Method for predicting degradation rate of petroleum hydrocarbon in site groundwater by efficient degrading flora based on complex hydrochemical conditions and BP neural network

Abstract

The invention belongs to the technical field of groundwater pollution remediation, and discloses a method for predicting the degradation rate of petroleum hydrocarbon in site groundwater by efficient degrading flora based on complex hydrochemical conditions and a BP neural network. In order to solve the technical problems that chemical conditions of water in underground water in different sites are different, and the composite influence of the chemical conditions on the remediation effect of the efficient degrading flora is difficult to predict, a data set of the petroleum hydrocarbon degradation rate, the water body dissolved oxygen, the water temperature, the salinity, the NH3-N concentration and the remediation time is constructed through laboratory simulation. Secondly, a BP neural network model for predicting the degradation rate of petroleum hydrocarbon in underground water through dissolved oxygen in water, water temperature, salinity, NH3-N concentration and restoration time is established, and it is proved that the model has good prediction capacity and can effectively predict the degradation rate of the petroleum hydrocarbon by flora under different underground water chemical conditions. A reference is provided for predicting the effect of repairing petroleum hydrocarbon in underground water by microorganisms.

Description

Technical Field

[0001] The invention belongs to the field of groundwater pollution and restoration, and relates to a method for predicting the degradation rate of petroleum hydrocarbons in site groundwater by using efficient degradation bacteria based on complex water chemical conditions and BP neural network. Background Art

[0002] During the oil extraction, transportation, storage and processing, there will be leakage, and oil leakage into the soil and groundwater of petrochemical sites will cause serious damage to the groundwater environment. Bioremediation, especially microbial remediation, is currently the most promising strategy for the remediation of petroleum hydrocarbon contaminated groundwater due to its environmental friendliness, low cost and no secondary pollution.

[0003] The hydrological conditions of coastal petrochemical sites are complex and vary greatly. The water temperature in groundwater is low, usually between 10-20°C. And the dissolved oxygen content in groundwater is low, usually between 0.1-2 mg / L. The salinity and ammonia nitrogen content in groundwater are high, far exceeding the Class IV groundwater standard. In the process of microbial remediation, the above factors will have a great impact on the degradation rate of petroleum hydrocarbons, making it difficult to judge the feasibility of remediation in actual site applications and predict the remediation effect. The cost of conducting experiments to evaluate the effect of microbial remediation is high, the cycle is long, and the samples are easily affected by environmental factors, resulting in deviations and delays in the measurement data. Therefore, it is particularly important to establish a predictive model for the remediation of petroleum hydrocarbon contamination in groundwater by efficient degradation bacteria to evaluate the feasibility of remediation and predict the degradation rate of petroleum hydrocarbons in groundwater on sites under complex water chemical conditions.

[0004] At present, the models for predicting the degradation rate of petroleum hydrocarbons in groundwater mainly include kinetic models, multivariate quadratic regression models, response surface methodology, etc. The literature (Khan AA, et al. "Establishing correlations and scale-up factor for estimating the petroleum biodegradation rate in soil." Bioremediation Journal, 2015, 19 (1): 32-46.) constructed a regression model for estimating the degradation rate of petroleum hydrocarbons in soil by including soil physicochemical properties (soil texture, soil organic matter, soil pH, soil water content) and the initial inoculation amount of petroleum-degrading bacteria. 2is 0.93, and the model only screens several parameters related to the physical and chemical properties of soil, and does not take into account the influence of groundwater hydrochemical conditions on the degradation of petroleum hydrocarbons by bacteria. The literature (Liang Qingxia, et al. "Comparison of detection methods for petroleum hydrocarbons in soil and analysis of degradation kinetics." China Environmental Science, 2023, 43(08): 4194-4201.) evaluated the fitting effect of zero-order reaction kinetic equation, first-order reaction kinetic equation, pseudo-first-order reaction kinetic equation and second-order reaction kinetic equation models on the biodegradation of petroleum hydrocarbons in contaminated soil, and found that the second-order reaction kinetics can better simulate the degradation effect of microorganisms on petroleum hydrocarbons, but the second-order reaction kinetics is only a simplified model, which is usually only applicable to specific conditions and ranges, has great limitations, and the prediction results deviate greatly from the actual situation in complex environments. Literature (He Jiani, et al. "Research on high-throughput evaluation methods for functional characteristics of petroleum-degrading bacteria." Contemporary Chemical Industry, 2024, 53(08): 1772-1779.) established a multivariate linear regression model for predicting the degradation rate of petroleum hydrocarbons, including the C23O enzyme activity of the bacterial community itself, the emulsification performance of surfactants, the dehydrogenase activity, and the lipase activity. This regression model was optimized, and the two parameters of lipase and dehydrogenase with no significant correlation were eliminated. The optimized multivariate linear regression equation was established with C23O enzyme activity and surfactant emulsification index as parameters, and the determination coefficient R of the model was 1.34. 2 The fitting accuracy is poor, the prediction effect is poor, and only the influence of the characteristics of the bacterial community on the degradation rate of petroleum hydrocarbons is focused on. The literature (Yang Y, et al. "Efficient Bioremediation of Petroleum-Contaminated Soil by Immobilized Bacterial Agent of Gordonia alkanivorans W33." Bioengineering, 10 (5). (2023)) uses the response surface methodology model to optimize the fermentation conditions of the petroleum hydrocarbon-degrading bacteria Gordonia alkanivorans W33 (including MgCl in the culture medium). 2 and CaCl 2 concentration), predict the biomass after fermentation, and then predict its degradation effect on petroleum in soil, but the model has not paid attention to other potential environmental factors, which affects the completeness and accuracy of the model.

[0005] In actual contaminated sites, the hydrochemical conditions of groundwater in different regions and sites are complex and have obvious differences. Currently, there is no parameter prediction model that comprehensively considers the hydrochemical conditions of groundwater, especially focusing on groundwater temperature, dissolved oxygen content, salinity, NH 3 -N concentration and other important influencing parameters.

[0006] Therefore, the present invention focuses on the technical problem that the hydrochemical conditions in groundwater of different sites are different, and their combined effects make it difficult to predict the remediation effect of efficient degradation bacteria. A prediction method for the degradation rate of petroleum hydrocarbons in groundwater of sites by efficient degradation bacteria based on complex hydrochemical conditions and BP neural network is established. First, a laboratory simulation is used to construct a relationship between the degradation rate of petroleum hydrocarbons and the dissolved oxygen, water temperature, salinity, and NH 3 -N concentration and restoration time. Secondly, a method was established to analyze the dissolved oxygen, water temperature, salinity, NH 3 The BP neural network model that uses five parameters, such as -N concentration and remediation time, to predict the degradation rate of petroleum hydrocarbons in groundwater provides a reference for predicting the effect of microbial remediation of petroleum hydrocarbons in groundwater. Summary of the invention

[0007] In view of the technical difficulty that the existing microbial remediation effect is difficult to predict due to the influence of complex factors of hydrological conditions, the present invention constructs a microbial remediation system based on dissolved oxygen, water temperature, salinity, NH 3 -BP neural network model for predicting petroleum hydrocarbon degradation rate in groundwater based on N concentration and remediation time.

[0008] The technical solution of the present invention is as follows:

[0009] A method for predicting the degradation rate of petroleum hydrocarbons in groundwater of a site by efficient degradation bacteria based on complex water chemical conditions and BP neural network comprises the following steps:

[0010] (1) Effect of bacterial communities on petroleum hydrocarbon degradation under different water chemistry conditions

[0011] The oil-degrading bacteria were cultured in a simulated groundwater medium with diesel as the only carbon source, the amount of diesel added was 1% of the volume of the simulated groundwater medium, and the culture was carried out in a constant temperature shaking incubator;

[0012] The petroleum-degrading bacterial community is the high-efficiency petroleum-degrading bacterial community disclosed in patent CN 116410864 A.

[0013] According to the hydrochemical conditions of the groundwater in the petrochemical site, the environmental parameters are set as follows: dissolved oxygen content is 1.8mg / L-7.2mg / L, culture temperature is 10-25℃, salinity in the system is 0-80g / L in terms of NaCl, and NH 3 -N concentration is 0-5g / L.

[0014] The culture period was 28 days, and samples were taken every 7 days. The residual petroleum hydrocarbon content was determined by the weight method to obtain the petroleum hydrocarbon degradation rate.

[0015] (2) Establishment of a prediction model for the degradation rate of petroleum hydrocarbons in groundwater

[0016] (2.1) Taking each environmental parameter as input variable and petroleum hydrocarbon degradation rate as output indicator;

[0017] Input variable: x 1 = dissolved oxygen content, x 2 = culture temperature, x 3 = Salinity, x 4 =NH 3 -N concentration, x 5 =Repair time;

[0018] Output variable: y 1 =Petroleum hydrocarbon degradation rate;

[0019] The obtained input variables and output variables are used as the sample data set X of the petroleum hydrocarbon degradation rate prediction model, and the mapminmax normalization function is used to normalize the data in the sample data set. The process is:

[0020]

[0021] In the formula, X i is the environmental data of sample data set X; imin and x imax are the minimum and maximum values ​​before normalization, respectively, completing the normalization operation of the sample data set. After normalization, the data sample set X′={x′ 1 , x′ 2 , x′ 3 , x′ 4 , x′ 5};

[0022] (2.2) Determine the network structure. According to the input and output indicators selected in step (2.1), determine the number of layers of the input layer, output layer and hidden layer of the petroleum hydrocarbon degradation rate prediction model based on the BP neural network. The number of nodes in the input layer of the petroleum hydrocarbon degradation rate prediction model for predicting the petroleum hydrocarbon degradation rate is m=5, and the number of nodes in the output layer is n=1. The number of neurons in the hidden layer is determined by the following formula:

[0023]

[0024] In the formula, S is the number of hidden layer nodes, m is the number of input layer nodes, n is the number of output layer nodes, and a is a constant between 1 and 10;

[0025] (2.3) Selection of activation function: the input layer uses the tansig activation function, the output layer uses the purelin activation function, and the trainlm function is selected as the training function of the petroleum hydrocarbon degradation rate prediction model;

[0026] (2.4) The obtained sample data set is used to train a prediction model for the degradation rate of petroleum hydrocarbons in the site groundwater by highly efficient degradation bacteria based on complex water chemical conditions;

[0027] The relevant parameter settings in the training process of the petroleum hydrocarbon degradation rate prediction model are as follows: learning rate is 0.01, number of trainings is 2000, and minimum training error is 0.00001;

[0028] (2.5) Using the statistical index determination coefficient R 2 To evaluate the prediction results of the petroleum hydrocarbon degradation rate prediction model; the determination coefficient R 2 The mathematical form of is as follows:

[0029]

[0030] In the formula, y i is the actual value, f i is the predicted value, is the average value of the actual value, and i is the data sequence number in the sample data set;

[0031] The closer the predicted value is to the actual value, the higher the degree of fit. 2 It can effectively reflect the degree of fit between the predicted value and the actual value and evaluate the quality of the model prediction results.

[0032] Beneficial effects of the invention: The invention takes into account the dissolved oxygen content, water temperature, salinity and NH 3 The influence of -N concentration on the degradation rate of petroleum hydrocarbon pollution remediation by bacterial flora was established by using BP neural network model. 3 The invention has good prediction effect and solves the problem that the degradation effect of microorganisms is difficult to predict due to the combined influence of hydrological conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 This is a comparison chart of the actual value and predicted value of the model training set.

[0034] Figure 2 This is a model validation diagram. DETAILED DESCRIPTION

[0035] The specific implementation of the present invention is further described below in conjunction with the accompanying drawings and technical solutions.

[0036] Example 1 Effect of bacterial flora on degradation of petroleum hydrocarbons at different dissolved oxygen and temperature levels

[0037] (1) Source of bacterial flora

[0038] The invention is derived from the highly efficient petroleum-degrading bacterial community disclosed in patent CN 116410864 A. The bacterial community is composed of a variety of bacteria and fungi. There are three dominant bacterial genera, namely Pseudomonas, Achromobacter and Stenotrophomonas, all of which are petroleum hydrocarbon-degrading genera. There are two dominant fungal genera, namely Mortierella and Fusarium, both of which are petroleum hydrocarbon-degrading genera. There are as many as 25 petroleum hydrocarbon-degrading genera in the bacterial community with a relative abundance greater than 0.1%.

[0039] (2) Effect of bacterial flora on the degradation of petroleum hydrocarbons at different dissolved oxygen and temperature levels

[0040] The activated microbial flora was inoculated into 50 mL of sterilized simulated groundwater culture medium at a dosage of 5% (v / v), and 1% (v / v) diesel was added as the sole carbon source. The dissolved oxygen content in the system was set to 2 (± 0.2) mg / L, 5 (± 0.2) mg / L, and 7 (± 0.2) mg / L, respectively, and the culture temperature was set to 10°C, 15°C, and 20°C, respectively. The dissolved oxygen content and temperature were combined in pairs, and cultured at 180 rpm for 28 days. Samples were taken at 7d, 14d, 21d, and 28d to detect the degradation rate of petroleum hydrocarbons.

[0041] The remaining petroleum hydrocarbons in the sample were extracted with dichloromethane, and the residual petroleum hydrocarbon content was determined by the weight method to obtain the petroleum hydrocarbon degradation rate.

[0042] The simulated groundwater culture medium contains: MgSO 4 ·H 2 O 0.3g / L, NH 4 Cl 1g / L, CaCl 2 ·H 2 O0.3g / L, KH 2 PO 4 0.4g / L, Na 2 SO 4 0.4g / L, MnSO 4 ·H 2 O 0.0015g / L, FeSO 4 7H 2 O0.002g / L.

[0043] (3) Effect of bacterial flora on the degradation of petroleum hydrocarbons at different salinity levels

[0044] The activated microbial flora was inoculated into 50 mL of sterilized simulated groundwater culture medium at a rate of 5% (v / v), and 1% (v / v) diesel was added as the sole carbon source. A certain amount of NaCl was added so that the NaCl content in the culture medium was 0, 20 g / L, 40 g / L, and 60 g / L. The activated microbial flora was inoculated into 50 mL of sterilized simulated groundwater culture medium at a rate of 5% (v / v), and the dissolved oxygen content of the system was maintained at 2 (± 0.2) mg / L. The culture was carried out at 20°C and 180 rpm for 28 days, and samples were taken at 7 days, 14 days, 21 days, and 28 days to detect the degradation rate of petroleum hydrocarbons.

[0045] (4) Bacterial flora in different NH 3 -N concentration level on the degradation of petroleum hydrocarbons

[0046] The activated microbial flora was inoculated into 50 mL of sterilized simulated groundwater medium at a dosage of 5% (v / v), and 1% (v / v) diesel was added as the sole carbon source. A certain amount of NH 3 -N (NH 4 Cl), so that the medium contains NH 3 -N content was 0, 1g / L, and 3g / L. The activated microbial flora was inoculated into 50mL of sterilized simulated groundwater culture medium, with an addition amount of 5% (v / v), and the dissolved oxygen content of the system was maintained at 2 (±0.2) mg / L. The culture was carried out at 20°C and 180rpm for 28 days, and samples were taken at 7d, 14d, 21d, and 28d to detect the degradation rate of petroleum hydrocarbons.

[0047] Example 2 Establishment of a prediction model for the degradation rate of petroleum hydrocarbon pollution by bacterial flora remediation

[0048] (1) The model data set is constructed based on the input and output data obtained, including dissolved oxygen content, water temperature, salinity, NH 3 -N concentration and repair time are used as input variables, and petroleum hydrocarbon degradation rate is used as output variable. The obtained data set is used to train the model. The specific model construction representative symbols are shown in Table 1:

[0049] Table 1 Model parameters representative symbols

[0050]

[0051] The input and output variables obtained in the above steps are used to establish a sample data set of the model, and the mapminmax normalization function is used to normalize the data in the data set. The process is as follows:

[0052]

[0053] In the formula, Xi is the environmental data of sample data set X; imin and x imax are the minimum and maximum values ​​before normalization, respectively, completing the normalization operation of the sample data set. After normalization, the data sample set X′={x′ 1 , x′ 2 , x′ 3 , x′ 4 , x′ 5}, directly used for the input and output variables required for subsequent model establishment;

[0054] (2) Determine the strategy of network structure. According to the input and output indicators selected in step (1), determine the input, output and hidden layer numbers of the network structure. The number of nodes in the input layer of the BP neural network for predicting the degradation rate of petroleum hydrocarbons is m=5, and the number of nodes in the output layer is n=1. The number of neurons in the hidden layer is determined by the following formula:

[0055]

[0056] In the formula, S is the number of hidden layer nodes, m is the number of input layer nodes, n is the number of output layer nodes, and a is a regulation constant between 1 and 10;

[0057] (3) Selection of activation function: the input layer uses the tansig activation function, the output layer uses the purelin activation function, and the trainlm function is selected as the training function of the petroleum hydrocarbon degradation rate prediction model;

[0058] (4) The obtained input and output variables are used to establish a sample data set of the model, and a BP neural network model is obtained through training to predict the degradation rate of petroleum hydrocarbons in the groundwater of the site by efficient degradation bacteria based on complex water chemical conditions.

[0059] The relevant parameter settings during the training process of the BP neural network model are: learning rate is 0.01, training times is 2000, and minimum training error is 0.00001.

[0060] (5) Using the statistical index determination coefficient R 2 To evaluate the prediction results of the petroleum hydrocarbon degradation rate prediction model; the determination coefficient R 2 The mathematical form of is as follows:

[0061]

[0062] In the formula, y i is the actual value, f i is the predicted value, is the average value of the actual value, and i is the data sequence number in the sample data set;

[0063] When the predicted value is closer to the actual value, that is, the degree of fit is higher, R 2 Therefore, R 2 It can effectively reflect the degree of fit between the predicted value and the actual value and evaluate the quality of the model prediction results.

[0064] Model fitting results: determination coefficient R of the model training set 2 is 0.90.

[0065] Example 3 Validation of the prediction model for degradation rate of petroleum hydrocarbon pollution by bacterial flora

[0066] The degradation rate of petroleum hydrocarbons in groundwater was obtained by conducting a petroleum hydrocarbon degradation experiment under water chemical conditions different from those in Example 1, which is the true value. At the same time, the experimental parameters were substituted into the established prediction model to obtain the predicted value. Table 2 shows the true value and predicted value of the degradation rate of petroleum hydrocarbons by the efficient degradation bacteria under different remediation conditions.

[0067] Table 2 Actual and predicted values ​​of degradation rates of petroleum hydrocarbons by efficient degradation bacteria under different conditions

[0068]

[0069] Model fitting results: determination coefficient R 2 =0.97, and the determination coefficient is above 0.9, which means that this method has a good effect in predicting the degradation rate of petroleum hydrocarbons by the efficient degradation bacteria in the groundwater of the site.

Claims

1. A method for predicting the degradation rate of petroleum hydrocarbons in groundwater on site by using efficient degradation bacteria based on complex water chemical conditions and BP neural network, characterized in that: The following steps are involved: (1) Effect of bacterial communities on petroleum hydrocarbon degradation under different water chemistry conditions The oil-degrading bacteria were cultured in a simulated groundwater medium with diesel as the only carbon source, the amount of diesel added was 1% of the volume of the simulated groundwater medium, and the culture was carried out in a constant temperature shaking incubator; The culture period was 28 days, and samples were taken every 7 days. The residual petroleum hydrocarbon content was determined by the weight method to obtain the petroleum hydrocarbon degradation rate. (2) Establishment of a prediction model for the degradation rate of petroleum hydrocarbons in groundwater (2.1) Taking each environmental parameter as input variable and petroleum hydrocarbon degradation rate as output indicator; Input variables: x1 = dissolved oxygen content, x2 = culture temperature, x3 = salinity, x4 = NH3-N concentration, x5 = repair time; Output variables: y1 = petroleum hydrocarbon degradation rate; (2.2) Determine the network structure. According to the input and output indicators selected in step (2.1), determine the number of layers of the input layer, output layer and hidden layer of the petroleum hydrocarbon degradation rate prediction model based on the BP neural network. The number of nodes in the input layer of the petroleum hydrocarbon degradation rate prediction model for predicting the petroleum hydrocarbon degradation rate is m=5, and the number of nodes in the output layer is n=1. The number of neurons in the hidden layer is determined by the following formula: In the formula, S is the number of hidden layer nodes, m is the number of input layer nodes, n is the number of output layer nodes, and a is a constant between 1 and 10; (2.3) Selection of activation function: the input layer uses the tansig activation function, the output layer uses the purelin activation function, and the trainlm function is selected as the training function of the petroleum hydrocarbon degradation rate prediction model; (2.4) The obtained sample data set is used to train a prediction model for the degradation rate of petroleum hydrocarbons in the site groundwater by highly efficient degradation bacteria based on complex water chemical conditions; (2.5) Using the statistical index determination coefficient R 2 To evaluate the prediction results of the petroleum hydrocarbon degradation rate prediction model.

2. The method for predicting the degradation rate of petroleum hydrocarbons in groundwater by efficient degradation bacteria according to claim 1, characterized in that: According to the hydrochemical conditions of the groundwater in the petrochemical site, the environmental parameters were set as follows: dissolved oxygen content was 1.8 mg / L-7.2 mg / L, the culture temperature was 10-25 °C, the salinity in the system was 0-80 g / L in terms of NaCl, and the NH3-N concentration in the system was set to 0-5 g / L.

3. The method for predicting the degradation rate of petroleum hydrocarbons in groundwater by efficient degradation bacteria according to claim 1, characterized in that: The obtained input variables and output variables are used as the sample data set X of the petroleum hydrocarbon degradation rate prediction model, and the mapminmax normalization function is used to normalize the data in the sample data set. The process is: Where, X i is the environmental data of sample data set X; imin and x imax are the minimum and maximum values ​​before normalization respectively, completing the normalization operation on the sample data set. The normalized data sample set X′={x′1, x′2, x′3, x′4, x′5}.

4. The method for predicting the degradation rate of petroleum hydrocarbons in groundwater by efficient degradation bacteria according to claim 1, characterized in that: The relevant parameter settings in the training process of the petroleum hydrocarbon degradation rate prediction model are as follows: the learning rate is 0.01, the number of training times is 2000, and the minimum training error is 0.00001.

5. The method for predicting the degradation rate of petroleum hydrocarbons in groundwater of a site by a high-efficiency degradation bacterial community according to claim 1, characterized in that: The simulated groundwater culture medium contains: MgSO4·H2O 0.3g / L, NH4Cl1g / L, CaCl2·H2O0.3g / L, KH2PO4 0.4g / L, Na2SO4 0.4g / L, MnSO4·H2O 0.0015g / L, and FeSO4·7H2O 0.002g / L.

Images