Light-hydrogen coal storage multi-source coupling methanol preparation system and collaborative optimization scheduling method

Abstract

The invention relates to a light hydrogen coal storage multi-source coupling methanol preparation system and a collaborative optimization scheduling method thereof, and belongs to the technical field of energy chemical engineering and renewable energy coupling system optimization scheduling. The system comprises a photovoltaic power generation unit, a water electrolysis hydrogen production unit, a coal gasification unit, a water-gas conversion unit, a low-temperature methanol washing unit, a methanol synthesis and rectification unit, a carbon capture unit, a CO2 methanol production unit, an electric energy storage unit and a hydrogen storage unit, and multi-energy coupling is realized through a material flow and an energy flow. Aiming at the uncertainty of photovoltaic output, an adaptive dynamic optimization scheduling method based on a moment information constraint set is provided, an optimization model including economy, carbon emission and light abandoning rate is constructed, and a linear fraction decision rule is adopted to convert a problem into a mixed integer linear programming problem for efficient solving. According to the method, the running reliability of the system is guaranteed, the carbon emission per ton of methanol is remarkably reduced, the running cost is effectively reduced, the photovoltaic fluctuation is efficiently absorbed, and the method is suitable for optimal scheduling of a large-scale low-carbon methanol production system.

Description

Technical Field

[0001] This invention belongs to the field of energy chemical and renewable energy coupled system optimization scheduling technology, specifically involving a multi-source coupled methanol production system integrating photovoltaic, hydrogen energy, coal chemical and energy storage and its collaborative optimization scheduling method. Technical Background

[0002] Methanol, as an important clean fuel and chemical feedstock, is currently mainly produced via coal, resulting in high carbon intensity. Utilizing renewable energy sources such as photovoltaics to produce "green hydrogen" and introducing it into the coal-to-methanol process to replace some of the "ash hydrogen" generated by coal gasification is an effective way to reduce the carbon intensity of methanol products. However, photovoltaic power generation is characterized by significant intermittency and high uncertainty, creating an irreconcilable contradiction with the continuous, stable, and rigid operating characteristics required by chemical production processes. The drastic fluctuations in photovoltaic output make it difficult to predict the hydrogen production of the directly coupled water electrolysis hydrogen production unit, leading to unstable hydrogen flow into the chemical system and hindering reliable coupling with continuous and stable chemical processes.

[0003] Among existing approaches, stochastic programming strategies heavily rely on accurate photovoltaic power output probability distribution models, but obtaining precise distributions in practice is extremely difficult, leading to high decision-making risks. Traditional robust optimization methods often generate overly conservative scheduling strategies, which, while improving system reliability, sacrifice operational economy and result in high costs. Furthermore, existing methods generally suffer from high computational complexity and insufficient inter-subsystem coordination, making it difficult to achieve globally optimal economic and environmental benefits while ensuring system operational reliability. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art and provide a collaborative optimization scheduling method for a multi-source coupled photovoltaic-hydrogen-coal-storage methanol production system, so as to improve the system's economy, low carbon emissions and operational reliability under photovoltaic uncertainty, and solve the problems of conservative scheduling results, complex calculations and insufficient system coordination capabilities of the prior art under high-dimensional uncertainty.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] First, a multi-source coupled methanol production system based on photovoltaic, hydrogen, coal, and storage is provided. This low-carbon methanol production system includes a pulverized coal preparation and gasification unit, a water-gas shift unit, a low-temperature methanol washing unit, a methanol synthesis and distillation unit, a photovoltaic power generation unit, an electrolytic water hydrogen production unit, a carbon capture unit, and a CO2 methanol production unit. The system organically couples these units through material and energy flows, wherein:

[0007] The photovoltaic power generation unit serves as the sole power source for the system, and its output terminal is connected to the power input terminal of each unit to provide them with DC or AC power.

[0008] The water electrolysis unit for producing hydrogen and oxygen uses proton exchange membrane (PEM) electrolyzer technology. Its oxygen outlet is connected to the gasifier inlet of the pulverized coal preparation and gasification unit, replacing the traditional air separation unit to provide oxygen for coal gasification. Its hydrogen outlet is connected to the inlet of the CO2 to methanol unit.

[0009] The pulverized coal preparation and gasification unit, the water-gas conversion unit, and the low-temperature methanol washing unit are connected in sequence to produce syngas (CO+H2) and remove acidic gases; the CO2-rich tail gas outlet of the low-temperature methanol washing unit is connected to the inlet of the carbon capture unit.

[0010] The carbon capture unit uses membrane separation to capture CO2 in the tail gas; the CO2 product outlet of the membrane separation unit is connected to the inlet of the CO2 to methanol unit; the other inlet of the CO2 to methanol unit is connected to the hydrogen outlet of the water electrolysis to hydrogen and oxygen unit, and methanol is synthesized by catalytic synthesis using the captured CO2 and green hydrogen.

[0011] The system also includes a hydrogen storage device and an oxygen storage device, which are connected between the hydrogen and oxygen outlets of the water electrolysis hydrogen and oxygen production unit and the downstream gas-using unit, respectively. The hydrogen storage device is used to store excess hydrogen generated by electrolysis during peak photovoltaic periods and release it when photovoltaic output is insufficient, so as to ensure the continuous and stable replenishment of hydrogen in the CO2 to methanol unit. The oxygen storage device is used to store excess oxygen and ensure a continuous and stable supply of oxygen to the gasifier.

[0012] The system also includes a battery energy storage system connected to the photovoltaic power generation unit to smooth the photovoltaic power output and provide the system with stable and continuous power.

[0013] Secondly, the present invention further provides a collaborative optimization scheduling method for the coal-to-methanol low-carbon production system, specifically including the following steps:

[0014] S1. Establish a scheduling optimization model for a multi-source coupled methanol production system based on photovoltaic, hydrogen, coal storage, and other sources, specifically:

[0015] S1.1. Mathematical modeling is performed on the general unit for coal-to-methanol production, the photovoltaic power generation unit, the water electrolysis hydrogen production unit, the electric energy storage unit, and the hydrogen storage unit;

[0016] S1.2. Construct the objective function and constraints;

[0017] S2. Uncertainty modeling is performed using moment information constraint sets;

[0018] S3. Establish a sequential decision-making framework and an adaptive dynamic scheduling model;

[0019] S4. An efficient solution method is proposed to solve this scheduling optimization problem, specifically:

[0020] S4.1. Use linear fractional decision rules to perform piecewise linear modeling of sequential decision variables;

[0021] S4.2. Transform the adaptive dynamic scheduling optimization problem into a mixed-integer linear programming problem;

[0022] S4.3. Use the solver to solve the reconstructed mixed-integer linear programming problem to obtain the optimal scheduling strategy;

[0023] Preferably, the specific steps of S1.1 in mathematical modeling the key unit are as follows:

[0024] S1.1.1. Establish a general unit mathematical model for coal-to-methanol production:

[0025] Considering the conservation of matter and energy, the general unit mathematical model for coal-to-methanol production is expressed as follows:

[0026] (1)

[0027] (2)

[0028] (3)

[0029] (4)

[0030] (5)

[0031] (6)

[0032] (7)

[0033] in, The processing flow rate of unit u; The raw material input flow rate for unit u; Output flow rate for unit u; , , The demand for hydrogen, steam, and electricity for unit u; The output ratio of the product in unit u; , , The demand coefficients for hydrogen, steam, and electricity in unit u; , The upper and lower limits of the flow rate processed by unit u; , The figures represent the methanol production of the methanol synthesis and distillation unit and the CO2-to-methanol unit, respectively. The target is the minimum methanol production.

[0034] S1.1.2. Establish the mathematical model of the photovoltaic power generation unit:

[0035] The output power of a photovoltaic power generation system is constrained by variables such as irradiance and temperature. The calculation formula is as follows:

[0036] (8)

[0037] in, The output power of a photovoltaic power generation system is expressed in kW. Represents global horizontal irradiance, W / m 2 ; Indicates the installed capacity of photovoltaic power, in kW; GHI STC GHI, W / m³ under standard test conditions 2 ; Temperature coefficient, % / ℃; T c T represents the temperature of the photovoltaic module under actual conditions, in °C. STC The temperature of the photovoltaic module under standard test conditions is expressed in °C. This indicates the power supply efficiency of the photovoltaic power generation system.

[0038] (9)

[0039] in, Indicates ambient temperature, in °C; The standard operating temperature of the component is indicated in °C.

[0040] The operating constraints of photovoltaic power generation systems are as follows:

[0041] (10)

[0042] S1.1.3. Establish the mathematical model of the battery energy storage unit:

[0043] The energy variation relationship of battery energy storage in adjacent time periods is as follows:

[0044] (11)

[0045] in, Represents the remaining energy stored in electrical energy storage, in kWh; Indicates the self-discharge rate; The charging power of electrical energy storage is expressed in kW. Indicates the charging efficiency of electrical energy storage; The discharge power of the electrical energy storage is expressed in kW. This indicates the discharge efficiency of the electrical energy storage.

[0046] (12)

[0047] in, This indicates the installed capacity of the energy storage module, expressed in kWh.

[0048] The operational constraints of battery energy storage are as follows:

[0049] (13)

[0050] (14)

[0051] (15)

[0052] (16)

[0053] (17)

[0054] in, , These represent the maximum charging power and the maximum discharging power.

[0055] S1.1.4. Establish a mathematical model for the water electrolysis hydrogen production unit:

[0056] An electrolyzer produces hydrogen by consuming electrical energy to electrolyze water. The energy conversion formula is as follows:

[0057] (18)

[0058] in, Input electrical power to the electrolytic cell, kW; Hydrogen production, in kg; The value represents the electrolyzer conversion efficiency; LHV is the lower heating value of hydrogen, 33.3 kWh / kg.

[0059] The operating constraints of the electrolytic cell are as follows:

[0060] (19)

[0061] (20)

[0062] in, The installed capacity of the electrolytic cell is in kW.

[0063] S1.1.5. Establish a mathematical model for the hydrogen storage unit:

[0064] The relationship between the hydrogen storage capacity of the hydrogen storage system in adjacent time periods is as follows:

[0065] (twenty one)

[0066] in, The mass of hydrogen is expressed in kg. and These represent the hydrogen input and output rates of the hydrogen storage tank, respectively, in kg / h.

[0067] (twenty two)

[0068] in, This indicates the installed capacity of the hydrogen storage tank, expressed in kg.

[0069] The operating constraints of the hydrogen storage tank are as follows:

[0070] (twenty three)

[0071] (twenty four)

[0072] (25)

[0073] (26)

[0074] (27)

[0075] in, , The maximum input and output rates of hydrogen in the hydrogen storage tank are given in kg / h.

[0076] Preferably, the objective function in S1.2 includes operating costs that measure economic efficiency, carbon emissions per ton of methanol that measure environmental impact, and photovoltaic curtailment rate that measures the efficiency of renewable energy utilization.

[0077] The formula for calculating the operating cost is as follows:

[0078] (28)

[0079] in, This represents the total operating cost. , , , The figures represent the following costs: photovoltaic power generation cost per kilowatt-hour (RMB / kWh); battery energy storage cost per kilowatt-hour (RMB / kWh); electrolyzer hydrogen production cost (RMB / kg); and hydrogen storage tank cost (RMB / kg).

[0080] The formula for calculating carbon emissions per ton of methanol is as follows:

[0081] (29)

[0082] in, The carbon emission factor per ton of methanol is expressed as tCO2e / t-MeOH. This represents the total direct carbon emissions, tCO2; tCO2e represents the total indirect carbon emissions. This represents the annual methanol production, expressed in tons (t).

[0083] The total direct carbon emissions The calculation formula is as follows:

[0084] (28)

[0085] in, The figure represents the carbon dioxide emissions (t) of the low-temperature methanol washing unit. This indicates the capture efficiency of the carbon capture system;

[0086] The total indirect carbon emissions The calculation formula is as follows:

[0087] (31)

[0088] in The term "carbon emissions implied by photovoltaic equipment" refers to the total CO2 emissions generated during the production, transportation, installation, and disposal of equipment such as photovoltaic panels, spread over the entire life cycle. The term "carbon emissions from other key equipment" refers to the CO2 emissions generated by major chemical equipment such as electrolyzers, hydrogen storage tanks, and reactors during their life cycle. This indicates the carbon emissions generated during the upstream production stage of raw materials.

[0089] The formula for calculating the light rejection rate is as follows:

[0090] (32)

[0091] in, This represents the light rejection rate at time t; This represents the photovoltaic output power at time t; This represents the charging power of the stored energy at time t; This represents the input power of the electrolytic cell at time t. This represents the electrical load of other equipment at time t.

[0092] Preferably, the constraints in S1.2 include operational constraints of key devices, reliability constraints of hydrogen supply, reliability constraints of power supply, charging and discharging power ramping constraints of battery energy storage system, and power ramping constraints of electrolyzer;

[0093] The hydrogen supply reliability constraints are as follows:

[0094] (33)

[0095] in, The hydrogen output rate of the electrolyzer is kg / h; and These represent the hydrogen input and output rates of the hydrogen storage tank, respectively, in kg / h; The hydrogen gas demand rate is kg / h.

[0096] The power supply reliability constraints are as follows:

[0097] (34)

[0098] in, The output power of a photovoltaic power generation system is expressed in kW. and The distribution represents the charging and discharging power of electrical energy storage, in kW; Input electrical power to the electrolytic cell, kW; The total electrical load of other equipment in the production system is expressed in kW.

[0099] The charging and discharging power ramp-up constraints of the battery energy storage system are as follows:

[0100] (35)

[0101] in, This indicates the net output power of the energy storage system; Indicates discharge. Indicates charging; Indicates the maximum power rise rate; Indicates the maximum power decrease rate;

[0102] The power ramp-up constraint of the electrolytic cell is as follows:

[0103] (36)

[0104] (37)

[0105] in, Indicates the operating power of the electrolytic cell; Indicates the maximum power ramp rate; Indicates the maximum power downhill ramp rate; and These represent the minimum and maximum technical output power of the electrolytic cell, respectively.

[0106] Preferably, in step S2, the daily forecast error of photovoltaic power generation is considered as the main source of uncertainty, and a moment information constraint set is used. Describe the distribution of uncertainty:

[0107] (38)

[0108] in, for In the dataset The empirical distribution on, For low-dimensional uncertainty components, The number of components, The radius is the Wasserstein radius. and It is a first-order moment boundary.

[0109] Preferably, in step S3, the sequential decision-making framework is as follows:

[0110] Variables related to the coal-to-methanol process, namely These are considered current decisions, which are determined before the uncertainty materializes;

[0111] In contrast, the scheduling variables for electricity and hydrogen were formulated as sequential response decisions. This was achieved by employing decision rules. These scheduling variables are represented by these rules. These rules explicitly define the decision at each time step as a function of realized uncertainties (such as the actual output of photovoltaic power). This framework allows the system to make adaptive, incremental scheduling based on the latest observed information, thus responding more flexibly and robustly to uncertainty.

[0112] Preferably, in step S3, the general form of the adaptive dynamic optimization scheduling problem is expressed as follows:

[0113]

[0114] st (39)

[0115]

[0116] in, and Indicates the decision-making stage; Includes stages Continuous and binary variables; , , , and It is a coefficient matrix or vector; Representation phase The uncertainty of history, its supporting set express; (That is, the current decision is certain.) This is the current decision; In the above formula, and In the set of moment information constraints The two are equivalent, both referring to the final uncertainty vector of renewable energy.

[0117] Preferably, in S4.1, the specific process of using linear fractional decision rules to perform piecewise linear modeling of sequential decision variables is as follows:

[0118] For the decision variables in formula (39) , is a continuous variable and binary variables A linear fractional decision rule was established, and the formula is as follows:

[0119] (40)

[0120] (41)

[0121] in, ; and It is a continuous coefficient matrix; and It is a discrete coefficient matrix, whose elements are restricted to... Inside; , , , It is a constant matrix that satisfies and The second equation in equations (40) and (41) explains the decision rule from the perspective of the uncertainty partitioning of the proposed constraint set; this decision rule ensures unpredictability, that is, for all constraints... The realization of all Established;

[0122] After using the linear fractional decision rule, the adaptive dynamic optimization scheduling problem formula (39) can be expressed as follows:

[0123] (42)

[0124] in, It is the coefficient matrix or parameters obtained after the transformation.

[0125] Preferably, in step S4.2, the specific process of transforming the adaptive dynamic optimization problem into a mixed-integer linear programming problem is as follows:

[0126] First, constraint reconstruction was performed, and a constraint was constructed. The closed convex hull of the polygon is represented as a polyhedron defined by the left and right limits of the breakpoints:

[0127] (43)

[0128] in,

[0129] (44)

[0130] (45)

[0131] Next, using duality theory, the equation (42) involving... Constraints, namely Equivalently reconstructed as follows:

[0132] (46)

[0133] in, and It is a dual variable. Indicates the number of rows in the constraint;

[0134] Subsequently, the objective function in the scheduling problem (42) is reconstructed into an equivalent mixed-integer linear programming problem:

[0135] (47)

[0136] in It is a satisfaction and The constant auxiliary matrix;

[0137] By combining the objective function reconstruction (47) with the constraint reconstruction (46), the equivalent mixed integer linear programming reconstruction form of the adaptive dynamic optimization scheduling problem (39) can be obtained.

[0138] Compared with the prior art, the technical solution of the present invention has the following advantages:

[0139] This invention deeply couples photovoltaic, hydrogen production, energy storage, and coal chemical processes through material and energy flows, achieving synergistic optimization between the energy and chemical sides. Addressing the uncertainty of photovoltaic output, the adaptive dynamic optimization model based on moment information constraint sets employed in this invention fully utilizes statistical information from historical data to generate robust scheduling strategies that cope with uncertainty disturbances without being overly conservative, effectively balancing the economy and reliability of decision-making. Compared to robust optimization methods, the proposed method reduces costs by 12.5%, and by 2.2% compared to multi-stage sub-robust optimization methods based on Wasserstein fuzzy sets.

[0140] In terms of optimization objectives, the model comprehensively considers operating costs, carbon emissions, and renewable energy integration rates, achieving multi-dimensional synergistic optimization of economics, environmental protection, and energy efficiency, and strongly supporting the green and low-carbon transformation of the system. Compared with traditional coal-to-methanol systems, carbon emissions per ton of methanol are reduced by more than 80%, carbon utilization is increased by 87%, and methanol production per ton of coal is increased by 84%.

[0141] Furthermore, in terms of solution strategy, this invention transforms the complex model into an efficiently solvable mixed-integer linear programming problem through linear fractional decision rules and duality theory. This solves the computational challenge under high-dimensional uncertainty, ensuring the operability and solution efficiency of the complex model in practical applications. Compared with radial decision rules, the proposed method improves performance by 3.1%.

[0142] In summary, this invention provides innovative solutions for system design, uncertainty handling, multi-objective optimization, and solution calculation, comprehensively improving the overall performance of the "photovoltaic-hydrogen-coal-storage" multi-source coupled methanol production system under real uncertain environments. Attached Figure Description

[0143] Figure 1 This is a schematic diagram of the multi-source coupled methanol production system based on photohydrogen and coal storage involved in the present invention.

[0144] Figure 2 This is a flowchart of the collaborative optimization scheduling method involved in the present invention.

[0145] Figure 3 A comparison chart showing the results of different optimization scheduling methods.

[0146] Figure 4 A comparison chart of results for different operating modes. Detailed Implementation

[0147] The present invention will now be described in detail, and the technical solutions in the embodiments of the present invention will be clearly and completely described. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0148] Example 1:

[0149] This embodiment uses the construction of a coal-to-methanol project with an annual output of 800,000 tons in Xinjiang, China as an example to illustrate the implementation of the present invention.

[0150] Reference Figure 1As shown, the coal-to-methanol low-carbon production system based on multi-energy coupling involved in this invention includes the following units: a coal gasification unit, a water-gas conversion unit, a low-temperature methanol washing unit, a methanol synthesis and distillation unit, a photovoltaic power generation unit, an electrolysis water hydrogen production unit, a carbon capture unit, and a CO2-to-methanol unit; the system organically couples each unit through material flow and energy flow, wherein:

[0151] The coal gasification unit employs dry pulverized coal gasification technology, with a designed capacity to meet the annual methanol production needs of 800,000 tons. The gasifier inlet is connected to the oxygen outlet of the water electrolysis hydrogen-oxygen production unit, replacing the traditional air separation unit. The designed raw coal processing capacity is 70 t / h, with the designed input coal composition being 1.36% moisture, 16.52% ash, 25.25% volatile matter, 56.86% fixed carbon, 86.3% carbon element content, and 4.62% hydrogen element content. The oxygen requirement is 40 t / h.

[0152] The water-gas shift unit is used to adjust the H2 / CO ratio in the synthesis gas to meet the requirements of subsequent methanol synthesis; the low-temperature methanol washing unit is used to remove acidic gases (such as H2S and CO2) from the synthesis gas; the methanol synthesis and distillation unit uses low-pressure gas-phase synthesis technology to synthesize crude methanol and distill it to obtain the product methanol.

[0153] The carbon capture unit uses membrane separation to capture CO2 in the tail gas of the low-temperature methanol washing unit, with a designed capture efficiency of 95%. The CO2 to methanol unit uses catalytic hydrogenation to synthesize methanol from the captured CO2 and green hydrogen, as a supplement to the main methanol synthesis unit.

[0154] The photovoltaic power generation unit is a ground-mounted photovoltaic power station, serving as the system's sole power source, with a total installed capacity of 4230MW. The battery energy storage system uses lithium-ion batteries, with an installed capacity of 1290MWh. The water electrolysis hydrogen production unit employs proton exchange membrane (PEM) electrolyzer technology, with its oxygen outlet connected to the gasification unit and its hydrogen outlet connected to the methanol synthesis and distillation unit to supplement insufficient hydrogen, with a rated power of 2048MW. The hydrogen storage device has a hydrogen storage tank capacity of 418t to meet the hydrogen demand for 24 consecutive hours.

[0155] To improve the system's economy, low carbon footprint, and operational reliability under uncertain photovoltaic environments, a collaborative optimization scheduling method for a multi-source coupled photovoltaic-hydrogen-coal-storage methanol production system is implemented. The specific steps are as follows:

[0156] S1. Establish a scheduling optimization model for a multi-source coupled methanol production system based on photovoltaic, hydrogen, coal storage, and other sources, specifically:

[0157] S1.1. Mathematical modeling is performed on the general unit for coal-to-methanol production, the photovoltaic power generation unit, the water electrolysis hydrogen production unit, the electric energy storage unit, and the hydrogen storage unit;

[0158] S1.1.1. Establish a general unit mathematical model for coal-to-methanol production:

[0159] Considering the conservation of matter and energy, the general unit mathematical model for coal-to-methanol production is expressed as follows:

[0160] (1)

[0161] (2)

[0162] (3)

[0163] (4)

[0164] (5)

[0165] (6)

[0166] (7)

[0167] in, The processing flow rate of unit u; The raw material input flow rate for unit u; Output flow rate for unit u; , , This refers to the hydrogen, steam, and electricity requirements of unit u; The output ratio of the product in unit u; , , The demand coefficients for hydrogen, steam, and electricity in unit u; , The upper and lower limits of the flow rate processed by unit u; , The figures represent the methanol production of the methanol synthesis and distillation unit and the CO2-to-methanol unit, respectively. The target is the minimum methanol production.

[0168] S1.1.2. Establish the mathematical model of the photovoltaic power generation unit:

[0169] The output power of a photovoltaic power generation system is constrained by variables such as irradiance and temperature. The calculation formula is as follows:

[0170] (8)

[0171] in, The output power of a photovoltaic power generation system is expressed in kW. Represents global horizontal irradiance, W / m2 ; Indicates the installed capacity of photovoltaic power, in kW; GHI STC GHI, W / m³ under standard test conditions 2 ; Temperature coefficient, % / ℃; T c T represents the temperature of the photovoltaic module under actual conditions, in °C. STC The temperature of the photovoltaic module under standard test conditions is expressed in °C. This indicates the power supply efficiency of the photovoltaic power generation system.

[0172] (9)

[0173] in, Indicates ambient temperature, in °C; The standard operating temperature of the component is indicated in °C.

[0174] The operating constraints of photovoltaic power generation systems are as follows:

[0175] (10)

[0176] S1.1.3. Establish the mathematical model of the battery energy storage unit:

[0177] The energy variation relationship of battery energy storage in adjacent time periods is as follows:

[0178] (11)

[0179] in, Represents the remaining energy stored in electrical energy storage, in kWh; Indicates the self-discharge rate; The charging power of electrical energy storage is expressed in kW. Indicates the charging efficiency of electrical energy storage; The discharge power of the electrical energy storage is expressed in kW. This indicates the discharge efficiency of the electrical energy storage.

[0180] (12)

[0181] in, This indicates the installed capacity of the energy storage module, expressed in kWh.

[0182] The operational constraints of battery energy storage are as follows:

[0183] (13)

[0184] (14)

[0185] (15)

[0186] (16)

[0187] (17)

[0188] in, , These represent the maximum charging power and the maximum discharging power.

[0189] S1.1.4. Establish a mathematical model for the water electrolysis hydrogen production unit:

[0190] An electrolyzer produces hydrogen by consuming electrical energy to electrolyze water. The energy conversion formula is as follows:

[0191] (18)

[0192] in, Input electrical power to the electrolytic cell, kW; Hydrogen production, in kg; The value represents the electrolyzer conversion efficiency; LHV is the lower heating value of hydrogen, 33.3 kWh / kg.

[0193] The operating constraints of the electrolytic cell are as follows:

[0194] (19)

[0195] (20)

[0196] in, The installed capacity of the electrolytic cell is in kW.

[0197] S1.1.5. Establish a mathematical model for the hydrogen storage unit:

[0198] The relationship between the hydrogen storage capacity of the hydrogen storage system in adjacent time periods is as follows:

[0199] (twenty one)

[0200] in, The mass of hydrogen is expressed in kg. and These represent the hydrogen input and output rates of the hydrogen storage tank, respectively, in kg / h.

[0201] (twenty two)

[0202] in, This indicates the installed capacity of the hydrogen storage tank, expressed in kg.

[0203] The operating constraints of the hydrogen storage tank are as follows:

[0204] (twenty three)

[0205] (twenty four)

[0206] (25)

[0207] (26)

[0208] (27)

[0209] in, , The maximum input and output rates of hydrogen in the hydrogen storage tank are given in kg / h.

[0210] S1.2. Construct the objective function and constraints;

[0211] The objective function includes operating costs that measure economic efficiency, carbon emissions per ton of methanol that measure environmental impact, and photovoltaic curtailment rate that measures the efficiency of renewable energy utilization.

[0212] The formula for calculating the operating cost is as follows:

[0213] (28)

[0214] in, This represents the total operating cost. , , , The figures represent the following costs: photovoltaic power generation cost per kilowatt-hour (RMB / kWh); battery energy storage cost per kilowatt-hour (RMB / kWh); electrolyzer hydrogen production cost (RMB / kg); and hydrogen storage tank cost (RMB / kg).

[0215] The formula for calculating carbon emissions per ton of methanol is as follows:

[0216] (29)

[0217] in, The carbon emission factor per ton of methanol is expressed as tCO2e / t-MeOH. This represents the total direct carbon emissions, tCO2; tCO2e represents the total indirect carbon emissions. This represents the annual methanol production, expressed in tons (t).

[0218] The total direct carbon emissions The calculation formula is as follows:

[0219] (28)

[0220] in, The figure represents the carbon dioxide emissions (t) of the low-temperature methanol washing unit. This indicates the capture efficiency of the carbon capture system;

[0221] The total indirect carbon emissions The calculation formula is as follows:

[0222] (31)

[0223] in The term "carbon emissions implied by photovoltaic equipment" refers to the total CO2 emissions generated during the production, transportation, installation, and disposal of equipment such as photovoltaic panels, spread over the entire life cycle. The term "carbon emissions from other key equipment" refers to the CO2 emissions generated by major chemical equipment such as electrolyzers, hydrogen storage tanks, and reactors during their life cycle. This indicates the carbon emissions generated during the upstream production stage of raw materials.

[0224] The formula for calculating the photovoltaic curtailment rate is as follows:

[0225] (32)

[0226] in, This represents the light rejection rate at time t; This represents the photovoltaic output power at time t; This represents the charging power of the stored energy at time t; This represents the input power of the electrolytic cell at time t. This represents the electrical load of other equipment at time t.

[0227] The constraints include operational constraints of key equipment, reliability constraints of hydrogen supply, reliability constraints of power supply, charging and discharging power ramping constraints of battery energy storage system, and power ramping constraints of electrolyzer.

[0228] The hydrogen supply reliability constraints are as follows:

[0229] (33)

[0230] in, The hydrogen output rate of the electrolyzer is kg / h; and These represent the hydrogen input and output rates of the hydrogen storage tank, respectively, in kg / h; The hydrogen gas demand rate is kg / h.

[0231] The power supply reliability constraints are as follows:

[0232] (34)

[0233] in, The output power of a photovoltaic power generation system is expressed in kW. and The distribution represents the charging and discharging power of electrical energy storage, in kW; Input electrical power to the electrolytic cell, kW; The total electrical load of other equipment in the production system is expressed in kW.

[0234] The charging and discharging power ramp-up constraints of the battery energy storage system are as follows:

[0235] (35)

[0236] in, This indicates the net output power of the energy storage system; Indicates discharge. Indicates charging; Indicates the maximum power rise rate; Indicates the maximum power decrease rate;

[0237] The power ramp-up constraint of the electrolytic cell is as follows:

[0238] (36)

[0239] (37)

[0240] in, Indicates the operating power of the electrolytic cell; This indicates the maximum power ramp rate. Indicates the maximum power downhill ramp rate; and These represent the minimum and maximum technical output power of the electrolytic cell, respectively.

[0241] S2. Uncertainty modeling is performed using moment information constraint sets;

[0242] Considering the daily forecast error of photovoltaic power generation as the main source of uncertainty, a moment information constraint set is adopted. Describe the distribution of uncertainty:

[0243] (38)

[0244] in, for In the dataset The empirical distribution on, For low-dimensional uncertainty components, The number of components, The radius is the Wasserstein radius. and It is a first-order moment boundary.

[0245] S3. Establish a sequential decision-making framework and an adaptive dynamic optimization scheduling model;

[0246] Variables related to the coal-to-methanol process, namely These are considered current decisions, which are determined before the uncertainty materializes;

[0247] In contrast, the scheduling variables for electricity and hydrogen were formulated as sequential response decisions. This was achieved by employing decision rules. These scheduling variables are represented by these rules. These rules explicitly define the decision at each time step as a function of realized uncertainties (such as the actual output of wind and solar power). This framework allows the system to make adaptive, incremental scheduling based on the latest observed information, thus responding more flexibly and robustly to uncertainty.

[0248] The general form of the adaptive dynamic optimization scheduling problem is expressed as follows:

[0249]

[0250] st (39)

[0251]

[0252] in, and Indicates the decision-making stage; Includes stages Continuous and binary variables; , , , and It is a coefficient matrix or vector; Representation phase The uncertainty of history, its supporting set express; (That is, the current decision is certain.) This is the current decision; In the above formula, and In the set of moment information constraints The two are equivalent, both referring to the final uncertainty vector of renewable energy.

[0253] S4. An efficient solution method is proposed to solve this scheduling optimization problem, specifically:

[0254] S4.1. Use linear fractional decision rules to perform piecewise linear modeling of sequential decision variables;

[0255] For the decision variables in formula (39) , is a continuous variable and binary variables A linear fractional decision rule was established, and the formula is as follows:

[0256] (40)

[0257] (41)

[0258] in, ; and It is a continuous coefficient matrix; and It is a discrete coefficient matrix, whose elements are restricted to... Inside; , , , It is a constant matrix that satisfies and The second equation in equations (40) and (41) explains the decision rule from the perspective of the uncertainty partitioning of the proposed constraint set; this decision rule ensures unpredictability, that is, for all constraints... The realization of all Established;

[0259] After using the linear fractional decision rule, the adaptive dynamic optimization scheduling problem formula (39) can be expressed as follows:

[0260] (42)

[0261] in, It is the coefficient matrix or parameters obtained after the transformation.

[0262] S4.2. Transform the adaptive dynamic optimization problem into a mixed-integer linear programming problem;

[0263] First, constraint reconstruction was performed, and a constraint was constructed. The closed convex hull of the polygon is represented as a polyhedron defined by the left and right limits of the breakpoints:

[0264] (43)

[0265] in,

[0266] (44)

[0267] (45)

[0268] Next, using duality theory, the equation (42) involving... Constraints, namely Equivalently reconstructed as follows:

[0269] (46)

[0270] in, and It is a dual variable. Indicates the number of rows in the constraint;

[0271] Subsequently, the objective function in the scheduling problem (42) is reconstructed into an equivalent mixed-integer linear programming problem:

[0272] (47)

[0273] in It is a satisfaction and The constant auxiliary matrix;

[0274] By combining the objective function reconstruction (47) with the constraint reconstruction (46), the equivalent mixed integer linear programming reconstruction form of the adaptive dynamic optimization scheduling problem (39) can be obtained.

[0275] S4.3. Use the solver to solve the reconstructed mixed-integer linear programming problem to obtain the optimal scheduling strategy.

[0276] The photovoltaic-hydrogen-coal-storage multi-source coupled methanol production system and collaborative optimization scheduling method described in this invention achieves efficient and low-carbon operation under all operating conditions by collaboratively optimizing key decision variables such as electrolyzer power, battery charge / discharge power, and hydrogen / oxygen storage rates. Application results show that the system can adaptively adjust the hydrogen production rate of the electrolyzer based on fluctuations in photovoltaic output and flexibly schedule battery energy storage and hydrogen storage facilities for energy and hydrogen storage, thereby ensuring a stable supply of hydrogen required for the methanol synthesis unit. Under this optimized scheduling, the system's carbon emissions per ton of methanol are reduced by more than 80% compared to traditional coal-to-methanol methods, the comprehensive utilization rate of carbon resources is increased to over 87%, and the curtailment of photovoltaic power and production fluctuations caused by photovoltaic fluctuations are significantly reduced, ensuring the safety, stability, and efficiency of the continuous chemical production process.

[0277] Example 2:

[0278] This embodiment aims to verify the superiority of the proposed adaptive dynamic optimization method based on moment information constraint sets in the scheduling of coal-to-methanol systems. Building upon Embodiment 1, a comparative analysis with existing optimization methods demonstrates the advantages of this invention in addressing photovoltaic uncertainties. This embodiment primarily studies the following methods:

[0279] Method 1: Adaptive dynamic optimization method based on moment information constraint set;

[0280] Method 2: Using the multi-stage sub-Bruker optimization method of Wasserstein fuzzy sets;

[0281] Method 3: Multi-stage robust optimization;

[0282] In this embodiment, the breakpoints of the linear fractional decision rule are set to uniform intervals. The single-segment linear fractional decision rule is equivalent to the affine decision rule. Since the affine decision rule only applies to continuous variables, the binary decision variable is set as the first-stage variable in this case.

[0283] Results from 100 samples show that the linear fractional decision rule outperforms the affine decision rule in terms of performance, with the two-segment and six-segment linear fractional decision rules improving performance by approximately 3.1% and 3.9%, respectively. This indicates that the linear fractional decision rule, by introducing more flexible decision adaptability, can more effectively handle uncertainty. Although increasing the number of segments in the linear fractional decision rule leads to increased computation time, this is still within an acceptable range for scheduling optimization problems. Furthermore, Method 3 yields the most conservative results because it considers the worst-case scenario of uncertainty across the entire support set. When using linear fractional decision rules, Method 1 reduces methanol system operating costs by approximately 12.5% ​​and carbon emissions per ton of methanol by approximately 8.8% compared to Method 3. Moreover, compared to Method 2, Method 1 reduces operating costs by approximately 2.2% and carbon emissions per ton of methanol by approximately 2.7%. The results show that the proposed method has more advanced performance compared with the state-of-the-art robust optimization method, and proves that the method achieves an ideal trade-off between computation time and solution quality by flexibly adjusting the number of segments in the linear fractional decision rule.

[0284] In this embodiment, performance comparisons were also conducted under different sample sizes, where the number of segments in the linear fractional decision rule was fixed at 2, and the sample sizes were 25, 50, 100, 200, and 400. Each experiment involved repeated sampling and recalculation of the solution. The results show that the results of Method 3 are not affected by the sample size because it always considers the worst-case scenario within the entire support set. Furthermore, with the same sample size, the performance of Method 1 proposed in this invention is consistently superior to Method 2, especially when the sample size is small. This indicates that the moment information constraint set constructed in this invention can still effectively extract reliable distribution information from finite samples even when data is scarce, demonstrating stronger robustness and practicality.

[0285] Example 3:

[0286] This embodiment aims to analyze the economic and environmental impacts of different operating modes on the scheduling results of the coal-to-methanol system, in order to verify the advantages of the optimization method proposed in this invention in achieving coordinated system operation. Based on the same photovoltaic-hydrogen-coal-storage multi-source coupled methanol production system with complete units, the following three operating modes were designed for comparison through control command switching:

[0287] This embodiment aims to analyze the economic and environmental impacts of different system operation modes on the scheduling results of the coal-to-methanol system. By comparing three different system operation modes, the superiority of the complete system of this invention is verified.

[0288] Based on the aforementioned coal-to-methanol system, the following three configuration examples are designed:

[0289] Operating Mode 1: The system operates in a complete and coordinated manner. This means that the carbon capture unit, CO2-to-methanol unit, and water electrolysis unit are all put into operation according to the optimized results.

[0290] Operating Mode 2: The carbon capture and CO2-to-methanol unit is not in operation. In this mode, the CO2 removed by the low-temperature methanol washing unit is directly emitted, and the system produces methanol only through the coal-to-methanol unit.

[0291] Operating Mode 3: The water electrolysis unit is not operating. In this mode, the oxygen required by the system is supplied by the supporting conventional air separation unit, all hydrogen is purchased externally, and the carbon capture and CO2 to methanol production unit operates normally.

[0292] All operating modes were scheduled using the optimization method of this invention, under the same condition of ensuring an annual methanol production of 800,000 tons. The results are compared below:

[0293] From an economic perspective, Operation Mode 1 has the lowest total operating cost, approximately 3.6% lower than Operation Mode 2 and approximately 7.1% lower than Operation Mode 3. Operation Mode 2 suffers from higher operating costs due to the lack of CO2 resource utilization benefits and higher coal consumption. Operation Mode 3 is the most expensive due to the need for external hydrogen procurement and the high energy consumption of traditional air separation units.

[0294] In terms of environmental benefits, the carbon emissions per ton of methanol in Operation Mode 1 are only 0.35 tons, which is 70.3% and 44.4% lower than those in Operation Modes 2 and 3, respectively. Operation Mode 1 achieves a carbon capture rate of 95%, realizing efficient carbon resource utilization, while Operation Mode 2 has no carbon capture function, has the highest CO2 emissions, and the lowest carbon utilization rate.

[0295] Further analysis of the dynamic operating characteristics of each case revealed that Operation Mode 1 effectively mitigated the volatility of photovoltaic power generation through the flexible scheduling of the hydrogen and oxygen storage system. During peak photovoltaic power generation periods, the electrolyzer load rate reached over 95%, and excess hydrogen was stored in the hydrogen storage tank. During periods of insufficient photovoltaic output, the stored hydrogen was supplied to the CO2-to-methanol unit, ensuring its continuous and stable operation. Operation Mode 2, lacking carbon capture and CO2 utilization, had a relatively simple system operation, but resulted in low methanol production per ton of coal and high carbon emissions. Although Operation Mode 3 achieved carbon capture, the need for external hydrogen procurement and the high energy consumption of the air separation unit significantly increased the system's operating costs.

[0296] The above comparison demonstrates that the collaborative optimization scheduling method proposed in this invention is key to maximizing the economic and environmental benefits of the system. This method integrates independent units into a highly efficient and collaborative organic whole by dynamically coordinating photovoltaic output, water electrolysis unit load, energy storage status, and hydrogen storage status. Any operational mode that disrupts this synergy cannot achieve the optimal economic and environmental benefits achievable by this method.

[0297] The present invention has been described in detail above with reference to the embodiments, but the content described is only a preferred embodiment of the present invention and should not be considered as limiting the scope of the present invention. All equivalent changes and improvements made in accordance with the scope of the present invention should still fall within the patent coverage of the present invention.

Claims

1. A system for producing methanol by coupling a light hydrogen coal storage with a multi-source, and a method for collaborative optimization and scheduling, characterized in that, The methanol system comprises a coal gasification unit, a low-temperature methanol washing unit, a methanol synthesis and rectification unit, a photovoltaic power generation unit, a water electrolysis hydrogen production unit, a carbon capture unit, a CO2 methanol production unit, an electric energy storage unit and a hydrogen storage unit; the system is organically coupled with each unit through material flow and energy flow, wherein: The photovoltaic power generation unit is the only power source of the system, and the output end thereof is connected with the power input end of each unit; the water electrolysis hydrogen production unit adopts a proton exchange membrane electrolytic cell technology, the oxygen outlet thereof is connected with the gasifier inlet of the coal gasification unit, and the hydrogen outlet thereof is connected with the inlet of the CO2 methanol production unit; the coal gasification unit, the water gas shift unit and the low-temperature methanol washing unit are sequentially connected, for producing synthesis gas and removing acid gas; the CO2-rich tail gas outlet of the low-temperature methanol washing unit is connected with the inlet of the carbon capture unit; the carbon capture unit adopts a membrane separation method, and the CO2 product outlet thereof is connected with the inlet of the CO2 methanol production unit; the hydrogen storage unit is connected between the hydrogen outlet of the water electrolysis hydrogen production unit and a downstream gas unit; the battery energy storage system is connected with the photovoltaic power generation unit; The collaborative optimization scheduling method comprises the following steps: S1. establishing a scheduling optimization model of the photovoltaic-hydrogen-coal storage multi-source coupled methanol production system, specifically: S1.

1. mathematically modeling the coal-to-methanol general unit, the photovoltaic power generation unit, the water electrolysis hydrogen production unit, the electric energy storage unit and the hydrogen storage unit; S1.

2. constructing an objective function and a constraint condition; S2. modeling uncertainty by using a set of matrix information constraints; S3. establishing a sequential decision-making framework and an adaptive dynamic scheduling model; S4. proposing an efficient solving method to solve the scheduling optimization problem, specifically: S4.

1. using a linear fractional decision rule to segmentally linearly model the sequential decision variables; S4.

2. converting the adaptive dynamic scheduling optimization problem into a mixed integer linear programming problem; S4.

3. using a solver to solve the reconstructed mixed integer linear programming problem to obtain an optimal scheduling strategy; S4.

3. using a solver to solve the reconstructed mixed integer linear programming problem to obtain an optimal scheduling strategy.

2. The synergistic optimization scheduling method of the photo-hydro-coal multi-source coupled methanol system according to claim 1, characterized in that: The objective function in S1.2 comprises an operation cost for measuring economy, a ton-methanol carbon emission for measuring environmental impact and a photovoltaic light abandonment rate for measuring renewable energy utilization efficiency; the constraint condition comprises an operation constraint of a key device, a hydrogen supply reliability constraint, an electric power supply reliability constraint, a charge-discharge power ramping constraint of the battery energy storage system and a power ramping constraint of the electrolytic cell; The calculation formula of the operation cost is as follows: (1)； wherein, represents the total operating cost, , , , represents the photovoltaic electricity cost, yuan / kWh; the battery energy storage electricity cost, yuan / kWh; the electrolytic tank hydrogen production cost, yuan / kg; the hydrogen storage tank hydrogen storage cost, yuan / kg; The calculation formula of the ton-methanol carbon emission is as follows: (2)； wherein, represents the ton methanol carbon emission factor, tCO2e / t-MeOH; represents the total direct carbon emissions, tCO2; represents the total indirect carbon emissions, tCO2e; represents the methanol annual production, t; The total amount of indirect carbon emissions The calculation formula is as follows: (3)； wherein represents the carbon footprint of the photovoltaic equipment, which refers to the total CO2 emissions generated in the production, transportation, installation and disposal processes of the photovoltaic panel and other equipment, and is allocated to the entire life cycle; represents the carbon footprint of other key equipment, which refers to the CO2 emissions generated in the life cycle of main chemical equipment such as electrolytic cell, hydrogen storage tank and reaction furnace; represents the carbon emissions generated in the upstream production stage of raw materials; The calculation formula of the photovoltaic light abandonment rate is as follows: (4)； wherein, represents the light rejection rate at time t; represents the photovoltaic output power at time t; represents the charging power of the electrical energy storage at time t; represents the input power of the electrolyzer at time t, represents the electrical load of other devices at time t; The hydrogen supply reliability constraint is as follows: (5)； wherein, is the hydrogen output rate of the electrolyzer, kg / h; and are the hydrogen input and output rates of the hydrogen storage tank, respectively, kg / h; is the hydrogen demand rate, kg / h; The electric power supply reliability constraint is as follows: (6)； wherein, Pout represents the output power of the photovoltaic power generation system, kW; and Pdis represents the charge-discharge power of the distributed electric energy storage, kW; Pelec represents the input electric power of the electrolyzer, kW; Pother represents the total electric load of other equipment of the production system, kW; The charge-discharge power ramping constraint of the battery energy storage system is as follows: (7)； wherein, represents the net output power of the energy storage system; represents discharging, represents charging; represents the maximum power rise rate; represents the maximum power fall rate; The power ramping constraint of the electrolytic cell is as follows: (8)； (9)； wherein, represents the operating power of the electrolytic cell; represents the maximum power ramp-up rate; represents the maximum power ramp-down rate; and respectively represent the minimum and maximum technical output power of the electrolytic cell.

3. The synergistic optimization scheduling method of the photo-hydro-coal multi-source coupled methanol system according to claim 1, characterized in that: In the S2, considering the daily prediction error of photovoltaic power generation as the main source of uncertainty, the set of information constraints is adopted Describe the uncertainty distribution: (10)； where, is the empirical distribution on the dataset is the low-dimensional uncertainty component, is the number of components, is the Wasserstein radius, and is the first-moment bound.

4. The synergistic optimization scheduling method of the photo-hydro-coal multi-source coupled methanol system according to claim 1, characterized in that: In S3, the sequential decision-making framework is as follows: Variables related to the coal-to-methanol process, i.e. are considered current decisions, which are determined before uncertainty realization. In contrast, the scheduling variables for electricity and hydrogen are formulated as sequential response decisions; these scheduling variables are represented by employing decision rules that explicitly define the decisions at each time step as a function of realized uncertainty.

5. The synergistic optimization scheduling method of the photo-hydro-coal multi-source coupled methanol system according to claim 1, characterized in that: In S3, the general form of the adaptive dynamic optimization scheduling problem is as follows: ； where and denote the decision stage; contains the continuous and binary variables in stage ; , , , and are coefficient matrices or vectors; denotes the uncertainty history in stage , whose support set is denoted by ; , is the current decision; and in the above equation are equivalent in the set of information constraints , both refer to the final renewable energy uncertainty vector.

6. The synergic optimization scheduling method of the photo-hydro-coal multi-source coupling system for methanol production according to claim 1, characterized in that: In S4.1, the specific process of segmentally linearly modeling the sequential decision variables by using a linear fractional decision rule is as follows: For the decision variables in equation (39) are continuous variables and binary variables Linear fractional decision rules are established and are given by the following equations: (12)； (13)； where ; and are continuous coefficient matrices; and are discrete coefficient matrices whose elements are restricted to ; , , , are constant matrices satisfying and ; the second equality in (12) and (13) explains the decision rule from the perspective of uncertainty partition of the proposed constraint set; this decision rule ensures the nonanticipativity, i.e., for all realizations satisfying , there holds ; After using the linear piecewise decision rule, the adaptive dynamic optimization scheduling problem formula (11) can be expressed as follows: (14)； wherein is the transformed coefficient matrix or parameter.

7. The synergistic optimization scheduling method of the photo-hydro-coal multi-source coupled methanol system according to claim 1, characterized in that: In S4.2, the specific process of converting the adaptive dynamic optimization scheduling problem into a mixed integer linear programming problem is as follows: First, the constraint reconstruction is performed, and the closed convex hull of is constructed and represented as a polyhedron defined by the left and right limits of the breakpoints: (15)； Wherein, (16)； (17)； Next, by using the duality theory, the constraint involved in formula (14), i.e. , is equivalently reconfigured as the following formula: ​ (18)； wherein and are dual variables, denotes the number of constraints. Subsequently, the objective function in the scheduling problem (14) is reconstructed into an equivalent mixed integer linear programming problem as follows: (19)； wherein is a constant auxiliary matrix satisfying and ​ By combining the objective function reconstruction formula (19) with the constraint reconstruction formula (18), the equivalent mixed integer linear programming reconstruction form of the multi-stage distribution robust optimization scheduling problem (11) can be obtained.

Images