Method and device for determining helium extraction process parameter threshold value of co-production liquefied natural gas (LNG)

Abstract

The invention discloses a method and a device for determining helium extraction process parameter thresholds for co-production of liquefied natural gas (LNG). The method comprises the following steps: acquiring basic data of a target helium extraction project and historical data of a similar helium extraction project; the initial gas inlet scale and the gas yield of the target helium extraction project are determined; determining the gas inlet flow of each device in the target helium extraction project and the investment of each device; inputting the cost data into a plurality of pre-trained cost analysis models to obtain a cost analysis result; according to the cost analysis result and the investment of each device in the target helium extraction project, an SLSQP algorithm is adopted, a minimum cost target function of the gas inflow and the helium concentration is established, the optimized gas inflow scale and the optimized ammonia concentration are obtained, the comprehensiveness of determining consideration factors of the parameter threshold value of the helium extraction process of the co-production LNG can be improved, and the reliability of the system is improved. The accuracy and efficiency of determining the parameter threshold value of the helium extraction process of the co-production LNG are improved, and the energy consumption and cost of the process are reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of natural gas processing and helium resource extraction, and particularly relates to a method and device for determining process parameter threshold values for co-production of liquefied natural gas (LNG) and helium. BACKGROUND

[0002] This section is intended to provide background or context to the embodiments of the application. The description herein does not constitute admission that the prior art is prior art nor does it constitute an admission of any description in this section as prior art to an application described herein and likewise, reference to a document herein does not constitute an admission that the document is prior art to an application described herein.

[0003] Generally, an oil and gas project can be judged to have exploitation value by calculating a minimum economic scale. The minimum economic scale refers to a scale of an oil and gas field (i.e., reserves when the net present value is equal to zero) with a minimum return rate obtained by deducting an oil and gas production cost from an oil and gas production condition, a geographical environment condition and a traffic condition of an oil and gas field, and the like, under current investment and economic environment. If an expected production scale of an oil and gas field is greater than the minimum economic scale, it is indicated that the oil and gas field has exploitation value. If the expected production scale is less than the minimum economic scale, it is indicated that the oil and gas field cannot reach a benchmark return rate, and a next investment plan should be carefully decided.

[0004] For a co-production LNG (Liquefied Natural Gas) and helium engineering project, the minimum economic inlet gas scale of the helium extraction engineering refers to an initial inlet gas scale when a profit and loss balance (i.e., a net present value is equal to zero) is met under an existing process. For a project of co-production of LNG and helium using a membrane separation + decarburization and dehydration + low-temperature distillation + nitrogen circulation refrigeration process + pressure swing adsorption process, the initial inlet gas scale is a gas flow rate estimated (maximum) according to a production capacity, and a processing scale of a next year is adjusted according to a natural gas field production decline and is included in a benefit calculation.

[0005] A conventional method for determining process parameter threshold values for co-production of LNG and helium does not systematically associate the initial inlet gas scale, a gas component and a project economic index, resulting in that a production decision depends on experience estimation, accuracy and efficiency are low, a cost and energy consumption of the co-production LNG and helium engineering are high, and a project production risk is high. SUMMARY

[0006] Embodiments of the present application provide a method for determining process parameter threshold values for co-production of LNG and helium, to improve comprehensiveness of factors considered in determining the process parameter threshold values for co-production of LNG and helium, improve accuracy and efficiency of determining the process parameter threshold values for co-production of LNG and helium, and reduce process energy consumption and cost. The method comprises:

[0007] The basic data of the target helium extraction project and the historical data of similar helium extraction projects are acquired; the similar helium extraction projects are the same as the target helium extraction project in a helium extraction process; the basic data comprises a gas field scale and a gas component of the target helium extraction project; and the historical data comprises a gas field scale, a gas component, an initial gas inlet scale, a gas yield, an inlet flow of each device, an investment of each device and a plurality of operation costs of the similar helium extraction projects.

[0008] An initial gas inlet scale and a gas yield of the target helium extraction project are determined according to the basic data and the historical data by using an analogy algorithm.

[0009] An inlet flow of each device and an investment of each device in the target helium extraction project are determined according to the basic data and the historical data by using a LASSO (Least Absolute Shrinkage and Selection Operator) regression analysis algorithm.

[0010] The initial gas inlet scale, the gas yield, the inlet flow of each device, the investment of each device and the basic data of the target helium extraction project are respectively input into a plurality of pre-trained cost analysis models to obtain cost analysis results; the plurality of pre-trained cost analysis models are obtained by training a plurality of LightGBM (Light Gradient Boosting Machine) models respectively using the historical data; and each LightGBM model uses different operation cost items for training.

[0011] According to the cost analysis results and the investment of each device in the target helium extraction project, a SLSQP (Sequential Least Squares Programming) algorithm is used to establish a minimum cost objective function of the gas inlet scale and the helium concentration, and the optimized gas inlet scale and the ammonia concentration are obtained after multiple iterations.

[0012] The embodiment of the present application also provides a co-production LNG helium extraction process parameter threshold determination device to improve the comprehensiveness of the co-production LNG helium extraction process parameter threshold determination consideration factor, improve the accuracy and efficiency of the co-production LNG helium extraction process parameter threshold determination, and reduce the process energy consumption and cost.

[0013] An acquisition module is configured to acquire basic data of a target helium extraction project and historical data of similar helium extraction projects; the similar helium extraction projects are the same as the target helium extraction project in a helium extraction process; the basic data comprises a gas field scale and a gas component of the target helium extraction project; and the historical data comprises a gas field scale, a gas component, an initial gas inlet scale, a gas yield, an inlet flow of each device, an investment of each device and a plurality of operation costs of the similar helium extraction projects.

[0014] an analog module configured to determine an initial air intake scale and a gas yield of a target helium extraction project according to basic data and historical data by using an analog algorithm;

[0015] a regression analysis module configured to determine an air intake flow of each device and an investment of each device in the target helium extraction project according to the basic data and the historical data by using a LASSO regression analysis algorithm;

[0016] a cost analysis module configured to input the initial air intake scale, the gas yield, the air intake flow of each device, the investment of each device and the basic data of the target helium extraction project into a plurality of pre-trained cost analysis models respectively to obtain a cost analysis result, wherein the plurality of pre-trained cost analysis models are obtained by training a plurality of LightGBM models respectively using the historical data, and each LightGBM model uses different operating cost items for training;

[0017] an optimization module configured to establish a minimum cost objective function of air intake and helium concentration according to the cost analysis result and the investment of each device in the target helium extraction project by using a SLSQP algorithm, and to obtain an optimized air intake scale and ammonia concentration after multiple iterations.

[0018] Compared with the prior art, the embodiment of the present application determines the process parameter threshold of the cogeneration LNG helium extraction process by obtaining the basic data of the target helium extraction project and the historical data of similar helium extraction projects; the helium extraction processes used by the similar helium extraction projects and the target helium extraction project are the same; the basic data includes the gas field scale and gas composition of the target helium extraction project; the historical data includes the gas field scale, gas composition, initial gas scale, gas yield, gas flow of each device, investment of each device, and multiple operation costs of the similar helium extraction projects; the initial gas scale and gas yield of the target helium extraction project are determined according to the basic data and the historical data by using an analogy algorithm; the gas flow of each device and the investment of each device of the target helium extraction project are determined according to the basic data and the historical data by using a LASSO regression analysis algorithm; the initial gas scale, gas yield, gas flow of each device, investment of each device, and basic data of the target helium extraction project are respectively input into a plurality of pre-trained cost analysis models to obtain cost analysis results; the plurality of pre-trained cost analysis models are obtained by training a plurality of LightGBM models using historical data; each LightGBM model uses different operation cost items for training; according to the cost analysis results and the investment of each device in the target helium extraction project, a SLSQP algorithm is used to establish a minimum cost objective function of the gas flow and the helium concentration, and the optimized gas scale and ammonia concentration are obtained after multiple iterations, which can improve the comprehensiveness of the consideration factors for determining the process parameter threshold of the cogeneration LNG helium extraction process, improve the accuracy and efficiency of the determination of the process parameter threshold of the cogeneration LNG helium extraction process, and reduce the process energy consumption and cost. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creating any inventive labor. In the drawings:

[0020] Figure 1 A flow chart of a cogeneration LNG helium extraction process parameter threshold determination method provided in an embodiment of the present application;

[0021] Figure 2 A flow chart of a specific example of a cogeneration LNG helium extraction process parameter threshold determination method provided in an embodiment of the present application;

[0022] Figure 3 A raw material gas and helium yield relationship diagram of a target helium extraction project in an embodiment of the present application;

[0023] Figure 4 A gas flow diagram of a target helium extraction project in an embodiment of the present application;

[0024] Figure 5 A threshold distribution diagram of the intake scale of the target helium extraction project in the embodiment of the present application under different helium concentrations;

[0025] Figure 6 A minimum economic helium concentration distribution diagram of the target helium extraction project in the embodiment of the present application under different intake scales;

[0026] Figure 7 A SLSQP joint optimization feasible region analysis diagram of the target helium extraction project in the embodiment of the present application;

[0027] Figure 8 A target function surface and optimal solution distribution diagram of the target helium extraction project in the embodiment of the present application;

[0028] Figure 9 A schematic diagram of a cogeneration LNG helium extraction process parameter threshold determination device provided in the embodiment of the present application;

[0029] Figure 10 A schematic diagram of a computer device provided in the embodiment of the present application. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, further detailed description of the embodiments of the present application will be given below with reference to the drawings. Herein, the schematic embodiments of the present application and the descriptions thereof are used to explain the present application, but not as a limitation on the present application.

[0031] The acquisition, storage, use, processing and the like of data in the technical scheme of the present application all comply with the relevant provisions of laws and regulations.

[0032] The term "and / or" herein merely describes an association relationship, which means that there can be three relationships, for example, A and / or B can mean that A exists alone, A and B exist together, and B exists alone. In addition, the term "at least one" herein means any one of a plurality or any combination of at least two of a plurality, for example, including at least one of A, B and C can mean including any one or more elements selected from the set consisting of A, B and C.

[0033] In the description of the present specification, "include", "including", "have", "has", and the like are open terms, that is, mean including but not limited to. The description referring to the terms "one embodiment", "one specific embodiment", "some embodiments", "for example", and the like means that the specific features, structures, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In the present specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. The order of steps involved in each embodiment is used to illustrate the implementation of the present application, and the order of steps is not limited. The order of steps can be adjusted as needed.

[0034] It should be noted that all the determining factors in the embodiments of the present application are objective factors, and have nothing to do with artificial factors and market factors. For example, the cost data, investment data, and net present value data used in the embodiments of the present application are objective data derived from historical data and are not affected by market factors. The tax and discount rate are fixed values and are objective data, and are not affected by market factors.

[0035] The analytical function method is generally applicable to mature blocks with some discoveries. By setting the project exploration and development cost, a nonlinear analytical function of resource quantity and net present value (NPV = f (Resources)) is established. When the function value is zero, the resource quantity is the minimum economic scale.

[0036] For the co-production LNG helium extraction engineering project using membrane separation and low-temperature distillation technology, the analytical function method is used to establish a nonlinear analytical function of initial gas intake scale and net present value (NPV = f (initial gas intake scale)). The initial gas intake scale is used as the control variable for planning solution. The discounted cash flow economic evaluation model of the project is used to calculate the initial gas intake scale that meets the break-even point (i.e., the net present value is equal to zero). The initial gas intake scale is the minimum economic gas intake scale of the helium extraction engineering.

[0037] The traditional economic evaluation method of helium purification engineering does not systematically correlate the initial gas intake scale, gas components, and project economic indicators, resulting in investment decisions relying on empirical estimates, large cost and benefit prediction deviations, and high project risks.

[0038] To solve the above problems, the embodiment of the present application proposes a calculation method for determining the threshold of helium extraction concentration and gas inlet scale in cogeneration LNG based on variable joint optimization solution, by establishing the dynamic functional relationship between gas production (raw gas, helium), device investment, operating cost, gas inlet scale and gas component, using Newton iteration method to solve the single variable minimum helium extraction concentration and single variable minimum gas inlet scale, establishing SLSQP optimization model, calculating the double variable optimal value of gas inlet scale and helium concentration, and providing basis for helium extraction investment decision. The economic quantification evaluation of helium purification project can be realized, the investment decision risk is reduced, the calculation error of minimum economic scale is reduced by more than 20% compared with the traditional experience method, and scientific basis is provided for the value-added of natural gas industry chain and the efficient development of key mineral resources (helium).

[0039] Figure 1 The flow chart of the method for determining the process parameter threshold of helium extraction in cogeneration LNG provided in the embodiment of the present application is shown in Figure 1 The method can include the following steps:

[0040] In step 101, the basic data of the target helium extraction project and the historical data of similar helium extraction projects are obtained; the helium extraction process of the similar helium extraction project is the same as that of the target helium extraction project; the basic data includes the gas field scale and gas component of the target helium extraction project; the historical data includes the gas field scale, gas component, initial gas inlet scale, gas production, gas inlet flow rate of each device, investment of each device and multiple operating costs of the similar helium extraction project;

[0041] In step 102, the initial gas inlet scale and gas production of the target helium extraction project are determined according to the basic data and historical data by using an analogy algorithm;

[0042] In step 103, the gas inlet flow rate of each device and the investment of each device in the target helium extraction project are determined according to the basic data and historical data by using a LASSO regression analysis algorithm;

[0043] In step 104, the initial gas inlet scale, gas production, gas inlet flow rate of each device, investment of each device and basic data of the target helium extraction project are input into a plurality of pre-trained cost analysis models respectively to obtain cost analysis results; the plurality of pre-trained cost analysis models are obtained by training a plurality of LightGBM models respectively using historical data; each LightGBM model uses different operating cost items for training;

[0044] In step 105, according to the cost analysis results and the investment of each device in the target helium extraction project, a SLSQP algorithm is used to establish a minimum cost objective function of gas inlet scale and helium concentration, and the optimized gas inlet scale and helium concentration are obtained after multiple iterations.

[0045] The embodiment of the present application can improve the comprehensiveness of the consideration factors for determining the cogeneration LNG helium extraction process parameter threshold value, improve the accuracy and efficiency of the determination of the cogeneration LNG helium extraction process parameter threshold value, and reduce the process energy consumption and cost.

[0046] Figure 2 A flowchart of a specific example of a cogeneration LNG helium extraction process parameter threshold value determination method provided in the embodiment of the present application is shown in FIG. 1, in one embodiment, the cogeneration LNG helium extraction process parameter threshold value determination method can include: Figure 2

[0047] S1: using an analogy method, referring to gas fields adopting the same helium extraction process and their scales, constructing a relationship function of gas production (raw gas and helium) and initial gas inlet scale, gas component, obtaining each gas production in each year of the service life period from the annual production decline law.

[0048] S2: determining each link device required by the membrane separation and low-temperature distillation technology process, constructing a relationship function of each device investment and initial gas inlet scale, gas component by LASSO regression analysis, determining annual investment from the annual investment proportion.

[0049] S3: using a gradient boosting decision tree LightGBM to construct a relationship function of itemized operation cost and initial gas inlet scale, gas component, determining each annual operation cost from the production change.

[0050] S4: calculating the amount of money outflow from the annual investment and operation cost, determining the project net present value. In the case that the helium extraction project net present value is zero, fixing the raw gas helium concentration, using the Newton iteration method to solve the minimum gas inlet scale.

[0051] S5: calculating the amount of money outflow from the annual investment and operation cost, determining the project net present value. In the case that the ammonia extraction project net present value is zero, fixing the raw gas inlet scale, using the Newton iteration method to solve the minimum helium extraction concentration.

[0052] S6: using a sequential least squares programming SLSQP optimization algorithm to establish a minimum comprehensive cost objective function of the gas inlet amount and the helium concentration, and calculating the optimal gas inlet scale and the optimal helium concentration after multiple iterations.

[0053] In one embodiment, in steps 101, 102 and S1, for a helium extraction engineering project adopting membrane separation and low-temperature distillation technology, based on the analogy method, the historical data of a similar helium-containing gas field of a similar process are selected to construct a dynamic relationship function of the initial gas inlet scale, the gas component and the raw gas / helium production; combined with the annual production decline law (exponential decline, hyperbolic decline or harmonic decline model), the raw gas and helium production in each year of the service life period is predicted.

[0054] ​In one embodiment, the initial gas intake scale and gas yield of the target helium extraction project are determined according to the basic data and historical data by using the analogy algorithm, which can include: the initial gas intake scale of the target helium extraction project is determined according to the gas field scale of the target helium extraction project, the gas field scale and the initial gas intake scale of the similar helium extraction project by using the analogy algorithm.

[0055] According to the scale of the target helium extraction project, the initial gas intake scale of the helium extraction project is determined according to the scale of the gas field using the same helium extraction process, that is,

[0056] Then, the yield of the raw gas and the helium gas is determined according to the gas field scale and the gas component (helium content) of the target helium extraction project, and the yield of the raw gas and the helium gas in each year of the life cycle is obtained by the annual yield decline rule. Figure 3 The relationship between the raw gas and the helium gas yield of the target helium extraction project obtained in this step is shown in the relationship diagram of the raw gas and the helium gas yield of the target helium extraction project in the embodiment of the present application. Figure 3 Thus, the relationship function between the gas yield and the initial gas intake scale and the gas component (gas yield = f (initial gas intake scale, gas component)) is established.

[0057] In one embodiment, the initial gas intake scale and gas yield of the target helium extraction project are determined according to the basic data and historical data by using the analogy algorithm, which includes: the relationship function between the gas component, the initial gas intake scale and the gas yield of the similar helium extraction project is determined according to the historical data; the initial gas yield of the target helium extraction project is determined according to the relationship function, the gas component and the initial gas intake scale of the target helium extraction project by using the analogy algorithm; the initial gas yield is the gas yield of the target helium extraction project in the first year; and the gas yield in each year of the life cycle of the target helium extraction project is determined according to the initial gas yield and the preset attenuation coefficient.

[0058] An already extracted helium project (similar helium extraction project) similar to the target gas field helium concentration, raw gas component and geological conditions in the target helium extraction project is selected as a reference.

[0059] The yield decline function is constructed by historical data regression:

[0060] P He,t = P He,0 × e -αt ;

[0061] In the formula, P He,t is the gas yield of the target helium extraction project at time t (which can be in years); P He,0 is the initial gas yield of the similar helium extraction project; and a is the attenuation coefficient, which can be 0.03-0.05 / year.

[0062] The initial gas yield PHe,0 The initial gas intake scale S0 and the gas component (which can be the helium concentration C He ) present a linear relationship:

[0063] P He,0 = k x S0 x C He ;

[0064] Wherein, k is the recovery coefficient, which can be 0.85-0.95.

[0065] Taking a certain to-be-built co-production LNG helium extraction project as an example, the project has the same helium extraction process as a certain operating helium extraction project, and the initial gas intake scale of the to-be-built helium extraction project is estimated by analogy with the raw gas yield scale of the known helium extraction project. The raw gas yield scale of the known helium extraction project is 120x10 4 Nm 3 / day, the initial gas intake scale is 120x10 4 Nm 3 / day, the raw gas yield scale of a certain gas field project is 900x10 4 Nm 3 / day, and the initial gas intake scale of the to-be-built helium extraction project is 900x10 4 Nm 3 / day by analogy. The average helium content of the to-be-built helium extraction project is 0.1076% (the industrial utilization lower limit is 0.05%), the N2 concentration is 0.881%, the raw gas yield scale is 900x10 4 Nm 3 / day, and the gas yield of each year during the service life is obtained according to the annual yield decline rule, as shown in Figure 3 . After analogy with the known helium extraction project, the first-year helium yield is predicted to be 477,100 m3, and the annual decline rate is 4.2%.

[0066] In one embodiment, after the initial gas yield of the target helium extraction project is determined according to the relationship function, the gas component of the target helium extraction project, and the initial gas intake scale by using the analogy algorithm, it further includes: if the nitrogen concentration in the gas component of the target helium extraction project is greater than a first preset threshold, and / or the carbon dioxide concentration is greater than a second preset threshold, the determined initial gas yield of the target helium extraction project is corrected by using a correction factor; the correction factor is determined by the nitrogen concentration and / or the carbon dioxide concentration in the gas component of the target helium extraction project.

[0067] For a gas field with high nitrogen N2 (> a first preset threshold, which can be 5%) or high carbon dioxide CO2 (> a second preset threshold, which can be 2%), a correction factor β needs to be introduced to reduce the effective yield:

[0068] P He,0 ′= P He,0 x β;

[0069]

[0070] Among them, P He,0 ' is the revised initial gas production of the target helium extraction project; N2% is the nitrogen concentration in the gas components of the target helium extraction project; CO2% is the carbon dioxide concentration in the gas components of the target helium extraction project.

[0071] In steps 103 and S2, the device investment is modeled: for helium extraction projects using membrane separation and cryogenic distillation technology, the devices required for each link of the process are determined, and the gas composition of the project is taken into consideration. The intake flow rate of each link of the helium extraction device is estimated. Through LASSO regression analysis, a relationship function between the investment of each device and the initial intake scale and gas composition is constructed, and the annual investment is determined based on the investment ratio of each year.

[0072] Currently, co-production LNG helium extraction projects primarily utilize a combination of membrane separation and cryogenic distillation processes. This process primarily utilizes membrane separation, post-expansion, nitrogen cycle refrigeration, and two-tower cryogenic distillation. Investments associated with helium extraction projects include those related to cryogenic distillation and storage and transportation. The inlet flow rate of the membrane separation unit in a target helium extraction project can be calculated using the following formula:

[0073] F membrane =S0×(1-η CH4 );

[0074]

[0075] In the formula: F membrane is the air flow rate of the membrane separation device, that is, the residual flow after the membrane, the total flow rate minus the permeation rate; S0 is the total air flow (unit: mol / s); η CH4 is the separation efficiency, that is, the ratio of permeation to the total intake flow; P CH4 is the permeability coefficient of methane (unit: mol / m 2 ·s·Pa), which represents the permeation rate of methane per unit area and unit pressure difference; A: effective area of ​​the membrane (unit: m 2 ); Δp: methane partial pressure difference on both sides of the membrane (unit: Pa).

[0076] The inlet flow rate of the cryogenic distillation tower can be designed according to the helium-rich flow meter and must meet the minimum economic scale (such as ≥50m 3 The intake flow rate of other devices can be determined using the LASSO regression analysis algorithm based on basic data and historical data.

[0077] In one embodiment, the LASSO regression analysis algorithm is used to determine the gas inlet flow rate of each device in the target helium extraction project and the investment of each device according to the basic data and historical data, including: using the LASSO regression analysis algorithm, establishing the following impurity sensitivity coefficient regression equation according to the gas components of the similar helium extraction project:

[0078] γ = γ0 + γ1C He + γ2C N2 + γ3C CO2 ;

[0079] Wherein, γ is the impurity sensitivity coefficient; γ0 is the methane sensitivity coefficient; γ1 is the helium sensitivity coefficient; γ2 is the nitrogen sensitivity coefficient; γ3 is the carbon dioxide sensitivity coefficient; C He is the helium concentration in the gas components of the similar helium extraction project or the target helium extraction project; C N2 is the nitrogen concentration in the gas components of the similar helium extraction project or the target helium extraction project; C CO2 is the carbon dioxide concentration in the gas components of the similar helium extraction project or the target helium extraction project; the gas inlet flow rate of each device in the target helium extraction project and the investment of each device are determined according to the basic data, historical data and impurity sensitivity coefficient regression equation.

[0080] In one embodiment, the gas inlet flow rate of each device in the target helium extraction project and the investment of each device are determined according to the basic data, historical data and impurity sensitivity coefficient regression equation, including: the gas inlet flow rate of each device in the target helium extraction project is determined according to the nitrogen concentration in the gas components of the target helium extraction project, the nitrogen concentration in the gas components of the similar helium extraction project, the initial gas inlet scale of the similar helium extraction project, the initial gas inlet scale of the target helium extraction project and the gas inlet flow rate of each device in the similar helium extraction project; the investment of each device in the target helium extraction project is determined according to the basic data, historical data and impurity sensitivity coefficient regression equation according to the following formula:

[0081]

[0082] Wherein, I j is the investment of the jth device in the target helium extraction project; I j.base is the investment of the jth device in the similar helium extraction project; γ is the impurity sensitivity coefficient; ΔC impurity is the percentage of impurity concentration exceeding the preset reference value; F j is the flow rate of the jth device of the target helium extraction project; F j .base is the flow rate of the jth device of the similar helium extraction project.

[0083] After obtaining the investment result, the total cost of each device can be proportionally allocated to the annual expenditure table according to the engineering progress (such as 30%: 50%: 20%).

[0084] In addition to the above method, the relationship function of each device investment and initial air intake scale, gas component can also be constructed, and the annual device investment is determined by the annual investment ratio. Among them, the low temperature distillation related investment includes the investment of booster device, demethanizer device, helium refining device and helium liquefaction device. Each device investment is related to gas component and initial air intake scale. According to the related gas component content of each device, the processing scale of each device in the low temperature distillation step of the project is estimated according to the processing scale of the device of the project adopting the same helium extraction process, that is Then the scale index method is used to estimate the low temperature distillation related investment, that is (Wherein, the scale index is valued according to the similar helium extraction project scale, and the adjustment coefficient is valued according to the cost situation of the target helium extraction project period and place). Thus, the relationship function of low temperature distillation related investment and initial air intake scale, gas component is established (low temperature distillation related investment = f (initial air intake scale, gas component)). Other device investment can be calculated by referring to this method.

[0085] Storage and transportation related investment includes helium storage and loading facilities and ground supporting facilities investment. Helium storage and loading facilities investment is mainly related to initial air intake scale, and the helium storage and loading facilities investment of the project is estimated by referring to the air intake scale and investment of the project adopting the same helium extraction process, that is

[0086] (Wherein, the scale index is valued according to the similar helium extraction project scale, and the adjustment coefficient is valued according to the cost situation of the target helium extraction project period and place). The ground supporting facilities investment is estimated according to the total investment scale, which is generally 55%-65% of the total investment. Thus, the relationship function of storage and transportation related investment and initial air intake scale, gas component is established (storage and transportation related investment = f (initial air intake scale, gas component)).

[0087] The target helium extraction project adopts similar helium extraction project process, and the device air intake of the target helium extraction project is estimated by analogy with the device air intake of similar helium extraction project and considering the difference of gas component, as shown in table 1. According to the air intake of each link and the corresponding scale index, the device investment is estimated, as shown in table 2.

[0088] Table 1

[0089]

[0090] Table 2

[0091]

[0092] The design air intake scale of the target helium extraction project is 9000000m3 / h, helium content 0.1076%, similar helium extraction project intake scale 1200000m 3 / h, helium content 0.1278%, helium refining device flow rate 3100m 3 / h, adjustment coefficient 1.55, target helium extraction project helium refining device flow rate:

[0093] cubic meters / hour;

[0094] Similar helium extraction project investment 918 million yuan, adjustment coefficient 0.6, target helium extraction project helium refining device investment:

[0095] million yuan.

[0096] In steps 104 and S3, the operating cost is modeled: identify the operating cost drivers of each device, including power consumption, membrane replacement frequency, refrigerant consumption, combined with the influence of intake scale and gas composition on energy consumption, according to the yield decline law of S1, the relationship function between the sub-item operating cost and the initial intake scale, gas composition is constructed by gradient boosting decision tree LightGBM, and the annual operating cost is determined by the change of yield.

[0097] In one embodiment, the plurality of operating costs includes one or any combination of operating power cost, maintenance cost, material cost, fuel cost, personnel cost, repair cost, management cost and business cost.

[0098] In one embodiment, before the initial intake scale, gas yield, intake flow rate of each device, investment of each device and basic data of the target helium extraction project are input into a plurality of pre-trained cost analysis models to obtain cost analysis results, it further includes: establishing a LightGBM model corresponding to each operating cost in the plurality of operating costs; using the gas field scale, gas composition, initial intake scale, gas yield, intake flow rate of each device, investment of each device and each operating cost of the similar helium extraction project in the historical data to train the LightGBM model corresponding to the operating cost, to generate a plurality of cost analysis models.

[0099] Model construction: use multi-output regressor to wrap gradient boosting decision tree LightGBM, train independent model for each cost target.

[0100] Input layer: 10 feature parameters, including the operating power cost, maintenance cost, material cost, fuel cost, personnel cost, repair cost, management cost, and business cost of similar helium extraction projects, and the feed gas composition and helium extraction device flow of the target helium extraction project. The mean and standard deviation of the training set are used for Z-score standardization, and are distributed to 8 gradient boosting decision tree LightGBM models.

[0101]

[0102] wherein x norm is the standardized training set; x is the training set; σ is the standard deviation; μ is the mean.

[0103] Hidden layer: 8 independent gradient boosting decision tree LightGBM models (each corresponding to one cost), each model independently performs decision tree splitting and prediction.

[0104] Output layer: 8-dimensional vector, including the operating power cost, maintenance cost, material cost, fuel cost, personnel cost, repair cost, management cost, and business cost of the to-be-built helium extraction project, and the 8 cost prediction values are combined into the final result.

[0105] Data segmentation: 70% of the data in the training set is used for model training, 15% of the data in the validation set is used for early stopping method to monitor overfitting, and 15% of the data in the test set is used to evaluate the performance of the final model.

[0106] Model evaluation and optimization: the mean absolute error directly reflects the deviation of cost prediction, and R 2 score measures the explanatory power of the model to the cost change (target R 2 > 0.8).

[0107] Hyperparameter tuning uses Bayesian optimization to search for the optimal parameter combination.

[0108] SHAP (Shapley Additive Explanations) value explains the contribution of each feature to the cost, and the key features such as the impact of flow on power cost are analyzed through SHAP value.

[0109] The advantage of this method is that each cost target can adapt to different tree structures, such as power cost which may depend more on flow, and personnel cost which may depend more on equipment quantity. Through the above process, LightGBM can be used efficiently to predict the sub-item operating cost of helium extraction project, and provide quantitative support for investment decision. The key parameters of gradient boosting decision tree LightGBM are shown in Table 3.

[0110] Table 3

[0111] Parameter Value / Type Role Target value Regression Regression task Mean absolute error MAE Evaluation metric (mean absolute error) Leaf number 31 Max leaf number of single tree, control model complexity Learning rate 0.05 Influence convergence speed and accuracy Feature fraction 0.9 Randomly use 90% features per iteration to prevent overfitting

[0112] In addition to the above steps, the following method can also be used to calculate the operating cost. According to the process used in the co-production LNG helium extraction engineering project, the operating cost drivers of each link device in the helium extraction are determined, and the annual consumption of each cost is estimated considering the gas scale of each device. Then, based on the unit price of each cost and the annual consumption, the total price of the helium extraction project operating cost is estimated, and the annual operation cost is determined by the yield change.

[0113] For the membrane separation and low-temperature distillation helium extraction process in the above step, the operating cost of the helium extraction project is composed of operating power cost, maintenance cost, material cost, fuel cost, personnel cost, repair cost, management cost and business cost. The calculation formula of each operating cost and the cost driver are shown in Table 4:

[0114] Table 4

[0115]

[0116]

[0117] According to the processing scale of each device of the target helium extraction project and the cost driver of each operating cost in Table 4, the annual consumption of each cost is estimated, and the total price of the operating cost of the project is estimated by referring to the device processing scale and operating cost unit price of the project using the same helium extraction process, that is: Thus, the relationship function between the initial gas scale and the gas component of the sub-item operating cost (sub-item operating cost = f(initial gas scale, gas component) is established, and then the annual operating cost of each item is determined according to the yield change.

[0118] The device processing scale and operating cost unit price of similar helium extraction projects are counted, and the annual total price of each operating cost of the target helium extraction project is shown in Table 5.

[0119] Table 5

[0120]

[0121] The helium extraction flow rate F of the target helium extraction project membrane = 30341 m 3 / h, C He = 0.1076%, C N2 = 1%, C CO2 = 0.2%, and the sub-item operating cost of the similar helium extraction project (as shown in Table 6). The data is preprocessed, assuming that the mean and standard deviation of the training set are:

[0122] The model mean = [30341, 0.1076, 23061, 6256, 4242, 2767, 12464, 983, 492, 346];

[0123] Standard deviation = [150, 0.05, 108, 65, 49, 28, 138, 89, 21, 29];

[0124] Normal range = [30341 ± 300, 0.1076 ± 0.05, 23061 ± 97, 6256 ± 62, 4242 ± 48, 2767 ± 25, 12464 ± 118, 983 ±, 492 ±, 346 ± 4].

[0125] The total operation cost of the helium extraction project is 320367 million yuan, as shown in Table 6.

[0126] Table 6

[0127]

[0128] In one embodiment, according to the cost analysis result and the investment of each device in the target helium extraction project, a sequence least squares programming (SLSQP) algorithm is used to establish a minimum cost objective function of the gas inlet amount and the helium concentration, and the optimized gas inlet scale and ammonia concentration are obtained after multiple iterations, including: determining the net present value of the target helium extraction project according to the cost analysis result and the investment of each device in the target helium extraction project; using the SLSQP algorithm, the net present value of the target helium extraction project is used to establish a minimum cost objective function of the gas inlet amount and the helium concentration, and the optimized gas inlet scale and ammonia concentration are obtained after multiple iterations.

[0129] In one embodiment, the SLSQP algorithm is used to establish a minimum cost objective function of the gas inlet amount and the helium concentration using the net present value of the target helium extraction project, and the optimized gas inlet scale and ammonia concentration are obtained after multiple iterations, including: using the SLSQP algorithm, fixing the ammonia concentration in the gas component of the target helium extraction project unchanged when the net present value is zero, and using the Newton iteration method to solve the minimum gas inlet scale; using the SLSQP algorithm, fixing the initial gas inlet scale of the target helium extraction project unchanged when the net present value is zero, and using the Newton iteration method to solve the minimum ammonia concentration; the minimum gas inlet scale and the minimum ammonia concentration are determined as the optimized gas inlet scale and ammonia concentration, respectively.

[0130] In S4, the gas inlet scale threshold is solved: the helium annual production of S1, the annual investment of S2, and the annual operation cost of S3 are input into the discounted amount flow model, the amount inflow is calculated from the helium price and the corresponding production, the amount outflow is calculated from the annual investment and operation cost, and the net present value of the project is determined. In the case where the net present value of the helium extraction project is zero, the minimum gas inlet scale is solved by fixing the helium concentration of the raw material gas and using the Newton iteration method.

[0131] Solve the helium extraction concentration threshold in S5: the project net present value calculation method is the same as that in S4. In the case of zero net present value of the helium extraction project, the raw material gas inlet scale is fixed, and the Newton iteration method is used to solve the minimum helium extraction concentration.

[0132] Solve the multivariate optimization in S6: solve the joint optimization of variables, take the initial inlet scale and helium concentration as the control variables of the planning, use the SLSQP optimization algorithm, establish the minimum comprehensive cost objective function of the inlet amount and helium concentration, calculate the optimal inlet scale and optimal helium concentration after multiple iterations, and obtain the target function surface and optimal solution distribution diagram according to the calculation result.

[0133] In S4, the annual helium production, investment and operation cost are brought into the discounted amount flow economic evaluation model, the amount inflow is calculated from the helium price and the corresponding production, the amount outflow is calculated from the annual investment and operation cost, and the net present value of the project is determined. In the case of zero net present value of the helium extraction project, the raw material gas helium concentration is fixed, and the Newton iteration method is used to solve the minimum inlet scale.

[0134] For the combined process of membrane separation and low-temperature distillation of the co-production LNG helium extraction engineering project, since the raw material gas is a pipeline commodity gas, the income and cost calculation does not consider the income of the pipeline commodity gas output and the inlet cost of the pipeline commodity gas. The annual helium production is calculated by S1 according to the gas field attenuation law. The helium price is referred to the market price. According to the helium production and the corresponding price, the amount inflow of the project is estimated.

[0135] The annual investment of the project is estimated according to S2 according to the known helium extraction scale, the construction investment forms the depreciation of assets; the annual operation cost is estimated according to the inlet scale of the raw material gas (gas field production); according to the annual investment and operation cost, the amount outflow of the project is estimated.

[0136] Figure 4 The amount flow diagram of the target helium extraction project in the embodiment of the present application is shown in the amount flow table. The net present value (NPV) of the helium extraction engineering project is calculated according to the amount inflow and the amount outflow in the amount flow table, as shown in Figure 4 .

[0137] The helium production, annual investment and annual operation cost of the co-production LNG helium extraction project of a certain gas field calculated in S1 to S4 are brought into the discounted amount flow economic evaluation model of the project.

[0138] The Newton iteration method is used to solve the minimum inlet flow Q that satisfies NPV (Q n ) = 0, and the iteration result is shown in Table 7.

[0139] Table 7

[0140] Iteration number <![CDATA[Q(m 3 / h)]]> NPV(Q) NPV'(Q) Next step Q 1 500 -158.2 1.24 627.58 2 627.58 12.5 1.31 632.45 3 632.45 -0.003 1.30 632.45

[0141] Convergence speed: 3 iterations can reach the accuracy requirement of 10 -6 .

[0142] Figure 5 This is a distribution diagram of the intake scale threshold value under different helium concentrations for the target helium extraction project in the embodiment of the present invention. By setting different intake scales and repeating this step, the minimum economic helium concentration under different intake scales can be calculated, such as Figure 5 When the helium price is 140 yuan / m3, the intake scale is 9000000Nm 3 / day, the answer is C He min =0.1047 mol%.

[0143] Input raw gas helium concentration C He 、Helium Price A He Curve (time series), discount rate (such as 6%), initial investment, etc. The discount rate and helium price are fixed values ​​and have nothing to do with the market economy.

[0144] Define the objective function: set NPV = 0, fix the helium concentration of the raw gas, and reverse calculate the initial intake scale.

[0145] Amount flowing into Revenue t =A He ×P He,t .

[0146] Inflows: Investment (apportioned over the construction period), operating costs, and taxes (15% income tax and refundable VAT). Taxes are fixed values ​​and have nothing to do with the market economy. The Net Present Value (NPV) formula is as follows:

[0147]

[0148] Among them, Revenue t is the revenue of the production cycle t, OPEX t is the operating fee for production cycle t, Tax t is the tax for the production cycle t, t is the production cycle in years, r is the discount rate; T is the full cycle; I t The first year's investment; content It is the construction period of the specified initial helium concentration of the feed gas, in years.

[0149] Newton iteration method is used to solve the problem of NPV(Q n )=0 minimum intake air flow Q:

[0150]

[0151] Derivative calculation: numerical difference method approximation:

[0152]

[0153] Where h is time, unit hour; n is iteration number.

[0154] The minimum threshold of air intake scale can be solved efficiently by Newton iteration method, and dynamic analysis combined with cash flow model provides quantitative support for the feasibility of helium extraction project.

[0155] In S5, the annual helium production, investment and operating cost are brought into the discounted cash flow economic evaluation model. The cash inflow is calculated from the helium price and the corresponding production, and the cash outflow is calculated from the annual investment and operating cost. The net present value of the project is determined. In the case of zero net present value of the helium extraction project, the fixed raw gas intake scale is used to solve the minimum helium extraction concentration C He min .

[0156] Cash flow Revenue t The net present value (NPV) and calculation method are the same as in S4. The minimum helium extraction concentration C He is solved by Newton iteration method to satisfy NPV(C He min , and the iteration results are shown in Table 8.

[0157] Table 8

[0158] Iteration number C He (%)]] NPV(C He )]]> NPV(C He )]]> Next C He ]]> 1 2 -125.8 9850 1.723 2 0.1723 68.4 8420 1.4416 3 0.14416 12.1 7980 1.3246 4 0.1176 -0.00002 7905 1.3246

[0159] Input parameters: raw gas intake scale Q, helium price curve (time series), discount rate (such as 6%), initial investment, etc.

[0160] Define the objective function: let NPV = 0, fix the intake scale, and back-propagate the minimum helium concentration C He min .

[0161] Cash flow Revenue t The net present value (NPV) and calculation method are the same as in S42. The minimum helium extraction concentration C He is solved by Newton iteration method to satisfy NPV(C He min :

[0162]

[0163] Derivative calculation: numerical difference method approximation:

[0164]

[0165] The Newton iteration method can be used to efficiently solve the minimum threshold of helium extraction concentration, and combined with the dynamic analysis of the monetary flow model, it can provide quantitative support for the feasibility of the helium extraction project.

[0166] Convergence speed: 4 iterations can reach the accuracy requirement of 10 -6 .

[0167] Figure 6 This is a distribution diagram of the minimum economic helium concentration under different intake scales for the target helium extraction project in the embodiment of the present invention. By setting different helium concentrations and repeating this step, the minimum intake scale under different initial intake scales can be calculated, such as Figure 6 When the price of helium is 140 yuan / m3 and the concentration of helium is 0.1176%, the solution is Q0 min =7250023m 3 / Year.

[0168] In S6, the variables are jointly optimized and solved. The initial intake scale and helium concentration are used as the control variables for planning and solving. The SLSQP optimization algorithm is used to establish the objective function of minimizing the comprehensive cost of intake volume and helium concentration. After multiple iterations, the optimal intake scale and optimal helium concentration are calculated.

[0169] Figure 7 This is a feasible domain analysis diagram of the SLSQP joint optimization of the target helium extraction project in an embodiment of the present invention. Figure 8 The target function surface and optimal solution distribution diagram of the helium extraction project in the embodiment of the present invention are shown in FIG. The optimal intake scale and optimal helium concentration calculated after S6 iteration are shown in FIG. Figure 7 As shown, the obtained objective function surface and optimal solution distribution diagram are as follows Figure 8 shown.

[0170] The problem of minimizing a constrained multivariate function is converted into multiple small quadratic programming subproblems. Through multiple iterations and updates, the inherent connections between the multivariate variables are continuously strengthened. The SLSQP algorithm has fast and global convergence, making it the optimal method for strengthening the inherent connections between the variables in the overall cost of helium extraction projects.

[0171] In order to save the energy consumption of helium extraction and reduce the operating cost of the helium extraction project, two decision variables, namely the intake air volume and helium concentration, were extracted based on the co-production LNG helium extraction process. A corresponding energy consumption optimization process was established, and the SLSQP algorithm was used to optimize the model.

[0172] First define the variable range and initial value X0:

[0173] 300≤Q≤15000;

[0174] 0.005%≤C He ≤0.1%;

[0175] X0 = [800, 0.015%];

[0176] Through the SLSQP joint optimization feasible region analysis, grid data and multiple iterations were generated, and the optimal solution was obtained by optimizing path tracking, as shown in Table 9.

[0177] Table 9

[0178] Iteration number Q(m 3 / h)]]> C He (%)]] Objective function value Constraint violation amount 0 800 0.15 866.67 -125.8 10 712.4 0.142 782.9 -12.3 20 634.8 0.133 757.2 0 Optimal solution 632.5 0.132 757.1 0

[0179] The optimal solution is Q = 63.25 m 3 / h, C He = 0.132%, and the objective function value is 7.571 million yuan. The parameter sensitivity ranking is helium price, initial investment and operating cost, as shown in Table 10.

[0180] Table 10

[0181]

[0182]

[0183] By imposing constraints on the decision variables, the total cost of helium extraction was mathematically simulated, and the objective function was set as the comprehensive cost of helium extraction (the weighted sum of gas flow and helium concentration), as follows:

[0184]

[0185] Where ω1 is the cost weight of gas flow, which is positively correlated with the size of the equipment; ω2 is the concentration cost weight, and the lower the concentration, the higher the cost of helium extraction.

[0186] The net present value (NPV) formula is described in S4.

[0187] Constraints:

[0188] NPV(Q, C He ) ≥ 0;

[0189] Q min ≤ Q ≤ Q max ;

[0190] C min ≤ C He ≤ C max ;

[0191] Process constraints: Q × C He ≥ minimum annual helium production.

[0192] The SLSQP joint optimization feasible region analysis is a gradient optimization method for processing nonlinear constraints, can accurately process boundary constraints, and is suitable for solving minimum value problems. The embodiment of the present application takes the intake flow and the helium concentration as the estimated value, takes the intake flow and the helium concentration of 20 measurement points as the estimated value for solving, iteratively solves the square sum of the actual value and the estimated value, and seeks the minimum value of f(Q,C He ) by the minimum value method.

[0193] By means of the SLSQP multivariable joint optimization, the economic optimal combination of process parameters can be accurately positioned, the investment scale and the helium extraction cost are balanced, and a quantitative basis is provided for the investment decision of the helium extraction project.

[0194] By comparing the single variable factor optimization method with the SLSQP global optimization scheme, it is shown that the SLSQP scheme is most widely applicable in the comparison of multiple schemes, as shown in Table 11.

[0195] Table 11

[0196]

[0197] According to the functional relationship among the initial intake scale of the cogeneration LNG helium extraction engineering project, the gas component and yield, the investment and operation cost, the embodiment of the present application estimates the technical and economic evaluation indexes of each link and establishes an economic evaluation model, and finally calculates the minimum economic intake scale of the cogeneration LNG helium extraction project by planning solving.

[0198] The embodiment of the present application is directed to a co-production LNG helium extraction engineering project using membrane separation and low-temperature distillation technology. The life period of each year and the gas production of each year are obtained by analogy method and exponential decline method. The relationship function of gas production (raw gas and helium), investment of each device, and operation cost of each device with initial gas inlet scale and gas component is established in sequence by using analytic function method. The relationship function of investment of each device with initial gas inlet scale and gas component is constructed by LASSO regression analysis, and the annual investment is determined by the annual investment proportion. The relationship function of sub-item operation cost with initial gas inlet scale and gas component is constructed by using LightGBM based on gradient boosting decision tree, and the annual operation cost of each item is determined by the change of production. The above calculation results are brought into the discounted amount flow economic evaluation model of the project, so as to establish a nonlinear analytic function of initial gas inlet scale and net present value, taking the initial gas inlet scale as the control variable of programming solution, and measuring the initial gas inlet scale when the break-even point (i.e. the net present value is equal to zero) is met. The initial gas inlet scale is the minimum economic gas inlet scale of the helium extraction engineering. The calculation method of co-production LNG helium extraction concentration and gas inlet scale threshold value disclosed in the embodiment of the present application can clearly identify the influencing factors of project economy and technical feasibility, provide quantitative basis for engineering decision, and provide method support for economically and effectively obtaining key mineral resources and natural gas industry chain value-added.

[0199] The embodiment of the present application also proposes a co-production LNG helium extraction process parameter threshold determination device, which has a similar principle to the co-production LNG helium extraction process parameter threshold determination method, which will not be described here.

[0200] Figure 9 A schematic diagram of a co-production LNG helium extraction process parameter threshold determination device provided in the embodiment of the present application is shown in Figure 9 As shown in the figure, the co-production LNG helium extraction process parameter threshold determination device can include:

[0201] The acquisition module 901 is configured to acquire basic data of a target helium extraction project and historical data of a similar helium extraction project. The similar helium extraction project has the same helium extraction process as the target helium extraction project. The basic data includes the gas field scale and the gas component of the target helium extraction project. The historical data includes the gas field scale, the gas component, the initial gas inlet scale, the gas production, the gas inlet flow of each device, the investment of each device, and the operation cost of the target helium extraction project.

[0202] The analogy module 902 is configured to determine the initial gas inlet scale and the gas production of the target helium extraction project according to the basic data and the historical data by using an analogy algorithm.

[0203] The regression analysis module 903 is configured to determine the gas inlet flow of each device and the investment of each device in the target helium extraction project according to the basic data and the historical data by using a LASSO regression analysis algorithm.

[0204] The cost analysis module 904 is configured to input the initial gas inlet scale, gas yield, gas inlet flow of each device, investment of each device, and basic data of the target helium extraction project into a plurality of pre-trained cost analysis models to obtain cost analysis results. The plurality of pre-trained cost analysis models are obtained by training a plurality of LightGBM models using historical data. Each LightGBM model is trained using different operating cost items.

[0205] The optimization module 905 is configured to establish a minimum cost objective function of the gas inlet amount and the helium concentration by using an SLSQP algorithm according to the cost analysis results and the investment of each device in the target helium extraction project, and obtain an optimized gas inlet scale and ammonia concentration after multiple iterations.

[0206] In an embodiment, the analogy module 902 is specifically configured to:

[0207] The analogy algorithm is used to determine the initial gas inlet scale of the target helium extraction project according to the gas field scale of the target helium extraction project, the gas field scale of the similar helium extraction project, and the initial gas inlet scale.

[0208] In an embodiment, the analogy module 902 is specifically configured to:

[0209] The relationship function between the gas components, the initial gas inlet scale, and the gas yield of the similar helium extraction project is determined according to the historical data.

[0210] The analogy algorithm is used to determine the initial gas yield of the target helium extraction project according to the relationship function, the gas components of the target helium extraction project, and the initial gas inlet scale. The initial gas yield is the gas yield of the target helium extraction project in the first year.

[0211] The gas yield of each year in the life cycle of the target helium extraction project is determined according to the initial gas yield and a preset attenuation coefficient.

[0212] In an embodiment, the cogeneration LNG helium extraction process parameter threshold determination apparatus can further include a correction module configured to:

[0213] If the nitrogen concentration in the gas components of the target helium extraction project is greater than a first preset threshold, and / or the carbon dioxide concentration is greater than a second preset threshold, the initial gas yield of the target helium extraction project is corrected by using a correction factor. The correction factor is determined by the nitrogen concentration and / or the carbon dioxide concentration in the gas components of the target helium extraction project.

[0214] In an embodiment, the regression analysis module 903 is specifically configured to:

[0215] The LASSO regression analysis algorithm is used to establish the following impurity sensitivity coefficient regression equation according to the gas components of the similar helium extraction project:

[0216] γ = γ0 + γ1C He + γ2C N2 + γ3C CO2 ;

[0217] wherein γ is the impurity sensitivity coefficient; γ0 is the methane sensitivity coefficient; γ1 is the helium sensitivity coefficient; γ2 is the nitrogen sensitivity coefficient; γ3 is the carbon dioxide sensitivity coefficient; C He is the helium concentration in the gas composition of the similar helium extraction project or the target helium extraction project; C N2 is the nitrogen concentration in the gas composition of the similar helium extraction project or the target helium extraction project; C CO2 is the carbon dioxide concentration in the gas composition of the similar helium extraction project or the target helium extraction project;

[0218] According to the basic data, the historical data and the impurity sensitivity coefficient regression equation, the gas inlet flow of each device in the target helium extraction project and the investment of each device are determined.

[0219] In an embodiment, the regression analysis module 903 is specifically configured to:

[0220] According to the nitrogen concentration in the gas composition of the target helium extraction project, the nitrogen concentration in the gas composition of the similar helium extraction project, the initial gas inlet scale of the similar helium extraction project, the initial gas inlet scale of the target helium extraction project and the gas inlet flow of each device in the similar helium extraction project, the gas inlet flow of each device in the target helium extraction project is determined.

[0221] According to the following formula, the investment of each device in the target helium extraction project is determined according to the basic data, the historical data and the impurity sensitivity coefficient regression equation:

[0222]

[0223] wherein I j is the investment of the jth device in the target helium extraction project; I j.base is the investment of the jth device in the similar helium extraction project; γ is the impurity sensitivity coefficient; ΔC impurity is the percentage of the impurity concentration exceeding the preset reference value; F j is the flow of the jth device of the target helium extraction project; F j .base is the flow of the jth device of the similar helium extraction project.

[0224] In an embodiment, the multiple operation costs include one or any combination of the operation power cost, the maintenance cost, the material cost, the fuel cost, the personnel cost, the repair cost, the management cost and the business cost.

[0225] In an embodiment, the co-produced LNG helium extraction process parameter threshold value determination device can further include a training module configured to:

[0226] establishing a LightGBM model corresponding to each of the plurality of operation costs;

[0227] The LightGBM model corresponding to each operation cost is trained by using the gas field scale, gas component, initial gas inlet scale, gas yield, gas inlet flow of each device, investment of each device, and each operation cost of a similar helium extraction project in historical data, to generate a plurality of cost analysis models.

[0228] In one embodiment, the optimization module 905 is specifically configured to:

[0229] determining the net present value of the target helium extraction project according to the cost analysis result and the investment of each device in the target helium extraction project;

[0230] using the SLSQP algorithm to establish a minimum cost objective function of the gas inlet amount and the helium concentration by using the net present value of the target helium extraction project, and obtaining the optimized gas inlet scale and ammonia concentration after multiple iterations.

[0231] In one embodiment, the optimization module 905 is specifically configured to:

[0232] using the SLSQP algorithm to fix the ammonia concentration in the gas component of the target helium extraction project unchanged in the case of zero net present value, and using the Newton iteration method to solve the minimum gas inlet scale;

[0233] using the SLSQP algorithm to fix the initial gas inlet scale of the target helium extraction project unchanged in the case of zero net present value, and using the Newton iteration method to solve the minimum ammonia concentration;

[0234] determining the minimum gas inlet scale and the minimum ammonia concentration as the optimized gas inlet scale and ammonia concentration, respectively.

[0235] Compared with the prior art, the embodiment of the present application determines the threshold of the co-production LNG helium extraction process parameter by obtaining the basic data of the target helium extraction project and the historical data of similar helium extraction projects; the helium extraction processes of the similar helium extraction projects and the target helium extraction project are the same; the basic data includes the gas field scale and the gas component of the target helium extraction project; the historical data includes the gas field scale, the gas component, the initial gas inlet scale, the gas yield, the gas inlet flow of each device, the investment of each device and multiple operation costs of the similar helium extraction projects; the initial gas inlet scale and the gas yield of the target helium extraction project are determined according to the basic data and the historical data by using the analogy algorithm; the gas inlet flow of each device and the investment of each device of the target helium extraction project are determined according to the basic data and the historical data by using the LASSO regression analysis algorithm; the initial gas inlet scale, the gas yield, the gas inlet flow of each device, the investment of each device and the basic data of the target helium extraction project are input into multiple pre-trained cost analysis models respectively to obtain cost analysis results; the multiple pre-trained cost analysis models are obtained by training multiple LightGBM models respectively using the historical data; the operation cost items used for training each LightGBM model are different; according to the cost analysis results and the investment of each device in the target helium extraction project, the SLSQP algorithm is used to establish a minimum cost objective function of the gas inlet scale and the helium concentration, and the optimized gas inlet scale and the ammonia concentration are obtained after multiple iterations, which can improve the comprehensiveness of the co-production LNG helium extraction process parameter threshold determination, improve the accuracy and efficiency of the co-production LNG helium extraction process parameter threshold determination, and reduce the process energy consumption and cost.

[0236] According to the embodiment of the present application, a calculation method and device for the threshold of the co-production LNG helium extraction concentration and gas inlet scale are provided, which are used for the co-production LNG helium extraction engineering project adopting membrane separation and low-temperature distillation technology. Analytic function method is used to establish the relationship function between the gas yield (raw material gas and helium), device investment, operation cost and initial gas inlet scale, gas component respectively and sequentially, and the discount amount flow economic evaluation model of the project is brought in, so as to establish a nonlinear analytic function of the initial gas inlet scale and the net present value. The initial gas inlet scale is taken as the control variable of the planning solution, and the initial gas inlet scale satisfying the break-even (i.e. the net present value is equal to zero) is measured and calculated. The initial gas inlet scale is the minimum economic gas inlet scale of the helium extraction engineering. The calculation method for the threshold of the co-production LNG helium extraction concentration and gas inlet scale can clearly identify the influencing factors of the project economy and technical feasibility, provide quantitative basis for engineering decision-making, and provide method support for economically and effectively obtaining key mineral resources and natural gas industry chain value-added.

[0237] The embodiment of the present application also provides a computer device, Figure 10As an example of the computer device in the embodiment of the present application, the computer device 1000 comprises a memory 1010, a processor 1020, and a computer program 1030 stored in the memory 1010 and executable on the processor 1020, and the processor 1020 implements the above-mentioned method for determining the threshold of the process parameters of the co-production LNG helium extraction process when executing the computer program 1030.

[0238] The embodiment of the present application also provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the above-mentioned method for determining the threshold of the process parameters of the co-production LNG helium extraction process.

[0239] The embodiment of the present application also provides a computer program product, which comprises a computer program, and the computer program is executed by a processor to implement the above-mentioned method for determining the threshold of the process parameters of the co-production LNG helium extraction process.

[0240] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program codes.

[0241] The present application is described with reference to flowcharts and / or block diagrams of the method, device (system), and computer program product according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to produce a machine, so that the instructions executed by the computer or other programmable data processing devices produce a device implemented in the flowcharts and / or block diagrams. Figure 1 The function of one flow or multiple flows and / or blocks Figure 1 The function of one block or multiple blocks.

[0242] These computer program instructions can also be stored in a computer readable memory capable of guiding the computer or other programmable data processing devices to work in a specific way, so that the instructions stored in the computer readable memory produce a manufactured product including instruction devices, which implement the flowcharts and / or block diagrams. Figure 1 The function of one flow or multiple flows and / or blocks Figure 1 The function of one block or multiple blocks.

[0243] These computer program instructions can also be loaded into a computer or other programmable data processing devices, so that a series of operational steps are performed on the computer or other programmable data processing devices to generate a computer implemented process, so that the instructions executed on the computer or other programmable data processing devices provide a process for implementing the flowchart Figure 1 one flow or multiple flows and / or the functions specified in the block Figure 1 one flow or multiple flows and / or the functions specified in the block

[0244] The above described specific embodiments, the purpose, technical solutions and beneficial effects of the present application are further described in detail, it should be understood that the above described is only a specific embodiment of the present application, and is not used to limit the protection scope of the present application, any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A method for determining process parameter thresholds for helium extraction from liquefied natural gas (LNG) co-production, characterized in that: include: Obtain basic data of the target helium extraction project and historical data of similar helium extraction projects; the similar helium extraction projects and the target helium extraction project use the same helium extraction process; the basic data includes the gas field size and gas composition of the target helium extraction project; the historical data includes the gas field size, gas composition, initial gas intake size, gas production, gas intake flow rate of each device, investment of each device, and various operating costs of similar helium extraction projects; Using analogy algorithms, based on basic data and historical data, determine the initial gas intake scale and gas production of the target helium extraction project; Using the least absolute shrinkage and LASSO regression analysis algorithms, based on basic and historical data, we determined the inlet flow rate and investment for each unit in the target helium extraction project. Input the initial gas intake scale, gas production, gas intake flow rate of each device, investment of each device and basic data of the target helium extraction project into multiple pre-trained cost analysis models to obtain cost analysis results; the multiple pre-trained cost analysis models are obtained by training multiple Lightweight Gradient Boosting Machine (LightGBM) models using historical data; each LightGBM model is trained using different operating cost items; According to the cost analysis results and the investment of each device in the target helium extraction project, the sequential least squares programming (SLSQP) algorithm is used to establish the cost minimization objective function of the intake volume and helium concentration. After multiple iterations, the optimized intake scale and ammonia concentration are obtained.

2. The method according to claim 1, wherein Using analogy algorithms, based on basic data and historical data, the initial gas intake scale and gas production of the target helium extraction project are determined, including: An analogy algorithm is used to determine the initial gas intake scale of the target helium extraction project based on the gas field scale of the target helium extraction project, the gas field scale of similar helium extraction projects, and the initial gas intake scale.

3. The method according to claim 2, wherein Using analogy algorithms, based on basic data and historical data, the initial gas intake scale and gas production of the target helium extraction project are determined, including: Based on historical data, determine the relationship function between gas composition, initial gas intake scale and gas production of similar helium extraction projects; An analogy algorithm is used to determine the initial gas production of the target helium extraction project based on the relationship function, the gas composition of the target helium extraction project, and the initial gas intake scale; the initial gas production is the gas production of the target helium extraction project in the first year; Based on the initial gas production and the preset attenuation coefficient, the annual gas production during the life of the target helium extraction project is determined.

4. The method according to claim 3, wherein After determining the initial gas production of the target helium extraction project by using an analogy algorithm based on the relationship function, the gas composition of the target helium extraction project, and the initial gas intake scale, the method further includes: If the nitrogen concentration in the gas components of the target helium extraction project is greater than a first preset threshold, and / or the carbon dioxide concentration is greater than a second preset threshold, the initial gas production of the determined target helium extraction project is corrected using a correction factor; the correction factor is determined by the nitrogen concentration and / or carbon dioxide concentration in the gas components of the target helium extraction project.

5. The method according to claim 4, wherein LASSO regression analysis algorithm is used to determine the inlet flow rate and investment of each unit in the target helium extraction project based on basic data and historical data, including: Using the LASSO regression analysis algorithm, based on the gas composition of similar helium extraction projects, the following impurity sensitivity coefficient regression equation is established: γ=γ0+γ1C He +γ2C N2 +γ3C CO2 ; Among them, γ is the impurity sensitivity coefficient; γ0 is the methane sensitivity coefficient; γ1 is the helium sensitivity coefficient; γ2 is the nitrogen sensitivity coefficient; γ3 is the carbon dioxide sensitivity coefficient; C He The helium concentration in the gas components of similar helium extraction projects or target helium extraction projects; C N2 C is the nitrogen concentration in the gas components of similar helium extraction projects or target helium extraction projects; CO2 The carbon dioxide concentration in the gas components of similar helium extraction projects or the target helium extraction project; Based on basic data, historical data and the impurity sensitivity coefficient regression equation, the intake flow rate of each device and the investment of each device in the target helium extraction project are determined.

6. The method according to claim 5, wherein Based on basic data, historical data and the impurity sensitivity coefficient regression equation, determine the inlet flow rate and investment of each device in the target helium extraction project, including: Determine the gas intake flow rate of each device in the target helium extraction project based on the nitrogen concentration in the gas components of the target helium extraction project, the nitrogen concentration in the gas components of similar helium extraction projects, the initial gas intake scale of similar helium extraction projects, the initial gas intake scale of the target helium extraction project, and the gas intake flow rate of each device in similar helium extraction projects; According to the following formula, based on basic data, historical data and the impurity sensitivity coefficient regression equation, the investment for each device in the target helium extraction project is determined: Among them, I j The investment in the jth device in the target helium project; j.base is the investment of the jth device in a similar helium extraction project; γ is the impurity sensitivity coefficient; ΔC impurity The percentage of impurity concentration exceeding the preset reference value; F j The flow rate of the jth device of the target helium supply project; F j .base is the flow rate of the jth device of a similar helium extraction project.

7. The method according to claim 1, wherein The plurality of operating costs include one or any combination of operating power costs, maintenance costs, material costs, fuel costs, personnel costs, repair costs, management costs and business costs.

8. The method according to claim 7, wherein Before obtaining the cost analysis results, the following steps are also included: First, the initial gas intake scale, gas production, gas flow rate of each unit, investment of each unit, and basic data of the target helium extraction project are input into multiple pre-trained cost analysis models. Second, the cost analysis results are obtained. Establish a LightGBM model corresponding to each of the multiple operation costs; Using historical data on gas field size, gas composition, initial gas intake scale, gas production, gas intake flow rate of each device, investment of each device, and each operating cost of similar helium extraction projects, we trained the LightGBM model corresponding to the operating costs and generated multiple cost analysis models.

9. The method according to claim 7, wherein Based on the cost analysis results and the investment in each device in the target helium extraction project, the Sequential Least Squares Programming (SLSQP) algorithm was used to establish a cost minimization objective function for the intake volume and helium concentration. After multiple iterations, the optimized intake scale and ammonia concentration were obtained, including: Determine the net present value of the target helium extraction project based on the cost analysis results and the investment in each device in the target helium extraction project; The SLSQP algorithm was adopted and the net present value of the target helium extraction project was used to establish the cost minimization objective function of the intake volume and helium concentration. After multiple iterations, the optimized intake scale and ammonia concentration were obtained.

10. The method according to claim 9, wherein The SLSQP algorithm was used to establish a cost minimization objective function for intake volume and helium concentration using the net present value of the target helium extraction project. After multiple iterations, the optimized intake scale and ammonia concentration were obtained, including: Using the SLSQP algorithm, under the condition that the net present value is zero, the ammonia concentration in the gas composition of the fixed target helium extraction project remains unchanged, and the Newton substitution method is used to solve the minimum intake scale; Using the SLSQP algorithm, under the condition that the net present value is zero, the initial gas intake scale of the fixed target helium extraction project remains unchanged, and the Newton selection method is used to solve the minimum ammonia extraction concentration; The minimum air intake scale and the minimum ammonia concentration are determined as the optimized air intake scale and ammonia concentration respectively.

11. A device for determining the threshold value of helium extraction process parameters for co-production of liquefied natural gas (LNG), characterized in that: include: An acquisition module is configured to acquire basic data of a target helium extraction project and historical data of similar helium extraction projects; the similar helium extraction projects use the same helium extraction process as the target helium extraction project; the basic data includes the gas field size and gas composition of the target helium extraction project; the historical data includes the gas field size, gas composition, initial gas intake size, gas production, gas intake flow rate of each device, investment of each device, and various operating costs of similar helium extraction projects; The analogy module is used to determine the initial gas intake scale and gas production of the target helium extraction project based on basic data and historical data using an analogy algorithm; The regression analysis module is used to determine the air intake flow rate and investment of each device in the target helium extraction project based on basic data and historical data using the LASSO regression analysis algorithm; The cost analysis module is used to input the initial gas intake scale, gas production, gas intake flow rate of each device, investment of each device, and basic data of the target helium extraction project into multiple pre-trained cost analysis models to obtain cost analysis results; the multiple pre-trained cost analysis models are obtained by training multiple Lightweight Gradient Boosting Machine (LightGBM) models using historical data; each LightGBM model is trained using different operating cost items; The optimization module is used to establish a cost minimization objective function of the intake volume and helium concentration based on the cost analysis results and the investment of each device in the target helium extraction project, using the sequential least squares programming (SLSQP) algorithm, and obtain the optimized intake scale and ammonia concentration after multiple iterations.

12. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 10 is implemented.

13. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 10 is implemented.

14. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 10 is implemented.

Images