A Natural Gas Purification Control Method and an Online Optimization Platform Based on Energy Efficiency Evaluation

Abstract

The present invention relates to a natural gas purification control method and an online optimization platform based on energy efficiency evaluation, comprising the following steps: collecting current raw natural gas parameters, and obtaining the key operation parameters of the natural gas purification device corresponding to the optimal energy efficiency condition under the condition of the raw natural gas parameters through intelligent algorithm optimization; controlling the corresponding natural gas purification device according to the key operation parameters of the natural gas purification device; the intelligent algorithm can adopt a genetic algorithm for iterative optimization; the optimal energy efficiency condition includes a three-level evaluation standard, and the evaluation standard value 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 natural gas purification control method and an online optimization platform based on energy efficiency evaluation, belonging to the technical field of natural gas extraction and purification. Background Art

[0002] As a high-quality clean energy, natural gas has been steadily increasing in proportion in China's energy consumption structure year by year. The purification device of a natural gas purification plant is used to purify the natural gas containing impurities into qualified natural gas and achieve full-process emission compliance. However, the natural gas purification process consumes a large amount of energy, especially in the purification process of high-sulfur natural gas, which has great energy-saving potential.

[0003] The user requirements and the incoming gas conditions of a natural gas purification plant often change. Especially in the later stage of the production cycle, the operation of the purification device deviates from the design conditions. The energy-saving measures in the production process mainly rely on the dispatcher to adjust and control some operating parameters based on the load of the natural gas purification device, operation specifications, and experience, and cannot accurately control and maximize energy utilization. In order to prevent the product gas from not meeting the standards due to insufficient energy and material supply, an excessive supply method is used for production, resulting in serious energy waste problems.

[0004] At present, the energy consumption evaluation of the purification process on-site mostly uses extensive macroscopic energy consumption evaluation indicators, such as the comprehensive energy consumption for processing 10,000 cubic meters of raw gas. However, this extensive long-cycle and large-span energy consumption statistics is not conducive to grasping the operation status and energy utilization level of the purification device in the natural gas purification process, and it is also difficult to locate the weak links in energy use.

[0005] In addition to the lack of refined and scientific energy consumption evaluation methods. The difficulty in optimizing the energy consumption in the natural gas purification process also lies in that the purification energy consumption is not only affected by factors such as the flow rate, pressure, and composition of the raw gas, but also related to the operation status of each unit and key equipment of the purification device. Moreover, the operation parameters involved in the key equipment are numerous and have strong coupling and correlation, and the impact on the system energy use is severely non-linear. The relationship between the raw gas parameters and numerous key operation parameters of the equipment and the energy consumption (such as power consumption, fuel gas consumption) and material consumption (such as nitrogen, steam, or fresh water consumption) of each equipment is intricate. Under different key operation parameters of the purification equipment, different operation parameters of the purification equipment are presented, and the energy consumption and material consumption of different equipment offset each other, and at the same time, it also affects the raw natural gas processing volume per unit time and the output of by-product sulfur in the purification process. Therefore, the conventional control means relying on the experience of operators or the method of adjusting and controlling based on the feedback of extensive energy consumption evaluation is difficult to achieve the overall optimum of the whole process. Summary of the Invention

[0006] The purpose of the present invention is to provide a natural gas purification control method and an online optimization platform based on energy efficiency evaluation to solve the problem that it is difficult to further optimize the high energy consumption in the whole process of natural gas purification.

[0007] To achieve the above object, the solution of the present invention includes:

[0008] A natural gas purification control method based on energy efficiency evaluation of the present invention includes the following steps:

[0009] 1) Collect the current raw natural gas parameters, and obtain the key operation parameters of the natural gas purification device corresponding to the optimal energy efficiency condition under the condition of the raw natural gas parameters through optimization by intelligent algorithm B;

[0010] The energy efficiency conditions include one or more of the ratio of the total process energy consumption to the corresponding raw natural gas throughput, the ratio of the total process energy consumption to the economic value of the corresponding product, the sum of the unit energy consumptions of each process unit of the purification device, and the sum of the unit energy consumptions of each device of the purification device; the unit energy consumption of the process unit is the ratio of the energy consumption of the process unit to the corresponding input raw material flow rate or output product flow rate, and the unit energy consumption of the device is the ratio of the energy consumption of the device to the flow rate of the corresponding processed substance;

[0011] 2) Control the corresponding natural gas purification device according to the key operation parameters of the natural gas purification device;

[0012] The process of optimization by intelligent algorithm B in step 1) includes: first, initializing the population, where the population includes the key operation parameters of the natural gas purification device; then calculating the population fitness, where the fitness is the energy efficiency condition, and the energy efficiency condition is calculated by prediction model A; finally, changing the population and iterating multiple times, and finally selecting the key operation parameters of the natural gas purification device corresponding to the population with the optimal energy efficiency condition as the optimization result;

[0013] The prediction model A is a machine learning model, which is trained by historical data of raw natural gas parameters, key operation parameters of the natural gas purification device, and corresponding energy efficiency conditions.

[0014] In the natural gas purification process, the energy requirements and material requirements of each purification device are related to parameters such as the flow rate, pressure, and impurity content of the raw natural gas to be processed, and are also related to multiple operating parameters of the purification device. The operating parameters of the purification device are coupled with each other and also determine the output of by-product sulfur. The present invention first focuses on the entire purification process, the purification units that implement different process processes in the purification process, or the purification devices that implement specific functions to establish a more refined and targeted energy consumption evaluation system. According to different evaluation systems, through an optimization algorithm, the key operating parameters corresponding to the lowest energy consumption of the purification process under the same raw natural gas parameters in historical data are found, and each purification device is adjusted and controlled according to these key operating parameters, effectively reducing the total energy consumption in the natural gas purification process. It breaks the traditional natural gas purification plant, where each purification series unit is controlled separately according to supply and demand, and fuel materials are supplied in an excessive manner based on experience. It realizes the reduction and optimization of the energy consumption in the whole process of natural gas purification.

[0015] Further, the total process energy consumption is the sum of the energy consumption and material consumption in the natural gas purification process; the energy consumption of the process unit is the sum of the energy consumption and material consumption of the process unit in the natural gas purification process; the energy consumption of the device is the sum of the energy consumption and material consumption of the device in the natural gas purification process.

[0016] Further, the sum of the energy consumption and material consumption is obtained by weighted summation of various consumed energies and substances according to their respective energy conversion coefficients.

[0017] Further, the prediction model A is a neural network model.

[0018] Further, the intelligent algorithm B is a genetic algorithm.

[0019] The prediction model A can be any machine learning model. Through a large amount of historical data, the machine learning model is trained. Based on this model, the selected energy efficiency evaluation indicators can be predicted under different raw gas parameters and key operating parameters of the purification device.

[0020] The intelligent algorithm B can adopt iterative optimization algorithms such as genetic algorithms or particle swarm algorithms.

[0021] Further, each process unit of the purification device includes a desulfurization and decarbonization unit, a dehydration unit, a sulfur recovery unit, a tail gas treatment unit, and an acid water stripping unit.

[0022] By distinguishing each purification unit according to the process of natural gas purification and conducting energy consumption evaluation and prediction for each purification unit as a unit, it is beneficial to achieve the optimal overall process energy consumption.

[0023] Further, each device of the purification device includes a desulfurization and decarbonization pump unit, a desulfurization and decarbonization reboiler, a cooler, a regeneration tower reboiler, a Claus fan, a waste heat boiler, a reaction feed heater, a sulfur condenser, a burner, an incinerator, a tail gas treatment pump unit, and a fan.

[0024] Taking the devices that achieve different basic functions as units for energy consumption evaluation, the evaluation is more refined. At the same time, the key operation parameters of the natural gas purification device obtained are more comprehensive, the control based on the key operation parameters is more thorough and sufficient, and the energy consumption optimization adjustment is faster and better.

[0025] Further, the raw natural gas parameters include the raw natural gas flow rate, pressure, hydrogen sulfide content, and carbon dioxide content.

[0026] An online optimization platform for natural gas purification based on energy efficiency evaluation according to the present invention includes a control system. The control system is controlled and connected to the natural gas purification device to adjust the key operation parameters of each natural gas purification device; the control system also executes instructions to implement the natural gas purification control method based on energy efficiency evaluation as described above. Description of the Drawings

[0027] Figure 1 is the flowchart of the method of the present invention;

[0028] Figure 2 is the purification process flowchart of the combined device of a high-sulfur natural gas purification plant;

[0029] Figure 3 is the schematic diagram of the structure of a typical artificial neural network model;

[0030] Figure 4 is the schematic diagram of the comparison between the prediction result of the neural network prediction model A and the actual value of the historical data;

[0031] Figure 5 is the schematic diagram of the relative error distribution of the prediction results of all samples of the neural network prediction model A;

[0032] Figure 6 is the schematic diagram of the flow of the iterative optimization algorithm B based on the genetic algorithm;

[0033] Figure 7 is the schematic diagram of the result of the iterative calculation of the raw natural gas with a flow rate of 100 kNm3 / h and a pressure of 7.80 Mpa using the optimization algorithm B;

[0034] Figure 8 is the schematic diagram of the unit energy consumption reference value under different raw gas treatment capacities;

[0035] Figure 9 is the schematic diagram of the comparison between the unit energy consumption reference value and the actual unit energy consumption under different raw gas treatment capacities;

[0036] Figure 10 It is a schematic diagram of the process energy consumption factor under different raw gas treatment capacities;

[0037] Figure 11 It is the schematic diagram of the principle of the online optimization platform for natural gas purification of the present invention. Specific embodiments

[0038] The present invention will be further described in detail below with reference to the accompanying drawings.

[0039] Method embodiments:

[0040] A method for controlling natural gas purification based on energy efficiency evaluation according to the present invention establishes three - level energy efficiency evaluation indexes related to the natural gas purification process and determines the corresponding raw gas parameters and key operation parameters of purification equipment according to the on - site natural gas purification process and device operation data. The method is as Figure 1 shown, and specifically includes the following steps:

[0041] 1) First, establish energy efficiency evaluation indexes; the energy efficiency evaluation indexes are divided into three - level energy efficiency evaluation indexes: process - level energy efficiency evaluation indexes, unit - level energy efficiency evaluation indexes, and equipment - level energy efficiency evaluation indexes. Among them, the process - level energy efficiency evaluation indexes include the process unit comprehensive energy consumption and the energy consumption per 10,000 yuan of product output value; the unit - level energy efficiency evaluation indexes include the unit energy consumption of each process unit; the equipment - level energy efficiency evaluation indexes include the unit energy consumption of key equipment in each purification process. The process unit comprehensive energy consumption is the ratio of the total process energy consumption to the corresponding raw natural gas treatment capacity, the energy consumption per 10,000 yuan of product output value is the ratio of the total process energy consumption to the economic value of the corresponding product, the unit energy consumption of the process unit is the ratio of the energy consumption of the process unit to the corresponding input raw material flow or output product flow, and the unit energy consumption of the key equipment is the ratio of the energy consumption of the equipment to the flow of the corresponding processed substance.

[0042] 2) Establish prediction model A; based on historical data, establish prediction model A of the influence of key operation parameters of different purification devices on energy efficiency evaluation indexes under different fuel natural gas parameters (simply referred to as fuel gas parameters). Select standard historical data to train this prediction model.

[0043] If the selected and established evaluation index is the process unit comprehensive energy consumption, then obtain the historical key operation parameters of the purification device, the corresponding total process energy consumption, material consumption, and natural gas treatment capacity corresponding to different fuel gas parameters according to historical data; calculate the process unit comprehensive energy consumption based on the total process energy consumption, material consumption, and natural gas treatment capacity as the corresponding evaluation index.

[0044] If the selected and established evaluation index is the energy consumption per ten thousand yuan output value of the product, then based on historical data, obtain the key operation parameters of the historical purification device corresponding to different fuel gas parameters, as well as the corresponding total process energy consumption, material consumption, and the purified natural gas (finished natural gas or commercial natural gas) produced, and the process sulfur output; calculate the energy consumption per ten thousand yuan output value of the product based on the total process energy consumption, material consumption, the produced commercial natural gas, process sulfur output, current market price of the commercial unit gas, and sulfur market price as the corresponding evaluation index.

[0045] If the selected and established evaluation index is the unit energy consumption of the process unit, then based on historical data, obtain the key operation parameters of the historical purification device of each process unit under different fuel gas parameters, as well as the input raw material flow or output product flow of each process unit and the energy consumption and material consumption of the corresponding process unit; calculate the unit energy consumption of each process unit, and sum up the unit energy consumption of each process unit as the corresponding evaluation index.

[0046] If the selected and established evaluation index is the unit energy consumption of key equipment, then based on historical data, obtain the key operation parameters of the historical purification device of each key equipment under different fuel gas parameters, as well as the flow of the material processed by each key equipment and the energy consumption and material consumption of the corresponding key equipment; calculate the unit energy consumption of each key equipment, and sum up the unit energy consumption of each key equipment as the corresponding evaluation index.

[0047] 3) Establish the optimization algorithm model B; based on the prediction model A, establish the optimization model of the raw material natural gas parameters and the key operation parameters of the purification process when the corresponding evaluation index is optimal; at the same time, it can also calculate the energy consumption benchmark value of the corresponding energy efficiency evaluation index under different raw material gas parameters, that is, the theoretically optimal energy consumption value calculated under the corresponding raw material gas parameters. The difference between the energy consumption benchmark value and the on-site actual value reflects the energy-saving potential, and can also effectively evaluate the operation level of the natural gas purification process and determine the weak links in energy consumption. While determining the energy efficiency evaluation index benchmark value, the key operation parameters of the purification process obtained by the optimization algorithm model B can be used as the optimization guiding value of the control parameters of the corresponding purification equipment (step 4).

[0048] 4) Energy-saving control of the natural gas purification process; by adjusting and controlling the corresponding equipment to make its control parameters reach the key operation parameter guiding value or make its operating parameters approach the key operation parameter guiding value, the purpose of energy-saving and consumption reduction in the natural gas purification process is achieved to reach the optimal energy consumption.

[0049] The following combines examples to explain each step of the present invention in more detail.

[0050] The selected natural gas purification process in this embodiment is a typical high-sulfur natural gas purification process, mainly including five main units: MDEA desulfurization and decarbonization (acid gas removal), TEG dehydration, conventional Claus sulfur recovery, hydrogenation reduction tail gas treatment, and low-pressure acid water stripping.

[0051] Taking a high-sulfur natural gas purification plant as an example, the main process flow of the purification process in this plant is as Figure 2 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, part of the 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 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 from 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. Establish energy efficiency evaluation indicators at three levels from the entire purification process to the main units and then to the key equipment. This link mainly determines the energy efficiency evaluation indicators at all levels.

[0053] (1) Process-level energy efficiency evaluation indicators.

[0054] ① Process unit energy consumption, or process unit comprehensive energy consumption:

[0055] Process unit energy consumption represents the ratio of the total energy consumption of the process to the raw gas treatment volume. Here, it is identified, with the unit of MJ / Nm 3 , and the calculation formula is:

[0056]

[0057] Among them, M 原料气 is the process raw gas treatment volume (or flow rate), with the unit of 10 4 Nm 3 / t, E 过程 is the total energy consumption of the process, with the unit of MJ / t, and E 过程 is obtained by weighted summation of various energy consumptions and material consumptions in the process, mainly including total power consumption, fuel gas consumption, fresh water consumption, etc., according to the national standard and their respective energy conversion coefficients, and is calculated by formula (2):

[0058]

[0059] Among them, E1, E2, E3...E N are the values after weighted conversion of the unified consumption dimension for each type.

[0060] ② Energy consumption per ten thousand yuan output value of the product:

[0061] During the purification process, by-product sulfur is produced while producing commercial natural gas. This indicator is defined from the perspective of product revenue, representing the ratio of the total system energy consumption to the economic value of the product. Here, it is marked with and the unit is MJ / ten thousand yuan output value. The calculation formula is:

[0062]

[0063] Among them, M 净化气 is the output of clean natural gas in the process, with the unit of 10 4 Nm 3 / t, M 硫磺 is the output of sulfur in the process, with the unit of ton / t, and a and b are the market unit prices of commercial natural gas and sulfur, with the units of ten thousand yuan / 10 4 Nm 3 and ten thousand yuan / ton respectively.

[0064] ③ Energy consumption factor:

[0065] The energy consumption factor represents the ratio of the actual total energy consumption of the process under a certain working condition to the corresponding reference energy consumption. Here, it is marked with and is a dimensionless quantity. The calculation formula is:

[0066]

[0067] Among them, E 基准 represents the reference energy consumption value of the process. The energy consumption factor reflects the gap between the actual total energy consumption of the process under a certain raw material gas parameter and the minimum energy consumption of the process during optimal operation under the key operating parameters obtained by the optimization algorithm of the purification equipment, and is an intuitive manifestation of the energy-saving potential of the process.

[0068] (2) Unit-level and equipment-level energy efficiency evaluation indicators.

[0069] ① Unit-level unit energy consumption:

[0070] For each main purification unit in the natural gas purification process, the unit energy consumption can be used as the unit-level energy consumption evaluation indicator. Here, it is marked with e U and the unit is MJ / Nm 3 or MJ / ton. The mathematical formula is:

[0071]

[0072]

[0073] Among them, M 单元 represents the input raw material flow rate or output product flow rate of the unit, with the unit of 10 4 Nm 3 / t or ton / t, and E 单元 represents the corresponding energy consumption of the unit, with the unit of MJ / t. E 单元 is obtained by weighted summation of various energy consumptions within the unit, mainly including total power consumption, fuel gas consumption, fresh water consumption, etc., according to the national standard based on their respective energy conversion coefficients, and is calculated by Equation (6). Among them, E1, E2, E3... E N are the values after weighted conversion of the unified dimension of various types of consumption.

[0074] ② Equipment-level unit energy consumption

[0075] For the main equipment within the unit, the equipment unit energy consumption can be used as the equipment-level energy consumption evaluation index. Here, it is denoted by e D with the unit of MJ / Nm 3 or MJ / ton, and the mathematical expression is:

[0076]

[0077]

[0078] Among them, M 设备 represents the flow rate of the substance processed by the equipment, and E 设备 represents the corresponding energy consumption of the equipment, which is obtained by weighted summation of various energy consumptions, mainly including total power consumption, fuel gas consumption, fresh water consumption, etc. E 设备 is obtained according to the national standard based on their respective energy conversion coefficients and is calculated by Equation (8). Among them, E1, E2, E3... E N are the values after weighted conversion of the unified dimension of various types of consumption.

[0079] Regarding the unit-level and equipment-level energy efficiency evaluation indicators, taking an example, the desulfurization and decarbonization unit is used to remove acidic components in the raw natural gas. The energy efficiency evaluation indicators of the unit and its main internal equipment are shown in Table 1.

[0080] Table 1 Energy efficiency evaluation indicators of desulfurization and decarbonization unit and main equipment

[0081]

[0082] The dehydration unit is used to remove moisture in the wet purified gas to meet the dew point requirement of pipeline natural gas. The energy efficiency evaluation indicators of the unit and its internal equipment are shown in Table 2.

[0083] Table 2 Energy efficiency evaluation indicators of dehydration unit and equipment

[0084]

[0085] The sulfur recovery unit is used to recover sulfur elements in acidic gases. The specific energy efficiency evaluation indexes of the unit and its main equipment are shown in Table 3.

[0086] Table 3 Energy Efficiency Evaluation Indexes of Sulfur Recovery Unit and Equipment

[0087]

[0088] The tail gas treatment device is used to further improve the process sulfur recovery rate and make the SO2 in the discharged flue gas meet the standards. The specific energy efficiency evaluation indexes of the unit and the equipment inside the unit are shown in Table 4.

[0089] Table 4 Energy Efficiency Evaluation Indexes of Tail Gas Treatment Unit and Equipment

[0090]

[0091] The sour water stripping unit is used to strip the acidic components in the acidic water generated during the process. The specific energy efficiency evaluation indexes of the device are shown in Table 5.

[0092] Table 5 Energy Efficiency Evaluation Indexes of Sour Water Stripping Unit and Equipment

[0093]

[0094] 2. Using the historical operation data training method, establish the relationship prediction model A between the key operation parameters and the corresponding selected energy efficiency evaluation indexes, and predict the energy efficiency evaluation index values under different working conditions.

[0095] (1) Determine the key operation parameters and raw gas parameters that affect the energy efficiency evaluation indexes.

[0096] In addition to the uncontrollable raw gas parameters, such as raw gas flow rate, pressure, hydrogen sulfide content and carbon dioxide content, the parameters that affect the energy efficiency evaluation indexes at all levels also include the key operation parameters of the purification equipment that can be adjusted and controlled on site. In the energy efficiency evaluation system, the key issue is to determine the key operation parameters that affect the energy efficiency evaluation indexes, which are the variable parameters in the evaluation system. According to the working principle of the purification process and the on-site operation experience, the key operation parameters of each level of evaluation indexes are obtained. The difficulty lies in determining the key operation parameters that affect the process-level energy efficiency evaluation indexes. Table 6 shows the key operation parameters corresponding to the energy efficiency evaluation indexes at all levels.

[0097] Table 6 Energy Efficiency Indexes at All Levels and Their Key Operation Parameters

[0098]

[0099]

[0100] The above operation parameters can all be obtained from the operation data of the natural gas purification plant and are used for the evaluation of operation energy efficiency.

[0101] (2) Based on historical data, use the intelligent algorithm of machine learning to establish a prediction model A between the raw natural gas parameters, the key operation parameters in the natural gas purification process, and the energy efficiency evaluation indicators.

[0102] ① Collect the historical big data of the raw gas parameters, key operation parameters, and original energy consumption and material consumption, and calculate the historical big data of the corresponding energy efficiency evaluation indicators.

[0103] Select the historical data collection interval, such as from January to March 2019, and collect the raw gas parameters, the key operation parameters of each device at the corresponding time, and the total energy consumption, material consumption, and product output data of the purification plant at the corresponding time (energy consumption, material consumption, and product output can be measured by "flow rate", such as MJ / h for energy consumption, tons per hour or cubic meters per hour for material consumption and product output, etc.; or measured by the output per unit time, etc.), as well as the energy consumption, material consumption, and intermediate material processing volume (including consumption and output) of each device, etc. data; or only collect the energy consumption, material consumption, and intermediate material processing volume data of each device, and the total energy consumption, material consumption, and output of the whole plant can be calculated based on the data of each device. The energy consumption and material consumption mainly include the consumption of electricity, fuel gas, and supply water, etc., and the product output is mainly the output of purified natural gas and sulfur. The historical data can be collected in groups of the above corresponding data at every set time interval. According to the collected data, calculate the energy efficiency evaluation indicators at all levels, or calculate one or more energy efficiency evaluation indicators according to the selection. The energy efficiency evaluation indicators and the raw gas parameters and key operation parameters at the corresponding time constitute the historical big data of the evaluation indicators.

[0104] ② According to the historical big data of the energy efficiency evaluation indicators, combine the intelligent algorithm to establish a relationship prediction model A between the key operation parameters and the corresponding energy efficiency evaluation indicators.

[0105] Exclude the shutdown and abnormal data (if any) in the historical data. According to the valid historical data of a single purification series (the historical data of producing qualified natural gas after purification), an artificial intelligence algorithm is used to obtain a prediction model A with key operation parameters as independent variables and energy efficiency evaluation indicators 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 relationships between inputs and outputs 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 raw gas parameters and the key operation parameters; the number of neurons in the output layer is equal to the number of prediction targets. Here, the prediction target is the selected energy efficiency evaluation indicator. 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 trial and error or optimized in combination with intelligent algorithms.

[0106] Figure 3 Figure 4 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, the corresponding artificial neural network model needs to be configured and trained. 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 outputs 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.

[0107] For the artificial neural network prediction model A, the inputs of the model are the raw gas parameters and the key influencing parameters, and the outputs of the model are all or some of the selected energy efficiency evaluation indicators. The samples used to establish the model come from the historical operation data from January to March 2019. During the production process, a set of data is extracted every hour as the valid historical data. The obtained sample data is randomly divided into three categories: used as the training set, the validation set and the test set respectively, with the proportions being 70%, 15% and 15% respectively.

[0108] Before inputting the sample data into the neural network model for training, the input data is 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", and the transfer function of the output layer is the linear function "purelin". The training function of the network adopts the "trainlm" function, which updates the weights and bias values based on the Levenberg-Marquardt algorithm. A single-hidden-layer network is used to establish the model, and the trial-and-error method is 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 is 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 different each time. Therefore, each neural network with a specific structure is trained 5 times, and the average value of the mean square error is taken. The average mean square error values calculated for different neural network structures are different. The study found 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 relatively ideal, so it is selected for subsequent research.

[0109] To quantify the difference between the predicted results of the energy carrier consumption and the energy and material consumption prediction model A and the true values, the average absolute deviation (AAD%) is 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.

[0110]

[0111] Figure 4 It is a comparison chart between the predicted value of the unit energy consumption of the process and the actual historical value under different working conditions, that is, different raw material gas parameters and key operating parameters. Figure 5 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.

[0112] 3. Based on the relationship prediction model A between the raw material gas parameters, the key operating parameters of the natural gas purification device, and the corresponding energy efficiency evaluation indicators determined in the above step 2, an optimization algorithm model B is established using the iterative optimization method.

[0113] Based on the established relationship prediction model A between the raw gas parameters, key operation parameters, and energy efficiency evaluation indicators, an optimization algorithm model B is established between the raw natural gas parameters and the corresponding key operation parameters of the purification device when the energy efficiency evaluation indicator is optimal. At each raw gas parameter value, by changing the values of the adjustable key operation parameters, the optimal value of the energy efficiency evaluation indicator, that is, the benchmark value with the lowest energy consumption, is obtained through optimization. When the benchmark value is determined, the corresponding guiding values of the adjustable key operation parameters are also determined simultaneously. These guiding values can be used to guide the on-site parameter adjustment of the key operation parameters, achieving the lowest energy consumption in the natural gas purification process.

[0114] The energy efficiency evaluation indicator can be selected from one or more of the above three-level evaluation indicators. After selecting the energy efficiency evaluation indicator, the corresponding key operation parameters to be concerned are also determined simultaneously (determined according to Table 6 above). After obtaining the raw gas parameters, through the optimization algorithm model B, the values of the key operation parameters corresponding to the optimal selected energy efficiency evaluation indicator are calculated respectively through optimization. The value of the key operation parameter corresponding to the optimal energy efficiency evaluation indicator is called the guiding value of the key operation parameter, and the optimal energy efficiency evaluation indicator value is also the energy efficiency evaluation indicator benchmark value when the corresponding energy consumption index is the lowest. Since there are the same key operation parameters corresponding to different energy efficiency evaluation indicators, different guiding values of the key operation parameters may be obtained for different key operation parameters when different energy efficiency evaluation indicators are optimal. When using the guiding value of the key operation parameter to control and adjust the operation of the natural gas purification device, one of them can be selected according to the actual situation or on-site experience, or the average value of different guiding values of the key operation parameters can be obtained, or weights can be set for different energy efficiency evaluation indicators, and the guiding value of the key operation parameter corresponding to the energy efficiency evaluation indicator is selected according to the weight for the control and adjustment of the natural gas purification device.

[0115] Taking the process unit energy consumption as an example of the energy efficiency evaluation indicator, the optimization relationship between the raw gas parameters and the corresponding key operation parameters when the process unit energy consumption is the lowest (the energy efficiency evaluation indicator is optimal) is established. In this embodiment, a genetic algorithm is used to establish the optimization algorithm model B, and iterative adaptive optimization is performed to finally find the process benchmark energy consumption and the guiding key operation parameters (guiding values of the key operation parameters) of the purification unit under different working conditions.

[0116] The calculation process of the genetic algorithm used in this embodiment, the technical roadmap is shown in Figure 6 。The basic elements of the genetic algorithm mainly include: chromosome encoding 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.

[0117] Each individual in the population is an array corresponding to different key operation parameter variables and raw gas parameter variables. The individual fitness value is the energy efficiency evaluation index calculated by using the energy efficiency evaluation index prediction model A established in step 2 (the process unit energy consumption in this embodiment). The prediction model A is the fitness function, and the raw gas parameters do not participate in the variation as a constraint condition. By calculating the individual fitness through the optimization algorithm, comparing the individual fitness values one by one, and then changing the key operation parameters in the individual through crossover and mutation until the minimum fitness value (the minimum energy consumption) is found, that is, the key operation parameters of the purification device when the process unit energy consumption is the lowest under the corresponding raw gas parameters.

[0118] In this embodiment, the initial population size, number of iterations, crossover probability, and mutation probability of the designed genetic algorithm are set to 300, 200, 0.5, and 0.05 respectively. Taking the 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 as Figure 7 shown. It can be seen that after the number of iterations reaches 130 generations, the calculation results tend to converge and remain stable. The output result after the iteration is the lowest value of the process unit energy consumption under 100 kNm 3 / h and 7.80 MPa of the device.

[0119] Figure 8 The black squares in represent the optimal process unit energy consumption obtained by the optimization algorithm model B through optimization calculation under different raw gas treatment volumes (flows), that is, the process unit energy consumption benchmark value. In on-site engineering applications, according to the calculation results of the optimization model, the relationship between the raw gas treatment volume and the process unit energy consumption benchmark value can also be fitted. The fitting curve in this embodiment is shown in Figure 8 the curve. From Figure 8 it can be seen that as the raw gas treatment volume increases, the process unit energy consumption benchmark value also decreases, and the trend is very consistent with the theory and practice of on-site operation.

[0120] 4. Substitute the raw gas parameters of the actual on-site operation into the optimization algorithm model B to calculate the energy efficiency evaluation index benchmark value, and at the same time determine the corresponding key operation parameter guiding value.

[0121] In this embodiment, the process unit energy consumption is the energy efficiency evaluation index. The process unit energy consumption benchmark values under different raw gas parameters on-site are obtained by solving the optimization algorithm model B. Figure 9 The black squares in represent the actual purification process unit energy consumption values, which are all higher than the process unit energy consumption benchmark values (i.e., the benchmark specific consumption represented by the curve) under the same raw gas treatment volume (raw gas flow).

[0122] As the processing volume of the raw gas increases, the gap between the actual value and the reference value of the on-site energy efficiency evaluation index gradually decreases, which is in good agreement with the actual on-site operation. From the difference between the actual value and the reference value, it can also be analyzed that the process energy-saving potential is greater under low raw gas processing volume. Figure 10 is the energy consumption factor under different raw gas processing volumes. According to this energy efficiency index, the operation level of the actual operating conditions can be intuitively judged. Under actual conditions, the difference between the actual operating value of the key operating parameter and the guiding value of the key operating parameter calculated by the optimization algorithm model B is the on-site optimization space, which provides direct support for optimized operation, that is, controlling each purification device according to the guiding value, and adjusting each key operating parameter towards the guiding value to effectively reduce the comprehensive energy consumption of the natural gas purification process.

[0123] The natural gas purification control method of the present invention not only considers the influence of the differences in raw gas parameters on the actual operating energy consumption level, but also considers the influence of key operating parameters on the actual operating energy consumption level under the same raw gas parameters. At the same time, multiple energy efficiency evaluation indexes at three levels are proposed, and the potential of natural gas energy consumption optimization is evaluated through the operation of on-site equipment and the obtained reference value with the lowest energy consumption. The method of the present invention not only realizes the scientific and intuitive evaluation of the operation level, energy consumption change law and energy-saving space under different working conditions, but also provides easy-to-implement operation parameter tuning guidance for the optimized operation of the purification process, realizing the energy-saving operation of natural gas purification.

[0124] Platform embodiment:

[0125] This embodiment provides an on-line optimization platform for natural gas purification based on energy efficiency evaluation, as Figure 11 shown, including a memory, a processor and an internal bus. The processor and the memory complete mutual communication through the internal bus.

[0126] The processor can be a microprocessor MCU, a programmable logic device FPGA and other processing devices.

[0127] The memory can be various memories that store information by means of electric energy, such as RAM, ROM, etc.; various memories that store information by means of magnetic energy, such as hard disks, floppy disks, magnetic tapes, magnetic core memories, bubble memories, USB flash drives, etc.; various memories that store information by means of optical methods, such as CDs, DVDs, etc. Of course, there are also other types of memories, such as quantum memories, graphene memories, etc.

[0128] The processor can call the logical instructions in the memory to implement a natural gas purification control method based on energy efficiency evaluation. This method is introduced in detail in the method embodiment and will not be repeated here.

Claims

1. A natural gas purification control method based on energy efficiency evaluation, characterized in that, It includes the following steps: 1) Collect the current raw natural gas parameters, and obtain the key operating parameters of the natural gas purification device corresponding to the optimal energy efficiency conditions under the current raw natural gas parameters through optimization by intelligent algorithm B; The energy efficiency conditions include one or more of the ratio of the total process energy consumption to the corresponding raw natural gas throughput, the ratio of the total process energy consumption to the economic value of the corresponding product, the sum of the unit energy consumptions of each process unit of the purification device, and the sum of the unit energy consumptions of each device of the purification device; the unit energy consumption of the process unit is the ratio of the energy consumption of the process unit to the corresponding input raw material flow or output product flow, and the unit energy consumption of the device is the ratio of the energy consumption of the device to the flow of the corresponding processed substance; Each process unit of the purification device includes a desulfurization and decarbonization unit, a dehydration unit, a sulfur recovery unit, a tail gas treatment unit, and an acid water stripping unit; 2) Control the corresponding natural gas purification device according to the key operating parameters of the natural gas purification device corresponding to the optimal energy efficiency conditions under the current raw natural gas parameters; The optimization process of intelligent algorithm B in step 1) includes: first, initializing the population, and each individual in the population includes the key operating parameters of the natural gas purification device and the current raw natural gas parameters; Then calculate the population fitness of each individual under the current raw natural gas parameters, and the population fitness is the energy efficiency condition, which is calculated by prediction model A; finally, change the key operating parameters of the natural gas purification device in the individual by crossover and mutation to change the population and iterate multiple times until the key operating parameters of the natural gas purification device corresponding to the optimal population fitness are found as the optimization result; The prediction model A is a machine learning model with the raw natural gas parameters and the key operating parameters of the natural gas purification device as inputs and the corresponding predicted energy efficiency conditions as outputs, and is trained by the historical data of the raw natural gas parameters, the key operating parameters of the natural gas purification device, and the corresponding energy efficiency conditions.

2. The natural gas purification control method based on energy efficiency evaluation according to claim 1, characterized in that, The total process energy consumption is the sum of the energy consumption and material consumption in the natural gas purification process; the energy consumption of the process unit is the sum of the energy consumption and material consumption of the process unit in the natural gas purification process; the energy consumption of the device is the sum of the energy consumption and material consumption of the device in the natural gas purification process.

3. The natural gas purification control method based on energy efficiency evaluation according to claim 2, characterized in that, The sum of the energy consumption and material consumption is obtained by weighted summation of various consumed energies and substances according to their respective energy conversion coefficients.

4. The natural gas purification control method based on energy efficiency evaluation according to claim 1, characterized in that, The prediction model A is a neural network model.

5. The natural gas purification control method based on energy efficiency evaluation according to claim 4, characterized in that, The intelligent algorithm B is a genetic algorithm.

6. The natural gas purification control method based on energy efficiency evaluation according to claim 1, characterized in that, The raw natural gas parameters include the raw natural gas flow rate, pressure, hydrogen sulfide content, and carbon dioxide content.

7. A natural gas purification online optimization platform based on energy efficiency evaluation, characterized in that, It includes a control system, and the control system is controlled and connected to the natural gas purification device to realize the adjustment of the key operating parameters of each natural gas purification device; the control system also executes instructions to realize the natural gas purification control method based on energy efficiency evaluation as described in any one of claims 1 to 6.

Images