A method for controlling a natural gas purification process

Abstract

The present invention relates to a control method for natural gas purification process, belonging to the technical field of natural gas purification. The method includes the following steps: collecting current raw natural gas parameters, obtaining the operation parameters of a natural gas purification device and the required amount of energy-carrying medium with the lowest overall process energy consumption under the condition of the raw natural gas parameters through intelligent algorithm optimization; controlling the utility engineering according to the required amount of energy-carrying medium, and controlling the natural gas purification device according to the operation parameters of the natural gas purification device; the intelligent algorithm can adopt iterative optimization such as genetic algorithm and particle swarm algorithm; the overall process energy consumption can be predicted by training a neural network model with historical data. The method of the present invention can guide the control strategies of natural gas purification devices and utility engineering facilities, reduce the total energy consumption and material consumption in the natural gas purification process, and reduce the production cost.

Description

Technical Field

[0001] The present invention relates to a control method for natural gas purification process, belonging to the technical field of natural gas purification. Background Art

[0002] As a high-quality clean energy, natural gas has been steadily increasing its proportion in China's energy consumption structure year by year. However, the natural gas purification process consumes a large amount of energy, especially the purification process of high-sulfur natural gas, which has great energy-saving potential.

[0003] A natural gas purification plant includes a natural gas purification device and utility engineering facilities. The natural gas purification device is used to purify the natural gas containing impurities into qualified natural gas and meet the emission standards throughout the process. The utility engineering facilities are used to provide the energy-carrying working fluids and energy required in the production process of the natural gas purification device, such as steam, electricity, fuel gas, etc., as well as nitrogen and purified circulating water, etc.

[0004] Traditional natural gas purification devices evenly distribute the processing volume of raw natural gas (the natural gas to be purified, hereinafter also referred to as raw gas) by the startup device, and establish mathematical models under ideal conditions such as energy balance and material balance. However, there is a large deviation between the energy consumption calculated by the traditional method and the actual operation situation. There is a problem that high-efficiency devices do not fully play their roles and the energy utilization rate is low. As the operation time of natural gas purification devices and energy-consuming equipment such as boilers and pumps in the utility engineering system prolongs, the energy-saving measures in the production process mainly rely on the dispatcher's empirical judgment of the demand and operation parameters of the natural gas purification device to obtain the quantity of energy-carrying working medium that the utility engineering needs to provide, which cannot accurately control and maximize energy utilization. At the same time, in order to prevent the product gas from not meeting the standards due to insufficient energy supply, an excessive supply method is often used for production, resulting in energy waste. Especially for multiple parallel purification series, it is more difficult for production scheduling to master when the flow rate of raw natural gas is lower than the full-load design condition. The difficulty of this work lies in that the purification process includes many process units such as acid gas removal, dehydration, sulfur recovery, tail gas treatment, and acid water stripping. There are many key parameters affecting energy consumption and they have strong coupling and correlation. The impact on system energy use is non-linear, and it is difficult to obtain the functional relationship between system energy use and operation parameters by classical mathematical regression analysis methods. The operation parameters of the purification device must first meet the processing requirements of the raw gas, which is first related to the flow rate and composition of the fuel gas. Under different operation parameters, the purification device has different demands for energy-carrying working medium, energy, circulating water, and nitrogen. And different demands for energy-carrying working medium further require different operation parameters of the utility engineering, such as the operation parameters of boilers and power pumps. The change of the operation parameters of the utility engineering further leads to the change of the energy consumption of the utility engineering. The difficulty of energy saving is that if only the flow rate and composition of the raw gas are considered for control to minimize the energy consumption (the consumption of energy and materials directly supplied such as energy, circulating water, and nitrogen) of the purification device, but at this time, the demand of the purification device for the energy-carrying working medium may make the energy consumption of the utility engineering relatively high. Therefore, it is very difficult to achieve the lowest energy consumption in the entire process of natural gas purification. Summary of the Invention

[0005] The purpose of the present invention is to provide a control method for natural gas purification process to solve the problem of high energy consumption in the entire process of natural gas purification.

[0006] To achieve the above purpose, the solution of the present invention includes:

[0007] A control method for natural gas purification process of the present invention includes the following steps:

[0008] 1) Collect the current raw natural gas parameters, and obtain the operation parameters of the natural gas purification device and the demand for energy-carrying working medium with the lowest comprehensive energy consumption in the process under the conditions of the raw natural gas parameters through optimization by intelligent algorithm B;

[0009] 2) Control the utility engineering according to the demand of the energy - carrying medium, and control the natural gas purification device according to the operating parameters of the natural gas purification device;

[0010] The process of optimizing by intelligent algorithm B in step 1) includes: First, initialize the population, where the population includes the operating parameters of the natural gas purification device; then calculate the population fitness, where the fitness is the comprehensive process energy consumption, and the comprehensive process energy consumption is calculated by prediction model A; finally, change the population and iterate multiple times, and finally select the operating parameters of the natural gas purification device corresponding to the population with the lowest comprehensive process energy consumption and the corresponding demand of the energy - carrying medium as the optimization result;

[0011] The prediction model A is a machine - learning model, which is trained by the historical data of raw natural gas parameters, the demand of the energy - carrying medium corresponding to the operating parameters of the natural gas purification device, energy consumption, and material consumption;

[0012] The comprehensive process energy consumption is the sum of the demand of the energy - carrying medium, energy consumption, and material consumption in the natural gas purification process.

[0013] The utility engineering of the natural gas purification plant needs to provide the necessary energy - carrying medium for the purification process, which is the key and difficult point of energy conservation and consumption reduction. The demand for the energy - carrying medium in the purification process is related to parameters such as the flow rate and pressure of the raw natural gas to be processed, and is also related to multiple operating parameters of the purification equipment. The operating parameters of the purification equipment determine the power consumption, fuel gas consumption, and consumption of materials such as water and nitrogen of the purification equipment. The present invention regards the consumption of the purification equipment itself and the consumption of the utility engineering as the total consumption based on historical data. Through the optimization algorithm, the lowest total consumption of the purification equipment and the utility engineering is found, and the purification device is adjusted and controlled according to the state with the lowest total consumption, and the utility engineering is controlled to provide the energy - carrying medium required at the lowest total consumption, effectively reducing the total energy consumption in the natural gas purification process. It breaks the traditional natural gas purification plant, where each purification series is controlled independently, and the utility engineering is also independently controlled. According to the total demand of the energy - carrying medium of each purification series of purification devices, an empirical and excessive supply method is adopted. In the traditional method, the energy consumption and material consumption of each purification series alone may be controlled to a relatively low level, but considering the energy consumption of the utility engineering, it is difficult for the whole plant to achieve the optimal energy consumption. The method of the present invention overcomes the traditional natural gas purification control idea, combines the demand of the energy - carrying medium, models and optimizes with the lowest total consumption as the goal, and realizes the reduction of the energy consumption in the whole process of natural gas purification.

[0014] Further, in step 2), the method of controlling the utility engineering according to the demand of the energy - carrying medium is to obtain the optimized operating parameters of the utility engineering equipment with the lowest energy consumption of the utility engineering equipment when meeting the demand of the energy - carrying medium through optimization by intelligent algorithm D, and control the corresponding equipment of the utility engineering according to the optimized operating parameters of the utility engineering equipment;

[0015] The process of optimizing with the intelligent algorithm D includes: First, initialize the population, which includes the operating parameters of utility engineering equipment; then calculate the fitness of the population, where the fitness is the energy consumption of utility engineering equipment, and the energy consumption of utility engineering equipment is calculated by the prediction model C; finally, change the population and iterate multiple times, and finally select the operating parameters of the utility engineering equipment corresponding to the population with the lowest energy consumption of the utility engineering equipment as the optimized operating parameters of the utility engineering equipment;

[0016] The prediction model C is a machine learning model, which is trained by the historical data of the energy-carrying medium supply amount and the operating parameters of the utility engineering equipment corresponding to the energy consumption of the utility engineering equipment.

[0017] In the present invention, the utility engineering is controlled to provide the energy-carrying medium according to the demand for the energy-carrying medium. For the utility engineering, the production of the energy-carrying medium is proportional to the energy consumption and material consumption of the utility engineering equipment. The lower the demand for the energy-carrying medium, the lower the energy consumption and material consumption of the utility engineering. However, based on historical data, under the condition of the lowest current demand for the energy-carrying medium, the present invention further optimizes to find the operating parameters of the utility engineering control that meet the current demand for the energy-carrying medium and have the lowest energy consumption of the utility engineering (generally, when the energy consumption of the utility engineering is the lowest, the material consumption is also the lowest), further reducing the total energy consumption in the process of natural gas purification.

[0018] Further, the prediction model C is a neural network model.

[0019] Further, the intelligent algorithm D is a particle swarm algorithm.

[0020] The prediction model C can be any machine learning model. The machine learning model is trained with a large amount of historical data. Based on this model, the supply amount of the energy-carrying medium and the corresponding energy consumption of the utility engineering under different operating parameters of the utility engineering can be predicted.

[0021] The intelligent algorithm D can adopt iterative optimization algorithms such as genetic algorithms or particle swarm algorithms.

[0022] Further, the operating parameters of the natural gas purification device are the parameters that affect the production and consumption of the energy-carrying medium by the purification device.

[0023] Through the research on the natural gas purification process and purification equipment, the operating parameters closely related to the demand for the energy-carrying medium of the purification equipment in the purification process are found, and only these parameters are optimized for the lowest comprehensive energy consumption of the process, thus eliminating the interference of the parameters that have no influence on the demand for the energy-carrying medium, simplifying the model, and increasing the optimization operation efficiency.

[0024] Further, the sum of the demand for the energy-carrying medium, energy consumption, and material consumption is obtained by weighted summation of various consumed energies and substances according to their respective energy conversion coefficients.

[0025] Further, the prediction model A is a neural network model.

[0026] Further, the intelligent algorithm B is a genetic algorithm.

[0027] The prediction model A can be any machine learning model. By training the machine learning model with a large amount of historical data, based on this model, the demand for energy-carrying media and other energy and material consumptions can be predicted under different raw material gas parameters and purification device operation parameters.

[0028] The intelligent algorithm B can adopt iterative optimization algorithms such as genetic algorithms or particle swarm algorithms.

[0029] Further, when multiple purification series operate in parallel with non-full load, the purification series with lower unit comprehensive energy consumption under the same raw material natural gas parameters undertakes more purification tasks; the unit comprehensive energy consumption is the ratio of the process comprehensive energy consumption to the corresponding raw material natural gas treatment volume.

[0030] In the natural gas purification series, the process is complex with numerous purification devices, and the operation times of the purification devices are different, resulting in certain individual differences between the purification series and different energy consumption performances of each purification series (for example, in the purification process with the same raw material gas parameters, some purification series have higher energy-carrying media or other energy and material consumptions, while some are lower). Therefore, after separately modeling and optimizing each purification series, the purification series with better energy consumption performance undertakes more purification tasks. After such load optimization, the energy consumption of the whole plant is further reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 is the flowchart of the optimization algorithm B for the optimal process comprehensive energy consumption under certain raw material gas parameters;

[0032] Figure 2 is the flowchart of the optimization algorithm D for the optimal utility energy consumption under a certain demand for energy-carrying media;

[0033] Figure 3 is the flowchart of the method of the present invention;

[0034] Figure 4 is the purification process flowchart of the combined device of a high-sulfur natural gas purification plant;

[0035] Figure 5 is the schematic diagram of the structure of a typical artificial neural network model;

[0036] Figure 6 is the schematic diagram of the comparison between the prediction results of the neural network prediction model A and the actual values of historical data;

[0037] Figure 7 is the schematic diagram of the relative error distribution of the prediction results of all samples of the neural network prediction model A;

[0038] Figure 8 It is a schematic diagram of Process B of an iterative optimization algorithm for the lowest comprehensive energy consumption based on a genetic algorithm;

[0039] Figure 9 It is the raw natural gas with a flow rate of 100 kNm 3 / h and a pressure of 7.80 Mpa, showing the result of iterative calculation using the optimization algorithm B;

[0040] Figure 10 It is a schematic diagram of the calculation result of the unit comprehensive energy consumption of the combined unit;

[0041] Figure 11 It is a schematic diagram of the calculation result of the medium-pressure steam output of the combined unit;

[0042] Figure 12 It is a schematic diagram of the calculation result of the consumption of low-pressure steam in the combined unit. Specific implementation mode

[0043] The following further describes the present invention in detail with reference to the accompanying drawings.

[0044] A natural gas purification process control method provided in this embodiment is as follows. First, as Figures 1 - 2 shown, 1.1) Based on historical data, a prediction model A of the consumption of energy-carrying working media such as steam produced and consumed at the battery limit, as well as the consumption of electricity, fuel gas, purified water, nitrogen, etc., by the operating parameters of different purification devices under different fuel gas parameters is established. 1.2) Then, according to the raw gas parameters such as the processing volume of the raw natural gas to be processed, the operating parameters of the purification device and the consumption of energy-carrying working media, etc., with the lowest total energy consumption in the purification process are obtained by optimization based on the prediction model A, and the optimization process uses algorithm B. 2.1) At the same time, based on historical data, a prediction model C of the supply amount of energy-carrying working media in the utility engineering corresponding to the operating parameters of the utility engineering equipment is established. 2.2) Then, according to the consumption of energy-carrying working media (as the supply amount of the utility engineering) obtained in 1.2), the operating parameters of the utility engineering with the lowest utility engineering energy consumption are obtained by optimization based on the prediction model C, and the optimization process uses algorithm D. The utility engineering generally includes multiple parallel-running devices that produce energy-carrying working media. The operating parameters of the utility engineering here should include the operating combination of the utility engineering equipment (not all equipment is running and producing energy-carrying working media), and the operating parameters of the corresponding equipment (or the shutdown of the equipment is also a kind of operating parameter). Therefore, the historical data should include the supply amount of energy-carrying working media, the corresponding operating combination of the utility engineering equipment, and the operating parameters of the corresponding equipment.

[0045] In the natural gas purification process, first, the raw natural gas flow rate (processing capacity), raw gas pressure, composition data, etc. are obtained as raw gas parameters. According to the working principle of the natural gas purification process, the operating parameters of the natural gas purification device are determined. Then, based on the operating parameters and operating status of the natural gas purification device, the energy-carrying working medium (such as low-pressure steam), energy (electricity, fuel gas), and materials (nitrogen, purified water) required by the purification device are determined. Next, the energy-carrying working medium, energy, and materials required by the utility engineering supply are controlled. Energy and materials can be directly supplied according to demand and quantity, but for the steam in the energy-carrying working medium, the parameters of the boiler in the utility engineering need to be controlled and adjusted to obtain and supply the required steam.

[0046] In the method of the present invention, as Figure 3 shown in 3.1 - 3.2), first, the current raw gas parameters are obtained and optimized using Algorithm B to obtain the operating parameters of the purification device, the demand for the energy-carrying working medium, the demand for energy, and the demand for materials when the total energy consumption of the entire purification process is the lowest under the current raw gas purification treatment. The lowest total energy consumption of the process means that the total consumption of the energy-carrying working medium, energy, and materials consumed is the lowest. 3.3) The demand for the energy-carrying working medium of the purification device obtained by optimizing with Algorithm B is used as the output quantity and optimized using Algorithm D to obtain the operating parameters of the corresponding facilities of the utility engineering (such as the boiler) when the energy consumption of the utility engineering is the lowest (and the material consumption is also the lowest) to meet the demand output of the energy-carrying working medium. 3.4) The operating parameters of the purification device and the operating parameters of the utility engineering obtained by the two optimizations respectively are used as control targets to control and adjust the purification device and the relevant facilities of the utility engineering to make the corresponding parameters reach the control target or approach the control target; at the same time, it meets the demand for energy and materials of the purification device obtained by the optimization algorithm B, and meets the energy and material requirements for the energy-carrying working medium output by the utility engineering. Finally, the integrated optimization of the high-sulfur natural gas purification device and the energy consumption of the utility engineering is realized, and the prediction and control of the lowest energy consumption of the natural gas purification device are realized.

[0047] In the natural gas purification process, the energy-carrying working medium, electricity, fuel gas, nitrogen, purified water, etc. required by the purification device are all provided by the utility engineering. Energy and materials such as electricity, fuel gas, nitrogen, and purified water can be accurately supplied according to quantity. For the steam as the energy-carrying working medium, etc., the output needs to be adjusted by controlling the boiler. Therefore, the operating parameters of the utility engineering in the prediction model C and the optimization algorithm D for the output quantity of the energy-carrying working medium in the present invention refer to the corresponding operating parameters of the equipment related to the generation of the energy-carrying working medium (such as the temperature of the boiler, the rotation speed of the water pump, etc.); energy and materials such as fuel gas, nitrogen, and purified water are directly supplied by the utility engineering according to quantity.

[0048] Meanwhile, those skilled in the art should understand that due to the certain lag and oscillation problems in the adjustment of operating parameters towards the control target, as well as the reasonable errors that may exist in the model, the supply of energy-carrying working fluids, energy, and materials in the control should have a certain redundancy, and this redundancy is different from excessive supply.

[0049] Specifically, the method of the present invention includes seven parts. The selected natural gas purification plant process is a typical high-sulfur natural gas purification process, and the purification process mainly includes five main units such as MDEA acid gas removal, TEG dehydration, conventional Claus sulfur recovery, hydrogenation reduction tail gas treatment, and low-pressure acid water stripping, as well as utilities.

[0050] 1. Based on the historical big data between the raw gas parameters, the operating parameters of the natural gas purification device, and their corresponding consumption of energy-carrying working fluids, establish a prediction model A for the consumption of energy-carrying working fluids of the natural gas purification device to predict the consumption of energy-carrying working fluids under different working conditions.

[0051] Taking a high-sulfur natural gas purification plant as an example, the main process flow of the purification process of this purification plant is as Figure 4 shown. The figure shows the process flow of a combined device with two purification series (I and II). The prediction model A and the optimization algorithm B are both for a complete purification series. The high-sulfur natural gas purification process is as follows: The raw natural gas is passed through the acid gas removal unit to remove hydrogen sulfide, partial organic sulfur, carbon dioxide, etc. using MDEA lean amine solution; after dehydration in the dehydration unit, it meets the requirements of the product gas and is transported out through the long-distance pipeline network; the acid gas generated in the acid gas removal unit enters the sulfur recovery unit, is mixed with air and enters the reaction furnace for reaction to recover the sulfur element in it as liquid sulfur, and industrial sulfur is produced through the sulfur forming unit. The reaction gas in this process generates medium-pressure saturated steam through the waste heat boiler; the tail gas of the sulfur recovery unit is treated in the tail gas treatment unit, and after meeting the environmental protection requirements, it is transported to the tail gas incinerator for incineration, and the flue gas is discharged through the chimney; the acid water generated in the purification process is sent to the acid water stripping unit, the purified water generated is recycled, and the acid gas generated by stripping is transported to the tail gas treatment unit for treatment.

[0052] (1) Determine the items of key influencing parameters that affect the steam production, consumption, energy consumption, and material consumption in the purification process.

[0053] If one wants to build a model to predict the demand and energy consumption of energy-carrying working fluids (low-pressure steam), it is necessary to find out the parameters in the operating parameters of the purification device that affect steam production, consumption and energy consumption (i.e., key influencing parameters). In the purification process of this embodiment, there are also devices in the purification device that can generate steam. For example, some waste heat boilers can generate medium-pressure steam, and the medium-pressure steam can be converted into low-pressure steam. Therefore, only the public works need to provide the remaining required low-pressure steam). According to the natural gas purification process flow and working principle, combined with the established process flow simulation model for parameter sensitivity analysis, the key equipment and corresponding influencing parameters related to energy consumption and the production and consumption of medium- and low-pressure water vapor of concern can be determined. That is, to find out the key influencing parameters that have a greater impact on steam production, consumption, energy consumption and material consumption. First of all, it is certain that the parameters of the raw natural gas to be processed (flow rate, composition, etc.) will have an important impact on energy consumption, material consumption and the demand for energy-carrying working fluids. Therefore, all or part of these key influencing parameters, excluding the raw gas parameters, can be used as the operating parameters of the natural gas purification device to achieve the control of the natural gas purification device.

[0054] Specifically, taking the acid gas removal unit as an example, MDEA solution is widely used as the absorbent for acid gas. Generally speaking, the process flow is that the raw natural gas contacts the lean amine solution countercurrently in the absorption tower to remove hydrogen sulfide and carbon dioxide. The rich amine solution at the bottom of the tower enters the amine solution regeneration tower after expansion and pressure reduction, flashing, and heat exchange with the lean amine solution. The reboiler in it consumes a large amount of low-pressure steam to desorb the hydrogen sulfide and carbon dioxide absorbed in the rich amine solution and regenerate the lean amine solution. The semi-lean amine solution generated by the tail gas treatment unit for treating the acid gas produced by acid water stripping is also recycled back to the amine solution regeneration tower for regeneration. Focusing on steam consumption, the reboiler in this unit is the main steam-consuming equipment. After parameter sensitivity analysis based on the simulation model, the key parameters affecting the low-pressure steam consumption are the lean amine solution flow rate, the ratio of low-pressure steam volume to lean amine solution volume, and the outlet temperature of the air cooler at the top of the amine solution regeneration tower. The determination of the main energy-consuming equipment and key parameters of other units is similar and will not be elaborated here.

[0055] In summary, according to the basic principle of the purification process and the on-site operation parameters, the key influencing parameters are determined as follows:

[0056] Raw natural gas flow rate, raw natural gas pressure, raw natural gas composition ratio (if the raw natural gas composition remains basically the same for a long time, this item does not need to be used as a key influencing parameter), lean amine liquid flow rate in the desulfurization and decarbonization unit, low steam consumption per unit lean amine liquid flow rate in the desulfurization and decarbonization unit, outlet temperature of the air cooler of the amine liquid regeneration tower in the desulfurization and decarbonization unit, steam drum pressure of the Claus waste heat boiler in the sulfur recovery unit, molar ratio of hydrogen sulfide to sulfur dioxide in the tail gas of the sulfur recovery unit, air distribution ratio of the hydrogenation furnace in the tail gas treatment unit, furnace temperature of the tail furnace in the tail gas treatment unit, oxygen content at the outlet of the tail furnace in the tail gas treatment unit, and steam temperature at the outlet of the waste heat boiler in the tail gas treatment unit. Among them, the raw natural gas flow rate, raw natural gas pressure, and raw natural gas composition ratio are used as raw natural gas parameters.

[0057] In the natural gas purification process of this embodiment, the purification device unit, equipment related to steam production and consumption, medium and low pressure steam production and consumption conditions, and key influencing parameter conditions affecting low-pressure steam demand are shown in the following table.

[0058] Table 1 Steam production and consumption table of the main units in the natural gas purification process

[0059]

[0060] (2) Calculation of the process comprehensive energy consumption and the unit comprehensive energy consumption of the process.

[0061] Calculate the unit comprehensive energy consumption of the purification device, which is the ratio of the total energy consumption of the process (process comprehensive energy consumption) to the raw natural gas processing volume. Here, it is marked, with the unit of MJ / Nm 3 , and the calculation formula is:

[0062]

[0063] Among them, M 原料气 is the raw natural gas processing volume of the process, with the unit of 10 4 Nm 3 / t; E 过程 is the process comprehensive energy consumption, with the unit of MJ / t, including the comprehensive total consumption of energy carriers, various energies and materials in the purification process, mainly including total power consumption, fuel gas consumption, nitrogen consumption, fresh water consumption, energy carrier consumption, etc., which is obtained by weighted summation according to the respective energy conversion coefficients in accordance with national standards and is calculated by formula (2):

[0064]

[0065] Among them, E1, E2, E3... E N is the value after weighted conversion of the unified dimension of each type of consumption.

[0066] (3) Based on intelligent algorithm machine learning, establish a prediction model A for the consumption of energy-carrying working fluids and energy and material consumption in the natural gas purification device.

[0067] Select the historical data collection interval, such as from January to March 2019, and collect the key influencing parameters and the production and consumption data of the energy-carrying working fluid consumption, energy consumption, and material consumption of the purification device corresponding to the time. Specifically, a set of the above-mentioned data can be collected at every set time interval.

[0068] Exclude the shutdown and abnormal data (if any) in the historical data. According to the effective historical data of a single purification series (the historical data of qualified natural gas produced after purification), use the artificial intelligence algorithm to obtain a prediction model A with the key influencing parameters as independent variables and the demand for energy-carrying working fluids, energy, and materials as dependent variables. In this embodiment, an artificial neural network is used to establish the prediction model A. The artificial neural network is a kind of simplification, abstraction, and simulation of the biological neural structure and is an empirical modeling tool. It can learn the complex relationship between input and output from the data collected in a specific problem domain and has good performance in accurate prediction and classification. Currently, it has been widely used in many fields of engineering applications. The multi-layer perceptron artificial neural network is one of the most widely used neural networks in various fields of engineering problems, including an input layer, a hidden layer, and an output layer, and each layer is composed of neurons. The number of neurons in the input layer is equal to the number of input parameters, that is, the number of key influencing parameters; the number of neurons in the output layer is equal to the number of prediction targets. Here, the prediction targets are the demand for energy-carrying working fluids, energy demand, and material demand. The hidden layer can be composed of one or more layers, and the number of layers and neurons in the hidden layer can be obtained by the trial-and-error method or combined with intelligent algorithms for optimization.

[0069] Figure 5 Shows a typical fully connected network structure. This model receives data input through the input layer nodes / neurons, transfers it to the hidden layer nodes, and finally transfers the information to the output nodes. Neurons are connected to any neuron in the next layer through communication links associated with connection weights (Synaptic Weight). Each neuron receives the outputs of the neurons in the previous layer, sums them weighted by the connection weights, and then adds a bias to calculate the single output of this neuron through an activation function. To adapt to specific data / problems, it is necessary to configure and train the corresponding artificial neural network model. The training process can be regarded as minimizing the error between the expected output and the actual output of the model. The training of the artificial neural network model is a process of introducing randomly selected samples with input data and expected output into the artificial neural network configuration model, determining the error between the expected output value and the actual output of the model, and minimizing the error by modifying / optimizing the connection weights and biases of the neurons.

[0070] Artificial neural network prediction model A. The inputs of the model are key influencing parameters, and the outputs of the model are the predicted demand for energy-carrying media, energy demand, and material demand. The historical operation data from January to March 2019 was used to establish the model samples, and a set of effective historical data was extracted every hour. The obtained sample data was randomly divided into three categories: as the training set, validation set, and test set, with proportions of 70%, 15%, and 15% respectively.

[0071] Before inputting the sample data into the neural network model for training, the input data was normalized to the range of [-1, 1] using the minmax function in Matlab. The transfer function of the hidden layer is the sigmoid function "tansig", the transfer function of the output layer is the linear function "purelin", and the training function of the network uses the "trainlm" function, which updates the weights and bias values based on the Levenberg-Marquardt algorithm. A single hidden layer network was used to establish the model, and the trial-and-error method was used to find the optimal number of hidden layer neurons. The network performance function uses the mean square error (MSE), and the correlation between the expected data and the actual output of the network was measured by the regression R value. Since the samples used to train the network and the initial values of the network weights and biases are randomly selected and generated, the training results of the neural network with the same structure are also different each time. Therefore, each structure of the neural network was trained 5 times, and the average value of the mean square error was taken. The average mean square error values calculated for different structures of the neural network are shown in Table 2. It can be seen that when the number of hidden layer neurons is 18, the mean square error value and the correlation coefficient value of the test set are both relatively ideal, so it was selected for subsequent research.

[0072] Finally, the neural network prediction model A with a structure of 11-18-5 was constructed; 11 is the number of neurons in the input layer, corresponding to 11 key influencing parameter input quantities; 18 is the number of neurons in the hidden layer; 5 is the number of neurons in the output layer, corresponding to the five output quantities of the demand for energy-carrying media, power consumption, fuel gas demand, purified water demand, and nitrogen demand as energy and material consumption.

[0073] Table 2 Calculation results of neural network outputs under different numbers of neurons in the hidden layer

[0074]

[0075]

[0076] After verification using the test set and validation set, in order to quantify the difference between the prediction results of the energy-carrying media consumption and the energy and material consumption prediction model A and the true values, the average relative deviation (AAD%) was defined and calculated by the following formula, where y i , x i and n represent the true value, the calculated value of the network model, and the number of samples respectively.

[0077]

[0078] The mean square error (MSE) of all samples calculated according to Model A is 0.005, and the average relative deviation is 1.9%. Among them, the proportion of samples with relative deviation in the interval [-1%, 1%] is 33.7%, the proportion of samples in the interval [-3%, 3%] is 79.1%, and the proportion in the interval [-5%, 5%] is 94.8%. The model accuracy is relatively reliable and can be used for engineering applications. Figure 6 It is a comparison chart between the predicted value and the historical actual value of the unit comprehensive energy consumption of the process under different working conditions, that is, different key operating parameters. Figure 7 It is the distribution of the relative error of the predicted value. The error rate is within ±5%, and the overall prediction accuracy is good.

[0079] 2. Based on the prediction model A obtained in step 1, an optimization algorithm B is established. Using the iterative optimization algorithm B, under the given raw gas parameters, the operating parameters of the purification device, the consumption (demand) of the energy-carrying medium of the purification device, the energy demand, and the material demand in the raw gas purification process with the lowest energy consumption are obtained.

[0080] In this embodiment, the genetic algorithm is used for iterative optimization, and the algorithm flow is shown in Figure 8 . The basic elements of the genetic algorithm mainly include: chromosome coding for a specific problem to be solved, population initialization, individual fitness calculation, selection operation, crossover operation, and mutation operation. Starting from an initial population, the genetic algorithm undergoes multiple generations of evolution and finally converges to one or several individuals with the best fitness, thereby obtaining the optimal solution or satisfactory solution to the problem.

[0081] In this embodiment, the individuals in the initial population correspond to different key operating parameters (the key operating parameters include raw gas parameters and purification device operating parameters), and the fitness is the process comprehensive energy consumption defined in step 1(2). The prediction model A and formula 1 are used as the fitness function, and the raw gas parameters among the key influencing parameters do not participate in the mutation as a constraint condition; calculate the fitness of the individuals, compare the individual fitness values, and then change the purification device operating parameters in the individuals through crossover and mutation until the minimum fitness value is found, that is, the operating parameters of the purification device, the energy-carrying medium demand, the energy demand, and the material demand of the purification device when the unit comprehensive energy consumption is the lowest under different raw gas parameters.

[0082] In this embodiment, the initial population size, the number of iterations, the crossover probability, and the mutation probability of the designed genetic algorithm are set to 300, 200, 0.5, and 0.05 respectively. Taking the individual with a raw natural gas treatment volume of 100 kNm 3 / h and 7.80 Mpa (raw gas parameters) as an example, the change curve of the best fitness value during the iterative calculation process of the genetic algorithm is asFigure 9 As shown, it can be seen that when the number of iterations reaches 130 generations, the calculation results tend to converge and remain stable. The output result after the iteration is the optimal consumption of energy-carrying working medium, energy consumption, and material consumption under the raw gas parameters of 100 kNm 3 / h and 7.80 MPa. The variation result of the corresponding key operation parameters (operation parameters of the purification device) is the operation parameters of the purification device to achieve the optimal unit comprehensive energy consumption.

[0083] 3. Using the optimization algorithm B based on the genetic algorithm for the raw gas parameter conditions such as the treatment volume of raw natural gas in the actual operation of the natural gas purification plant, the optimal operation parameters of the natural gas purification device, as well as the corresponding consumption of energy-carrying working medium, energy consumption, and material consumption are obtained.

[0084] Allocate the current treatment volume of raw natural gas in the actual operation of the natural gas purification plant to each purification series of the combined device. Each purification series, according to its own treatment volume and other raw gas parameters, through the optimization algorithm B aiming at the lowest comprehensive energy consumption of the natural gas purification device, obtains the optimal process comprehensive energy consumption and process unit comprehensive energy consumption corresponding to the treatment volume of each purification series. Figure 10 For different raw natural gas flow rates, the total optimal process unit comprehensive energy consumption of the combined device calculated by the optimization algorithm B is shown as black data points. In on-site engineering applications, according to the calculation results of the optimization algorithm, the relationship between the raw natural gas flow rate and the process unit energy consumption can also be fitted. The fitting curve in this embodiment is shown in Figure 10 the curve in. It can be seen from the curve that as the treatment volume of raw natural gas increases, the unit comprehensive energy consumption of the combined device decreases, and the trend is very consistent with the on-site actual operation.

[0085] The essence of the consumption or demand of the energy-carrying working medium of the purification device, that is, the demand for steam, is the total demand for steam of the purification device during the purification process minus the steam generation amount of the purification device during the purification process. Figure 11 And Figure 12 are the optimized values of the medium-pressure steam output and the low-pressure steam consumption and the fitted curves corresponding to the purification process obtained based on the optimization algorithm B under different raw gas flow rates.

[0086] The optimization algorithm B constructed in the present invention aiming at the lowest comprehensive energy consumption of the natural gas purification device is for example Figure 4One of the purification series, Purification Series I or Purification Series II. When multiple purification series are operating in parallel, it is necessary to calculate the optimal operating parameters of the purification device, the consumption of energy-carrying media, and the consumption of energy materials for each purification series under the raw gas parameters such as the flow rate of the raw gas processed by this series. Due to the differences between the purification series, the optimization results of each purification series according to the optimization algorithm B based on historical operation data will also have significant differences, that is, under the same raw gas parameters, the optimal operating parameters of the purification device, the consumption of energy-carrying media, and the consumption of energy materials of different purification series are not the same. Therefore, when a natural gas purification plant with multiple purification series is operating at a non-full load, in order to maximize energy conservation, the principle of load optimization distribution for different series is: The purification series with a lower unit comprehensive energy consumption under the same raw gas flow rate should undertake more natural gas purification tasks, but not exceed its maximum processing capacity.

[0087] According to the method of the present invention, in this embodiment, the load optimization distribution scheme of three combined devices in six series of a purification plant under different raw natural gas loads is calculated. The specific calculation results are shown in Table 3. Compared with the actual operation data, according to the optimized distribution, the energy-saving potential under the conditions of Table 3 for different series corresponding to the processing volume is 4% - 10%.

[0088] Table 3 Calculation results of load optimization distribution of combined devices under different total processing amounts

[0089] Total throughput 111 columns 112 columns 121 columns 122 columns 131 columns 132 columns <![CDATA[kNm 3 / h]]> <![CDATA[kNm 3 / h]]> <![CDATA[kNm 3 / h]]> <![CDATA[kNm 3 / h]]> <![CDATA[kNm 3 / h]]> <![CDATA[kNm 3 / h]]> <![CDATA[kNm 3 / h]]> 400 109.5 104.8 0 0 66.8 119 500 129.5 125.3 0 117.7 0 127.5 600 125.7 121.7 0 109.6 114.9 128.0 700 123.1 122.3 95.9 112.9 121 124.7

[0090] After the load (processing volume, that is, the flow rate of the raw gas) of each purification series of the combined device is distributed, the raw gas parameters are brought into the optimization algorithm B of each purification series to calculate the optimal operating parameters of the purification device, the demand for energy-carrying media, the demand for energy, and the demand for materials of each purification series. According to the optimal operating parameters of the purification device, each purification series is controlled. A natural gas purification plant usually configures a utility engineering system. Therefore, the production of energy-carrying media by the utility engineering is controlled according to the sum of the demands for energy-carrying media of each purification series, and is distributed to each purification series according to the calculated demands. At the same time, the energy and material demands calculated are provided and distributed.

[0091] Table 4 shows the production and consumption of low-pressure steam in a combined device (two purification series).

[0092] Table 4 Production and consumption of low-pressure steam in the combined device

[0093]

[0094] 4. Use the historical big data of the operating parameters of the utility engineering equipment and the flow rate of the energy-carrying media supplied to establish a prediction model C of the operating parameters of the utility engineering equipment corresponding to different supply amounts of the energy-carrying media.

[0095] First, identify the key utility equipment and operating parameters related to the production of energy-carrying fluid steam. A typical boiler energy balance relationship is shown in Equation (3). Among them, FB u,c,t and SB u,c,t represent the boiler fuel gas consumption and steam production respectively, H c represents the fuel enthalpy value of fuel c, that is, the lower calorific value of natural gas, η u,c represents the energy efficiency of boiler u when consuming fuel c, H stm and H wat represent the enthalpy values of the boiler outlet steam and the inlet boiler water.

[0096] FB u,c,t ·H c ·η u,c =(H stm -H wat )SB u,c,t (3)

[0097] The steam-driven equipment of the utility includes steam-driven feed water pumps, steam-driven circulating water fields, and steam-driven air separation and air compression, and cooperates with the supporting electric-driven equipment to jointly provide the driving force required for boiler feed water, circulating water, and compressed air. The energy consumption models of the boiler feed water pump unit, circulating water pump unit, and air compressor unit are expressed as follows:

[0098] ∑ u∈U B u,c,t ·FT u,c,t (hi stm -ho stm )η u,c +∑ v∈V B v,c,t ·ET v,c,t ·η v,c =XET c,t (4)

[0099] Among them, FT u,c,t represents the steam consumption of steam turbine u, hi stm and ho stm represent the enthalpy values of the inlet steam and the outlet steam respectively, η u,c represents the steam-driven efficiency, ET v,c,t represents the indicated power of the motor, η v,c represents the efficiency of the motor, XET c,t represents the shaft power required to drive the boiler feed water, circulating water, or compressed air, B u,c,t and B v,c,t are 0-1 variables.

[0100] The material balance and energy conservation equations of the desuperheater and pressure reducer are shown in (5) and (6):

[0101] FOu,c,t = FI u,c,t + FI′ u,c,t (5)

[0102] FO u,c,t · ho stm = FI u,c,t · hi stm + FI′ u,c,t · hi wat (6)

[0103] wherein, FI u,c,t is the flow rate of the inlet steam (and high-pressure steam) of the attemperating and pressure-reducing valve during the period t, hi stm is the enthalpy value of the inlet steam, FO u,c,t is the flow rate of the steam at the outlet of the attemperating and pressure-reducing valve (i.e., low-pressure steam), ho stm is the enthalpy value of the outlet steam, FI′ u,c,t and hi wat are respectively the flow rate and enthalpy value of the water required by the attemperating and pressure-reducing valve.

[0104] In the above equipment, the efficiencies of the boiler, steam drive and pump are all given by fitting historical big data. The modeling method adopts the neural network algorithm, and the process is similar to the purification process, which will not be elaborated here.

[0105] The operating parameters of the utility engineering equipment related to the generation of the energy-carrying working medium, i.e., steam, found in this embodiment include the fuel gas consumption of the power station boiler, the operating status of the machine pumps, the efficiency of the machine pumps, the steam flow rate of the attemperating and pressure-reducing valve, the steam consumption of the steam turbine, the power consumption of the power-consuming equipment, etc. The historical big data of the flow rate of the energy-carrying working medium supplied by the utility engineering includes the low-pressure steam flow rate.

[0106] Select the historical data acquisition interval, such as January to March 2019, and collect the operating parameters of the key utility engineering equipment put into use and the key equipment, and the quantity of the energy-carrying working medium produced in the utility engineering corresponding to the time.

[0107] Establish a prediction model C for the quantity of the energy-carrying working medium supplied corresponding to the key utility engineering equipment and operating parameters according to the above historical big data.

[0108] 5. Based on the prediction model C for the consumption of the energy-carrying working medium by the utility engineering equipment, establish an optimization algorithm D. Using the optimization algorithm D, when a given demand for the energy-carrying working medium is given, obtain the key utility engineering equipment and operating parameters with the optimal energy consumption.

[0109] This embodiment adopts the intelligent optimization algorithm Particle Swarm Optimization (PSO). The Particle Swarm Optimization algorithm is an evolutionary computing technique, which originates from the study of the foraging behavior of bird flocks. Its basic idea is to find the optimal solution through the cooperation and information sharing among individuals in the group. The basic principle is that PSO is initialized as a group of random particles (random solutions), and then the optimal solution is found through iteration. In each iteration, the particle updates itself by tracking two "extreme values" (pbest, gbest). After finding these two optimal values, the particle updates its velocity through the following formula (7). The first part in the formula represents the influence of the previous velocity magnitude and direction; the second part is a vector pointing from the current point to the particle's own best point, indicating the part of the particle's action that comes from its own experience; the third part is a vector pointing from the current point to the best point of the population, reflecting the collaborative cooperation and knowledge sharing among particles. c1 and c2 are the self-learning factor and the group learning factor respectively, and rand is a random number between 0 and 1.

[0110] v i =v i +c1×rand()×(pbest i -x i )+c2×rand()×(gbest i -x i ) (7)

[0111] On this basis, the position update of the particles in the continuous particle swarm is realized through the following formula.

[0112] x i =x i +v i

[0113] And the position update of the particles in the discrete particle swarm is realized through the following formula.

[0114]

[0115]

[0116] In the design of this embodiment, the output of the three gas boilers adopts continuous variables, and the medium-pressure boiler feed pumps (a total of 2 steam-driven and 3 motor-driven), the air compressor station compressors (a total of 3 steam-driven and 1 motor-driven), and the circulating water pumps (a total of 7 steam-driven and 7 motor-driven) adopt binary variables. The utility energy consumption can be calculated based on the key equipment participating in the operation to generate energy-carrying working fluids and the equipment operation parameters, or obtained from the energy consumption data in historical data. If obtained from historical data, the historical statistical data of the utility energy consumption corresponding to the utility control parameters needs to be added to the prediction model C in step 4, that is, the prediction model C of the operation parameters of the utility equipment and the utility energy consumption corresponding to different energy-carrying working fluid supply amounts is established in step 4. Using the lowest comprehensive utility energy consumption as the fitness objective function, the particle swarm algorithm is used for optimization and solution, which is the optimization algorithm D for the lowest utility energy consumption.

[0117] In step 3 above, according to the method of the present invention, the optimal purification device operation parameters, the energy-carrying working fluid demand, the energy demand, and the material demand corresponding to each purification series of the natural gas purification plant under the corresponding working conditions (that is, under the raw gas parameters) are calculated. The total energy-carrying working fluid demand is substituted into the optimization algorithm D for the lowest utility energy consumption to obtain the operation combination of the utility equipment with the lowest energy consumption, as well as the operation parameters such as the operation status of the key equipment and the flow control. Then, the obtained utility operation parameters are used as the control target, and the corresponding equipment is adjusted and controlled according to the optimal utility equipment combination to approach and reach this control target, that is, the energy-carrying working fluid demand of the purification series is met under the condition of the lowest energy consumption.

[0118] The operation combinations of the key utility equipment and the corresponding raw gas volumes that can be processed under different low-pressure steam supply amounts (energy-carrying working fluid supply amounts) calculated are shown in Table 5.

[0119] Table 5 Results of the operation combinations of the key utility equipment under different natural gas processing volumes

[0120]

[0121]

[0122] Generally speaking, the method of the present invention needs to first establish an optimization algorithm model B with the lowest comprehensive energy consumption per unit of the purification device and an optimization algorithm model D with the lowest utility energy consumption according to the above steps 1, 2, 4, and 5. First, optimize the load distribution of each purification series according to the amount of fuel gas to be processed in the whole plant. According to the load, obtain the raw gas parameters (such as raw gas flow rate) of the corresponding purification series, and substitute the raw gas parameters into the optimization algorithm model B of the purification series to obtain the optimal operation parameters of the purification device, the demand for energy-carrying media, the energy demand, and the material demand of each purification series. Then, substitute the sum of the energy-carrying medium demands of each purification series into the optimization algorithm model D of the utility engineering of the plant to obtain the best equipment combination and operation parameters of the utility engineering. Then, control the purification device of the corresponding purification series according to the optimal operation parameters of the purification device, control the corresponding equipment of the utility engineering to generate sufficient energy-carrying media according to the best equipment combination and operation parameters of the utility engineering and distribute them according to the demands of each purification series, and at the same time control the utility engineering to supply the required energy and materials and distribute them according to the demands of each purification series.

[0123] The utility engineering of a natural gas purification plant needs to provide the necessary energy and materials for the purification process, such as low-pressure steam, purified circulating water, and necessary nitrogen, etc. It is the main unit of the energy-carrying medium consumption and also the key point and difficulty of energy conservation and consumption reduction. This is because, on-site, it is necessary to obtain the operating characteristics of key equipment in the utility engineering, such as boilers, steam-driven steam turbines, and pressure-reducing valves, based on historical operating data, and at the same time, establish an optimal operation strategy for the utility engineering according to the dynamic needs of the series purification units or combined devices for energy and material consumption, schedule the start and stop of equipment, and achieve coordinated optimization and precise allocation with the purification process. The difficulty of optimal operation is that there are many operating equipment, and it is difficult to incorporate the operating performance and startup of the equipment into the optimization strategy at the same time, which is difficult to achieve by classical mathematical analysis methods.

[0124] In summary, to deeply explore the energy-saving potential of a natural gas purification plant, two problems need to be solved. One is the optimization problem of the production scheduling of the series purification units, and the other is the precise allocation problem of the utility engineering. A natural gas purification process control method provided by the present invention determines the key influencing parameters according to the working principle of the purification process, and based on the on-site historical data, establishes a prediction and optimization algorithm model for the consumption of energy-carrying media such as steam produced and consumed at the boundary of the purification device under different natural gas processing volumes; at the same time, based on the historical data, establishes a prediction model for the consumption of energy-carrying media of key equipment in the utility engineering and an optimization algorithm model for the operation combination of equipment; taking into account the energy and material consumption of both the purification device and the utility engineering, realizes the integrated optimization of the energy consumption of the natural gas purification device and the utility engineering, and achieves the lowest energy consumption control of the natural gas purification process.

[0125] Regardless of whether the scale of the natural gas purification plant has only one purification train or multiple parallel purification units, the method of the present invention can be used to analyze and mine historical operation data to obtain how to adjust the key operation parameters of the purification equipment and the operation parameters of the utility engineering under different raw natural gas flow rates, so as to minimize the total energy consumption of the purification train and the utility engineering.

Claims

1. A method for controlling a natural gas purification process, characterized in that, It includes the following steps: 1) Collect the raw natural gas parameters including flow rate, pressure and composition at present, and obtain the operation parameters of the natural gas purification device and the demand for energy-carrying working medium with the lowest overall process energy consumption under the conditions of the raw natural gas parameters through optimization by intelligent algorithm B; The operation parameters of the natural gas purification device include: the lean amine solution flow rate in the desulfurization and decarbonization unit, the low steam consumption per unit lean amine solution flow rate in the desulfurization and decarbonization unit, the outlet temperature of the air cooler of the amine solution regeneration tower in the desulfurization and decarbonization unit, the steam drum pressure of the Claus waste heat boiler in the sulfur recovery unit, the molar ratio of hydrogen sulfide to sulfur dioxide in the tail gas of the sulfur recovery unit, the air distribution ratio of the hydrogenation furnace in the tail gas treatment unit, the furnace temperature of the tail furnace in the tail gas treatment unit, the oxygen content at the outlet of the tail furnace in the tail gas treatment unit, and the steam temperature at the outlet of the waste heat boiler in the tail gas treatment unit; 2) Control the utility engineering according to the demand for energy-carrying working medium, and control the natural gas purification device according to the operation parameters of the natural gas purification device; The process of optimization by intelligent algorithm B in step 1) includes: first, initialize the population, and the population includes the operation parameters of the natural gas purification device; then calculate the fitness of the population, and the fitness is the overall process energy consumption, and the overall process energy consumption is calculated by prediction model A; finally, change the population and iterate multiple times, and finally select the operation parameters of the natural gas purification device corresponding to the population with the lowest overall process energy consumption and the corresponding demand for energy-carrying working medium as the optimization result; The prediction model A is a machine learning model, which is trained by the historical data of raw natural gas parameters, the demand for energy-carrying working medium corresponding to the operation parameters of the natural gas purification device, energy consumption, and material consumption; The overall process energy consumption is the sum of the demand for energy-carrying working medium, energy consumption, and material consumption in the natural gas purification process; in step 2), the method of controlling the utility engineering according to the demand for energy-carrying working medium is to obtain the optimized operation parameters of the utility engineering equipment with the lowest energy consumption of the utility engineering equipment when meeting the demand for energy-carrying working medium through optimization by intelligent algorithm D, and control the corresponding equipment of the utility engineering according to the optimized operation parameters of the utility engineering equipment; The process of optimization by intelligent algorithm D includes: first, initialize the population, and the population includes the operation parameters of the utility engineering equipment; then calculate the fitness of the population, and the fitness is the energy consumption of the utility engineering equipment, and the energy consumption of the utility engineering equipment is calculated by prediction model C; finally, change the population and iterate multiple times, and finally select the operation parameters of the utility engineering equipment corresponding to the population with the lowest energy consumption of the utility engineering equipment as the optimized operation parameters of the utility engineering equipment; The prediction model C is a machine learning model, which is trained by the historical data of the supply amount of energy-carrying working medium and the energy consumption of the utility engineering equipment corresponding to the operation parameters of the utility engineering equipment; the operation parameters of the utility engineering equipment include: the fuel gas consumption of the boiler in the power station, the operation status of the pump, the pump efficiency, the steam flow rate of the desuperheater and pressure reducer, the steam consumption of the steam turbine, and the power consumption of the power-consuming equipment.

2. The natural gas purification process control method according to claim 1, characterized in that, The prediction model C is a neural network model.

3. The natural gas purification process control method according to claim 2, characterized in that The intelligent algorithm D is a particle swarm algorithm.

4. The natural gas purification process control method according to claim 1 or 3, characterized in that The operation parameters of the natural gas purification device are the parameters that affect the production, consumption and energy-carrying working medium of the purification device.

5. The natural gas purification process control method according to claim 4, characterized in that, The sum of the demand for energy-carrying media, energy consumption, and material consumption is obtained by weighted summation of various consumed energies and substances according to their respective energy conversion coefficients.

6. The natural gas purification process control method according to claim 1, characterized in that, The prediction model A is a neural network model.

7. The natural gas purification process control method according to claim 6, characterized in that The intelligent algorithm B is a genetic algorithm.

8. The natural gas purification process control method according to claim 1, wherein When multiple purification series operate in parallel with non-full load, the purification series with lower unit comprehensive energy consumption under the same raw natural gas parameters undertakes more purification tasks; the unit comprehensive energy consumption is the ratio of the process comprehensive energy consumption to the corresponding raw natural gas processing volume.

Images