the individuals after crossover are then set: values ​​exceeding the maximum value are replaced with the maximum value, and values ​​below the minimum value are replaced with the minimum value. The individuals that have completed the crossover are copied into the population, and then the population is looped to generate a random number and compare it with the crossover mutation probability. If a mutation occurs, a random individual is selected to prepare for the mutation. Each gene of the chromosome is looped, and the selected chromosome is subjected to polynomial mutation. Then the mutated individuals are copied into the population, and the targets of the offspring that have completed the crossover mutation are recalculated to obtain the optimal target and the worst target. Then the current optimal target is compared with the global optimal target. Replacing the worst target with the historical global optimal target can increase the probability of the population iterating to produce a better individual. The average target and optimal target of this generation are recorded. After the iteration, C and δ are derived. The iterative curve of the genetic algorithm to optimize the SVM regression parameters is as follows. Figure 4 As shown in Figure 1, the iterative curve has converged when the number of iterations is 16, that is, the minimum value of MSE can be obtained after the 16th iteration.

[0106] In the genetic algorithm optimized SVM regression method based on principal component analysis extracted from time domain and frequency domain of PID signals, 9 principal component features reflecting VOCs concentration signals generated by 84 different PIDs were used as signal features, 67 groups of characteristic data after principal component analysis of signals were used as training sets, and 17 groups of characteristic data after principal component analysis of signals were used as test sets. Each group of data had 10 values, of which the first 9 values ​​were characteristic parameters after principal component analysis of the PID response signal to VOCs concentration, and the 10th value was the actual concentration of VOCs.

[0107] The training set and test set are normalized, and the optimal parameters of the radial basis kernel function are found to be 0.0101 and the optimal penalty factor C is 7.8783 through the genetic algorithm. Based on the selection of parameters, the support vector machine regression function F(x) = w*x+b is derived. The final test set results are as follows: Figure 5 As shown; the mean square error of the model is 0.000059, R 2 is 0.9999; compared with the R obtained by Wang Jin 2 is 99.8%, Li Hai obtained R 2 The results are 98.2%, 98.5%, 96.9%, etc. This model method is superior.

[0108] 3) The impact of sample number on PCA-GA-SVM model

[0109] In order to verify the validity of the model with high accuracy of 84 samples and to find the minimum number of samples for PCA-GA-SVM model construction, on the basis of 67 training sets and 17 test sets, 4 training sets and 1 test set were reduced in turn to find the minimum number of samples for model construction; A is the original data; B is the data of 4 training sets and 1 test set; C is the data of 8 training sets and 2 test sets; D is the data of 12 training sets and 3 test sets; E is the data of 16 training sets and 4 test sets; F is the data of 20 training sets and 5 test sets; the accuracy results of PCA-GA-SVM model are shown in the figure. Figure 6 shown.

[0110] When reducing the data to 12 training sets and 3 test sets, the R 2 It can still be maintained above 0.99. When the training set and test set are reduced again, the accuracy of the PCA-GA-SVM model will drop sharply. Therefore, when the PCA-GA-SVM regression model is used to calculate the VOCs concentration based on the PID signal, the number of samples should be greater than 69 groups.

[0111] The PID response to VOCs signals is 84 groups, so it meets the requirements of model establishment.

[0112] 4) Application of PCA-GA-SVM model before and after denoising in humidity environment

[0113] In order to verify the effectiveness of the wavelet packet decomposition node energy adaptive weight method for denoising VOCs signals collected in a high humidity environment, a wet cotton ball was placed at the PID air inlet to affect the PID signal and generate noise. Twelve groups of benzene gases with different concentrations were prepared using a gas distribution device. The accuracy of the PCA-GA-SVM model for quantitative analysis of VOCs in a humidity environment was compared with that without the denoising method. Among them, nine features, including mean value, standard deviation, skewness, kurtosis, maximum value, frequency mean, center of gravity frequency, frequency root mean square, and frequency standard deviation, were determined as the principal components of the time and frequency characteristic parameters of the PID signal. The 12 groups of noisy PID response VOCs signals each had nine characteristic parameters in the time domain and frequency domain.

[0114] The 9 characteristic parameters of the 12 groups of noisy signals and the 12 groups of denoised signals were respectively imported into the PCA-GA-SVM formula, and the concentrations of the noisy and denoised VOCs were obtained as shown in Tables 8 and 9.

[0115] Table 8 Comparison of calculations of 10ppm-400ppm VOCs noisy signal and denoised signal by PCA-GA-SVM model

[0116]

[0117] Table 9 Comparison of calculations of 401ppm-1000ppm noisy signal and denoised signal by PCA-GA-SVM model

[0118]

[0119] The data in the table above shows that humidity has the most serious impact on the accuracy of low-concentration VOC monitoring. The PCA-GA-SVM model achieves a maximum VOC monitoring accuracy of 99.5% in humid environments. Denoising the PID signal using adaptive threshold-weighted wavelet packets improves the accuracy by up to 14.9%. This validates the PCA-GA-SVM model's applicability to VOC monitoring in humid environments. It also demonstrates that the constructed denoising method, based on wavelet packet decomposition node energy adaptive weighting, can improve VOC monitoring accuracy while simultaneously increasing the signal-to-noise ratio.

[0120] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions or improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A PID-based support vector machine regression monitoring model for VOCs concentration in water, soil and air, characterized by: include: Based on the collected PID signal data, the characteristic parameters of the PID signal are extracted from the time domain and frequency domain; The principal component analysis method is used to extract the principal components of the characteristic parameters; Normalize the training set and test set in the process of quantitative analysis of VOCs concentration using support vector machine regression; The Gaussian radial basis kernel function is selected as the kernel function of SVM regression, and the optimal parameter δ and the optimal penalty factor C of the radial basis kernel function are sought through genetic algorithm. The support vector machine regression method is used with the optimal parameter δ and the optimal penalty factor C to construct a model of PID signals under different VOCs concentrations.

2. The PID-based support vector machine regression monitoring model for VOCs concentration in water, soil and air according to claim 1 is characterized in that: In the process of quantitative analysis of VOCs concentration using support vector machine regression, in order to prevent the prediction model from overfitting, the data set needs to be reasonably divided. For small sample data, the data is divided into a training set (model construction) and a test set (performance evaluation) in a ratio of 8:

2.

3. The PID-based support vector machine regression monitoring model for VOCs concentration in water, soil and air according to claim 1 is characterized in that: The amplitude analysis method in the time domain analysis method is used to obtain the time domain characteristics of the PID signal. The time domain characteristics of the signal are represented by 12 characteristic parameters, including the average value, standard deviation, skewness, kurtosis, maximum value, minimum value, peak-to-peak value, root mean square, crest factor, form factor, impulse factor, and margin factor of the signal generated by PID at different VOCs concentrations. The signal frequency domain analysis method is used to obtain the frequency domain characteristics of the PID signal, and four types of frequency domain characteristics are selected: average frequency, center of gravity frequency, mean square frequency, and frequency standard deviation.

4. The PID-based support vector machine regression monitoring model for VOCs concentration in water, soil and air according to claim 1 is characterized in that: The time domain characteristics and frequency domain characteristics of the PID response signal to VOCs are used as the original data and the principal component analysis method is used to extract the principal components. The nine features are: average value, average frequency, center of gravity frequency, root mean square frequency, standard deviation of frequency, standard deviation, skewness, and kurtosis.

5. The PID-based support vector machine regression monitoring model for VOCs concentration in water, soil and air according to claim 1 is characterized in that: By utilizing the search space and global search capabilities of the genetic algorithm, the optimal penalty factor C and the optimal kernel function parameter δ are automatically selected during SVM regression, thereby establishing a genetic algorithm-optimized SVM regression model and determining the VOCs concentration through PID signal generation.

6. A PID-based support vector machine regression monitoring model for VOCs concentration in water, soil and air according to claim 1 or 5, characterized in that: In the genetic algorithm, the larger the chromosome fitness value, the better the individual represented by the chromosome. After initialization, the optimal target and the optimal chromosome are found, and then iteration begins, operator selection, calculation of the current target fitness, calculation of the fitness ratio of each chromosome, and cyclic generation of non-zero random numbers for the population. Starting from the first chromosome, accumulation and accumulation of fitness ratios are started. When the accumulation is greater than the random number, the last accumulated individual is selected as the individual we choose. Select offspring of the same size as the original population, calculate the optimized variable dimension and population size, then determine the crossover probability, generate a random number and compare it with the crossover probability to see if there is a crossover. If so, randomly select two different chromosomes to connect, otherwise the crossover is meaningless; We export the two chromosomes we selected, loop through each dimension of the chromosome, simulate binary crossover with the crossover operator, and then limit the boundaries of the individuals after the crossover, replacing the values ​​exceeding the maximum value with the maximum value and those below the minimum value with the minimum value; Copy the individuals after crossover into the population, then loop through the population, generate a random number and compare it with the crossover mutation probability. If there is a mutation, randomly select an individual to prepare for the mutation. Loop through each gene of the chromosome, perform polynomial mutation on the selected chromosome, then copy the mutated individuals into the population, and recalculate the targets of the offspring after crossover mutation to obtain the optimal and worst targets; Then compare the current optimal goal with the global optimal goal. Replacing the worst with the historical global optimal goal can increase the probability of the population iterating out better individuals. Record the average goal and optimal goal of this generation. After the iteration, derive the value of the optimal penalty factor C and the optimal kernel function parameter δ.

7. The PID-based support vector machine regression monitoring model for VOCs concentration in water, soil and air according to claim 1, characterized in that: The monitoring PID signal of VOCs obtained in a humidity environment needs to be denoised by adaptive threshold weighted wavelet packet.

Images