Polysaccharide system asphaltene molecular flocculation dynamics evaluation and prediction method

Abstract

The invention discloses a polysaccharide system asphaltene molecule flocculation dynamics evaluation and prediction method, and belongs to the technical field of oil and gas pipeline flow safety guarantee. The method comprises the following steps: constructing polysaccharides with different properties; constructing a model asphaltene; constructing a system of different polysaccharide and model asphaltene mass ratios; energy minimization and pre-balance of the simulation system; carrying out molecular dynamics simulation of the simulation system; an asphaltene molecule flocculation evaluation method system with the molecular cluster number, the molecular contact number and the accessible surface area of the solvent as key evaluation indexes is constructed, and the dynamic flocculation process of asphaltene molecules is evaluated; a machine learning model is established, trained and tested, and the asphaltene flocculation dynamic process is rapidly and accurately predicted. The method is used for evaluating the influence of the key component polysaccharide of the biological membrane in the oil and gas pipeline on the asphaltene flocculation dynamic process.

Description

Technical Field

[0001] The invention relates to an evaluation and prediction method for the flocculation dynamics process of asphaltene molecules in a system in which polysaccharides, a key component of microbial membranes in oil and gas pipelines, exist, and specifically belongs to the technical field of flow safety assurance of oil and gas pipelines. Background Art

[0002] The oil and gas resources developed in my country are all transported to onshore terminals through submarine pipelines. Oil and gas pipeline transportation has become one of my country's five major transportation industries, playing a very important role in the national economy and is called the "energy lifeline."

[0003] Pipeline blockage caused by asphaltene flocculation and deposition in oil and gas pipelines and pipeline leakage caused by microbial corrosion of the inner wall of the pipeline are two technical problems that have long restricted the safe long-distance transportation of deep-sea oil and gas. On the one hand, asphaltene is the non-hydrocarbon component with the largest molecular weight and the strongest polarity in crude oil. It is easy to form micelles by molecular association, and then aggregate into larger floccules. After precipitation, aggregation, surface contact and adhesion, it finally settles at the bottom of the pipeline, resulting in a significant reduction in oil transportation efficiency, a decrease in pipeline oil and gas transmission volume and an increase in pipeline pressure difference, which damages pipe fittings and causes accidents. On the other hand, impurities such as water, organic acids, carbon dioxide, hydrogen sulfide gas mixed in the pipeline and microorganisms introduced during the oil extraction and transportation process will adhere to the pipe wall to form a biofilm. The biofilm contains a large number of microorganisms and bacterial extracellular polymers. These gases and corrosive substances can cause pipeline corrosion, causing huge economic losses and safety hazards.

[0004] Asphaltene flocculation deposition and pipeline biocorrosion processes often coexist, seriously threatening the safety of pipeline oil and gas flow. At present, both domestic and foreign countries take separate prevention and control measures for these two processes, ignoring the interaction between the two, which makes it difficult for comprehensive prevention and control technology to move from the laboratory to industrialization. For example, the main source of biocorrosion is the biofilm attached to the pipe wall, which is a porous structure composed of microbial cells and extracellular polymers (EPS). EPS is mainly composed of polysaccharides and proteins, with strong viscosity, which can promote the adhesion between biofilms and between biofilms and metal pipe walls. It can also enhance the adsorption of organic matter and inorganic ions and change the flow characteristics of the surrounding fluid, which will affect the flocculation between petroleum asphaltene molecules and the adhesion characteristics between asphaltene and metal pipe walls, which is crucial to the pipeline blockage process caused by asphaltene. However, there is currently a lack of a method to evaluate the effect of key components of biofilms (such as polysaccharides) on the kinetic process of asphaltene molecule flocculation in real time; more importantly, at different stages of oil and gas pipeline biocorrosion, the concentration and molecular structure of polysaccharides are in dynamic changes, and there is a lack of a method that can quickly and accurately predict the changes in the kinetic characteristics of asphaltene molecule flocculation over time and the properties of polysaccharides.

[0005] Based on this, the present invention discloses a method for evaluating and predicting the kinetic process of asphaltene molecular flocculation in a polysaccharide system, establishes an asphaltene molecular flocculation evaluation index system, and can accurately predict the asphaltene flocculation index according to the polysaccharide concentration and molecular structure characteristics, thereby overcoming the deficiencies of the existing technology, providing technical support for the rapid prediction, comprehensive inhibition, and efficient unblocking of asphaltene flocculation blockage in bio-corrosion pipeline systems, and promoting the improvement and perfection of the flow safety assurance system for long-distance oil and gas pipelines. Summary of the invention

[0006] The main purpose of this patent is to overcome the shortcomings of the existing technology and provide a method for evaluating and predicting the flocculation dynamics of asphaltene molecules under the influence of polysaccharides widely present in biocorroded oil and gas pipelines. To achieve the above purpose, the present invention mainly adopts the following technical solutions:

[0007] S1. Construction of polysaccharide molecules and asphaltene molecular structures with different properties

[0008] S2. Construction of mixed solvent systems of polysaccharides and asphaltene with different properties

[0009] S3. Molecular dynamics simulation of the mixed system

[0010] S4. Analyze the equilibrium structure of polysaccharides and asphaltene, radial distribution function of asphaltene, etc. during the simulation process, and construct an asphaltene flocculation evaluation method with the number of asphaltene molecular clusters, molecular contact number, and solvent accessible surface area as key indicators.

[0011] Among them, S1 includes:

[0012] S1-1. Use Charmm-gui online tool to draw the molecular structure of polysaccharides, and manually adjust and modify the molecular structure file and topology file parameters according to the gromos54A7 force field parameters of gromacs. The polysaccharide structure is composed of mannuronic acid (M) and guluronic acid (G) connected by α-1,4 glycosidic bonds. Depending on the different sources and extraction conditions, the ratio of mannuronic acid (M) and guluronic acid (G) is different, resulting in differences in their physical and chemical properties. Three ratios were designed: all G is 1, G / M is half and half is 0.5, and all M is 0.

[0013] S2 includes:

[0014] S2-1, Construction of 9×9×11nm 3 into which polysaccharide chains of different properties are then randomly inserted.

[0015] S2-2. Select VO-79 as the asphaltene model and randomly insert 12 asphaltene molecules into the box. S2-3. Fill the box with 10108 water molecules. The water molecule model is the SPC model.

[0016] S3 includes:

[0017] S3-1. For each simulation system, the system energy is minimized using the steepest descent method, so that the system energy converges to 1000 kJmol -1 nm -1 .

[0018] S3-2, molecular dynamics equilibrium was performed using NVT ensemble (constant number of atoms, volume and temperature) and NPT ensemble (constant number of atoms, pressure and temperature) for 0.5 ns each, so that the system reached the preset temperature (298 K) and pressure (1 bar) to obtain a reasonable molecular configuration and molecular density.

[0019] S3-3, the pre-equilibrium system was subjected to a 30ns molecular dynamics simulation under the NPT ensemble, during which all molecules were free to move without artificial restrictions. In all simulations, periodic boundary conditions were used and applied in the X, Y, and Z directions to minimize the influence of boundary effects of finite systems.

[0020] S4 includes:

[0021] S4-1. The microstructure of polysaccharides and asphaltene was analyzed by VMD software.

[0022] S4-2. Use the GROMACS analysis module to calculate the number of asphaltene molecular clusters, molecular contacts, solvent accessible surface area, etc.

[0023] S5. Build a machine learning model, use the data obtained in S4 to train, verify and test the machine learning model, and establish a prediction method for the asphaltene molecular flocculation kinetic process. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 : Flow chart of the steps of the patent method;

[0025] Figure 2 :Composition table of each simulation system;

[0026] Figure 3 :Represents 150g L -1 Concentration of 12G polysaccharide and asphaltene simulated equilibrium structure diagram;

[0027] Figure 4 :Represents 150g L -1 Concentration of 12G polysaccharide and asphaltene simulation initial system diagram;

[0028] Figure 5 : Radial distribution g(r) diagram of asphaltene in 12G polysaccharide simulation system;

[0029] Figure 6 : The number of clusters of asphaltene in the 12G polysaccharide simulation system;

[0030] Figure 7 : The solvent accessible surface area of ​​asphaltene in 12G polysaccharide simulation system;

[0031] Figure 8 : Molecular contact number diagram of asphaltene in 12G polysaccharide simulation system;

[0032] Fig. 9 : Coefficient of determination and root mean square error plots for machine learning training data and test data. DETAILED DESCRIPTION

[0033] The following will combine the drawings in the embodiments of this patent to clearly and completely describe the technical solutions in the embodiments of this patent. Obviously, the described embodiments are only part of the embodiments of this patent, not all of the embodiments; based on the embodiments in this patent, other persons in this field use the materials used in this patent without other evaluation characteristics, which are all within the scope of protection of this patent.

[0034] The patent is further described in detail below in conjunction with the accompanying drawings.

[0035] S1. Molecular structure construction

[0036] The three-dimensional structural models of polysaccharides and asphaltene (ASP) molecules with different properties were constructed, and the molecular coordinates were extracted after preliminary geometric optimization. The molecular coordinate files were then converted into topological files and coordinate files recognizable by GROMACS through the ATB website. According to the molecular structure information of the same type in the molecular force field library, the atomic type, bond length, bond angle, dihedral angle and other parameters in the topological structure were manually adjusted. The water molecule model adopts the SPC model.

[0037] S2. Initial system construction of the model

[0038] The 9×11×11nm 3 The polysaccharide chains with different properties were randomly inserted into the boxes in different systems. The degree of polymerization was set to 4, 8, and 12, the G / M ratio was set to 1, 0.5, and 0, and the polysaccharide concentration was set to 0 g L -1 , 50g L -1 , 100g L -1 , 150g L -1 At the same time, a corresponding number of sodium ions were added to the box to neutralize the charge, and 12 VO79 model asphaltene molecules were added to each system. Then 10108 water molecules were added to the simulation box, so that the entire system was simulated in the water phase. The number of water molecules and asphaltene molecules in all simulation systems remained consistent. The simulation systems are as follows Figure 2 shown.

[0039] S3. Energy minimization

[0040] For each simulation system, the system energy was first minimized using the steepest descent method until the system energy converged to 1000 kJ mol -1 nm -1 In order to prevent the system from collapsing due to inappropriate initial configuration, a restricted dynamics simulation is required to balance the system before the formal molecular dynamics simulation, that is, using a force constant of 1000 kJ mol -1 nm -2 The simple harmonic force restricts the movement of asphaltene and polysaccharide molecules within a certain range.

[0041] S4. Pre-balance

[0042] The NVT ensemble (constant number of atoms, volume and temperature) and the NPT ensemble (constant number of atoms, pressure and temperature) were used for molecular dynamics equilibrium for 0.5 ns each, so that the system reached the preset temperature (298 K) and pressure (1 bar) to obtain a reasonable molecular configuration and molecular density.

[0043] S5. Formal simulation

[0044] The pre-equilibrated system was subjected to a 30 ns molecular dynamics simulation under the NPT ensemble, during which all molecules were free to move without artificial restrictions. In all simulations, periodic boundary conditions were used and applied in the X, Y, and Z directions to minimize the influence of boundary effects of finite systems. The V-rescale thermostat and Berendsen barostat algorithms were used to keep the temperature and pressure at around 298K and 1 bar, respectively, with coupling constants of 0.1ps and 2.0ps, respectively. The cutoff radius for short-range van der Waals and electrostatic force calculations was set to 1.2nm. Particle Mesh Ewald was used to calculate long-range electrostatic force interactions, with an order of magnitude of 4 and a Fourier spacing of 0.16nm. During the entire simulation process, the simulation step size was set to 1fs, and a frame of motion trajectory was saved every 0.1ns for subsequent data analysis.

[0045] S6. Evaluation of the kinetics of asphaltene flocculation

[0046] S6-1 Using the molecular visualization program VMD, we can observe the changes in the configuration of the system over time. Figure 3 As shown, taking the 12G polysaccharide system as an example, the VMD software displays the structural composition of the polysaccharide chain and asphaltene in the system after equilibrium. Red represents polysaccharide molecules and green represents asphaltene molecules. Water molecules and sodium ions are not shown in the figure.

[0047] S6-2 uses the GROMACS analysis module to analyze the simulation results. For example, the solvent accessible surface area, the number of clusters, the radial distribution function, etc. can be directly obtained through the corresponding program commands. The radial distribution function (RDF) between the molecular mass centers can be used to obtain the stacking mode of the flocs.

[0048] S6-2-1 Solvent accessible surface area refers to the surface area of ​​a molecule that can contact with solvent molecules in a specific solvent (such as water). It is usually used to evaluate the hydrophobicity, interaction and biological activity of molecules. The calculation is obtained through the gromacs built-in command gmxsasa. The principle is to roll a virtual probe ball (usually a water radius of 0.14nm) on the surface of the molecule, and the surface that can contact with the molecule becomes the solvent accessible surface area. The larger the solvent accessible surface area value, the larger the contact area between the molecule and the solvent, and the lower the degree of clustering of the corresponding molecules, and the more dispersed the clusters.

[0049] S6-2-2 The number of clusters is calculated using the command gmx clustsize, and the cutoff distance is determined based on the radial distribution function between asphaltene. The greater the number of clusters, the lower the degree of cluster aggregation, which is consistent with the solvent accessible surface area and the more dispersed the molecular aggregation.

[0050] S6-2-3 Radial distribution function formula

[0051]

[0052] Among them: g(r) is the probability of other particles appearing at a position r away from the central particle; dN is the number of other particles in the spherical area at a position r away from the central particle; ρ is the system density composed of the asphalt molecular model; dr is the thickness of the spherical area; r is the distance from the central particle.

[0053] In gromacs, it can be directly obtained through the gmx rdf command, and the stacking method of asphaltene can be analyzed through g(r).

[0054] S6-3 Molecular contact number. To quantify the flocculation strength between asphaltene and asphaltene, the minimum distance between any two asphaltene molecules (polycyclic aromatic core) was calculated using the GROMACS standard module gmx midist. When the minimum distance between two molecules is less than or equal to 0.5nm, it indicates that the two molecules have flocculated, and the number of intermolecular interactions is counted based on this.

[0055] Example 1: Taking VO-79 as an asphaltene model and a monomer consisting of 12 guluronic acids (G) to form an alginate polysaccharide system as an example, the specific implementation steps of the evaluation method of the effect of molecular simulation polysaccharide on asphaltene cluster aggregation of the present invention are described.

[0056] Step 1: Establish the polysaccharide model. Select 12 guluronic acids (G) on the Charmm-gui website and then save and export the pdb file. Then import the file into the ATB website to obtain the topology file and structure file and modify the various parameters of the file to make it conform to the corresponding force field file.

[0057] Step 2: Construct a simulation box system of polysaccharides and asphaltene. Figure 4 Shown in 9*11*11nm 3 The boxes were randomly inserted with different numbers (4, 8, 13) of alginate polysaccharide chains representing different concentrations and the corresponding number of sodium ions to neutralize the charge and 12 asphaltene molecules were inserted, filling with 10108 water molecules.

[0058] Step 3: Minimize the energy of the system and pre-balance it. Use the steepest descent method to minimize the energy of the system so that the energy of the system converges to 1000 kJ mol -1 nm -1 The NVT ensemble (constant number of atoms, volume and temperature) and the NPT ensemble (constant number of atoms, pressure and temperature) were used for molecular dynamics equilibrium for 0.5 ns each, so that the system reached the preset temperature (298 K) and pressure (1 bar) to obtain a reasonable molecular configuration and molecular density to prevent the system from collapsing during formal simulation.

[0059] Step 4: Formal simulation: Perform a 30 ns molecular dynamics simulation in the NPT ensemble.

[0060] Step 5: Evaluation of the effect of polysaccharides on asphaltene aggregation. The molecular flocculation process of asphaltene at three representative concentrations of polysaccharide chains, 4, 8, and 13, was simulated and calculated. The radial distribution function, cluster number, solvent accessible surface area, and molecular contact number of asphaltene at three concentrations were obtained through simulation and calculation, as shown in the following table. Figure 5 , Figure 6 , Figure 7 , Figure 8 shown.

[0061] By comparing the radial distribution function, cluster number, solvent accessible surface area, and molecular contact number of asphaltene under different polysaccharide concentrations, the effect of polysaccharides on asphaltene aggregation under current conditions can be evaluated and analyzed. Overall, after the simulated equilibrium, as the polysaccharide concentration increases, the parallel stacking of asphaltene also increases, the number of molecules interacting between asphaltene molecules also increases, and the number of clusters formed also shows an increasing trend, and the solvent accessible surface area of ​​asphaltene also increases. This shows that the higher the polysaccharide concentration, the smaller the asphaltene flocculation tendency, and high concentrations of polysaccharides can hinder the flocculation of asphaltene molecules.

[0062] Similarly, this method can be used to simulate polysaccharide chains with different MG ratios and different numbers of monomers to obtain the effects of different properties of polysaccharides on asphaltene flocculation.

[0063] S7. Prediction of the kinetics of asphaltene flocculation

[0064] In order to predict the asphaltene aggregation process over time, a random forest (RF) model was established to perform data mining using the initial trajectory and 100 trajectories output from each production simulation run. Before formally conducting data mining, the data was normalized to be unified in the range of [0,1]. Data normalization was processed using the following formula.

[0065]

[0066] x i Represents a data set, min(x i ) represents the minimum value of the data set, max(x i ) represents the maximum value of the data set, Represents normalized data.

[0067] RF is a decision tree algorithm with easy-to-set parameters. It randomly divides the training data into a set of decision trees and trains these decision trees in parallel. This patent involves a total of 3636 data sets that are randomly divided into two groups, including a training data set (80%) and a test data set (20%). The data set includes three dependent variables, namely the number of molecular clusters, the number of molecular contacts, and the solvent accessible surface area of ​​asphaltene; and four independent variables, namely the polysaccharide concentration (0, 50, 100, and 150 g L -1 ), the number of monomers in each polysaccharide chain (4, 8 and 12), the proportion of G-monomers in each polysaccharide chain (0, 0.5 and 1) and time (from 0 to 30ns, with an interval of 0.3ns). In order to reduce the complexity of the model, RF models were established for each dependent variable separately, and the training process was verified using a 10-fold cross-validation method. During the training process, the accuracy of the RF model was affected by five parameters, such as the maximum depth of the tree, the random number of seeds used, the number of execution slots for building the ensemble, the number of features on each node, and the number of trees in the random forest. In this patent, the first three parameter settings were set to default, and the number of features and the number of trees were adjusted using a trial and error method, varying between 0 and 6 and 0 to 2000, respectively. Finally, the performance of the trained model was further verified using a test dataset that was not used for model development. The coefficient of determination (R 2 ) and root mean square error (RMSE) were used to evaluate the predictability of the model.

[0068] The coefficient of determination and root mean square error of the random forest model of the solvent accessible surface area test data after training in this patent are 0.9827 and 1.7529 respectively. The coefficient of determination and root mean square error of the random forest model of the molecular contact number test data are 0.9628 and 1.0609 respectively. The coefficient of determination and root mean square error of the random forest model of the molecular cluster number test data are 0.9643 and 0.2855 respectively. It can be seen from the results that the test results of the random forest model established in this patent are highly accurate and have strong generalization ability, which can be used to quickly and accurately predict the flocculation kinetics of asphaltene molecules in polysaccharide systems.

Claims

1. A method for evaluating and predicting the flocculation kinetics of asphaltene molecules in a polysaccharide system, characterized by: The method comprises the following steps: Step 1: Construct polysaccharide and asphaltene molecules. Polysaccharide molecules are polysaccharides with different properties composed of (M) mannuronic acid and (G) guluronic acid with different compositions and quantities. Petroleum asphaltene is generally divided into continental type with more aromatic rings and fewer side chains and archipelago type with fewer aromatic rings and more side chains. VO79 is selected as the model asphaltene, which can well represent the continental asphaltene. Step 2: Use the ATB website to optimize the structure of the constructed molecules and obtain ITP and GRO files of polysaccharide and asphaltene molecules with suitable optimized structural energy. Step 3: Randomly place the optimized molecules into a 9*11*11nm 3 The reaction system is constructed by using a box, and the amount of asphaltene and water molecules in the box is fixed. By changing the number of polysaccharide chains and controlling the mass ratio of polysaccharide chains to asphaltene, different concentrations of polysaccharides are expressed, and by changing the monomer composition chain length and GM ratio, polysaccharides of different properties are expressed. Step 4: Minimize the energy of the constructed system for 50,000 steps and set the maximum force to 1000 kJ mol -1 nm -1 Then, 0.5 ns of NVT and NPT pre-equilibrium were performed to make the energy of the system reasonable. Finally, a 30 ns formal simulation was performed, using a V-rescale temperature controller and a Berendsen voltage controller. Step 5: Use VMD software to observe the structure and changes of the system molecules during the simulation process and after equilibrium, and calculate the number of clusters, solvent accessible surface area, radial distribution function, number of molecular contacts and other parameters of asphaltene molecules through the GROMACS analysis module. Take the properties and time of polysaccharides as independent variables, calculate the number of clusters, solvent accessible surface area, number of molecular contacts and other parameters of asphaltene as dependent variables, and use WEKA machine learning software for random forest analysis. Step 6. Describe and evaluate the molecular flocculation kinetics of asphaltene through parameters such as the radial distribution function of asphaltene molecules, the number of clusters, the solvent accessible surface area, and the number of molecular contacts. Step 7. Establish a machine learning model to predict the flocculation kinetic parameters of asphaltene molecules in the polysaccharide system.

2. The method for evaluating and predicting the flocculation kinetics of asphaltene molecules in a polysaccharide system according to claim 1, characterized in that: A method system for evaluating the asphaltene flocculation kinetics in polysaccharide systems was established with the number of asphaltene molecular clusters, molecular contact number, and solvent accessible surface area as key evaluation indicators. This can evaluate the influence of polysaccharides, key components of biofilms in oil and gas pipelines, on the flocculation kinetics of asphaltene molecules. A mathematical model based on a machine learning algorithm was established that can quickly and accurately predict the changes in asphaltene molecular flocculation kinetics parameters over time and as a function of polysaccharide structural properties and concentration at different stages of biocorrosion.

Images