Industrial micro-grid group double-layer optimization scheduling method for electricity-hydrogen shared energy storage

Abstract

The invention discloses an industrial micro-grid group double-layer optimization scheduling method for electricity-hydrogen shared energy storage, and aims to make full use of energy consumption complementation characteristics of different parks. A refinement model of various devices, electricity-hydrogen shared energy storage components and the like in the polycrystalline silicon micro-grid is constructed by considering a hydrogen recovery system. Based on a Stackelberg game theory, a double-layer optimization scheduling model with shared energy storage as a leader and a plurality of polycrystalline silicon micro-grids as followers is constructed. The upper layer model is an EHSES multi-target optimization scheduling model considering economy, environmental protection and sustainability, the lower layer model is an economic scheduling model of each polycrystalline silicon microgrid, and iterative optimization of the electro-hydrogen price of the EHSES and the polycrystalline silicon microgrid is realized. And solving the model by using a multi-target myxomycete algorithm based on reverse learning and a reference point and a CPLEX solver. Simulation results show that the model can effectively improve the economy, environmental protection and sustainability of the polycrystalline silicon micro-grid cluster, and the effectiveness of the model is verified.

Description

Technical Field

[0001] This invention relates to the field of two-layer optimal scheduling of industrial microgrid groups, and particularly to a two-layer optimal scheduling method for industrial microgrid groups with shared electricity and hydrogen energy storage. Background Technology

[0002] Building a new power system based on new energy sources on the power supply side is a new challenge facing industry under the "dual carbon" environment. Currently, many energy-intensive enterprises have built their own coal-fired power plants to reduce production costs. Among them, polysilicon enterprises have a huge demand for hydrogen, and the traditional method of using self-owned power plants to produce hydrogen results in serious carbon emissions. Therefore, providing new energy power generation for traditional polysilicon enterprises and realizing the coordinated, economical, and efficient operation of multiple microgrids is of positive significance.

[0003] Energy storage sharing is an emerging operating model that fully leverages the economies of scale of energy storage. It takes into account residential customers installing solar photovoltaic (PV) panels and integrated energy storage.

[0004] To address the optimal scheduling problem of shared energy storage, existing technologies propose a two-stage energy sharing model that considers the uncertainties of photovoltaic (PV) processing to minimize the costs for both PV producers and consumers and shared energy storage. To cope with the challenges posed by renewable energy fluctuations, a coordinated optimal scheduling model for shared energy storage and multiple microgrids is proposed to minimize costs. Considering optimal economics, a strategy for wind power participation in day-ahead real-time market bidding based on shared energy storage is proposed. These studies provide good ideas for the optimal scheduling problem of shared energy storage, but they only consider the system's economics and neglect the optimization of the system's environmental impact. Other existing technologies propose optimal scheduling models for multi-regional energy systems that consider economics, environmental protection, and shared energy storage, aiming to minimize operating costs and maximize net environmental impact. Alternatively, a multi-constraint bi-objective objective function model has been designed to achieve optimal scheduling of shared energy storage in microgrids. This partially considers economics and environmental protection but lacks optimization for system sustainability. Furthermore, the Pareto boundary can be used to address the trade-offs between different objectives.

[0005] Regarding the pricing mechanism for shared energy storage, some existing technologies propose a cost-sharing mechanism based on cooperative game theory. This method requires precise calculation of user usage of shared energy storage, which is computationally complex. Others propose a comprehensive energy optimization and scheduling strategy that includes shared energy storage, proposing a dynamic leasing pricing strategy based on demand ratio. This strategy dynamically adjusts the service price of shared energy storage according to market supply and demand, but this mechanism is easily affected by market fluctuations and can impose a certain economic burden on users. Still others introduce a combined auction mechanism to realize transactions between users and shared energy storage, determining the price of energy storage services through auctions. However, this method requires advanced technologies such as blockchain, making implementation complex. Some have proposed a contract pricing mechanism for shared energy storage, requiring users and energy storage operators to sign long-term contracts, which lacks flexibility and cannot adapt to market changes. Others have proposed a novel energy sharing framework considering hydrogen trading, constructing a two-tier pricing mechanism based on Stackelberg game theory, treating shared energy storage as the leader and users as followers, and verifying the effectiveness of this pricing mechanism. Therefore, pricing methods based on Stackelberg game theory deserve further research.

[0006] To address the research gaps mentioned above, this invention introduces Electricity-Hydrogen Shared Energy Storage (EHSES) into polycrystalline silicon industrial microgrids (PMs), and proposes a two-layer optimal scheduling model based on Stackelberg game theory. This model achieves multi-objective optimal scheduling of EHSES and energy sharing of PMs. The main contributions are as follows: (1) Considering the high energy consumption of PMs, an Electricity-Hydrogen Shared Energy Storage and hydrogen energy recovery device are introduced to realize economic energy management of PMs; (2) By introducing the self-reliance factor index, an optimal scheduling model for shared energy storage that considers economic efficiency, environmental protection, and sustainability is proposed; (3) By treating EHSES as the leader and PMs as the followers, a novel two-layer optimal scheduling model is constructed using Stackelberg game theory. Summary of the Invention

[0007] To address the problems existing in the prior art, the present invention aims to provide a two-layer optimized scheduling method for industrial microgrid groups with shared electric-hydrogen energy storage.

[0008] To achieve the above objectives, the technical solution of the present invention is as follows:

[0009] A two-layer optimal scheduling method for industrial microgrid groups with shared electricity and hydrogen energy storage consists of the following steps:

[0010] Step 1: System Architecture and Component Model

[0011] Step 1.1 System Architecture

[0012] The Siemens process is often used in the production of polysilicon, which consumes a large amount of hydrogen. Traditional polysilicon companies mainly use a power supply model of "self-owned power plant + purchased electricity" to produce hydrogen, which will generate a large amount of carbon emissions.

[0013] To enhance the low-carbon nature, reliability, and sustainability of PMs, new energy sources are used to replace self-owned power plants for hydrogen production, and an EHSES (Environmentally Friendly Hybrid System) is introduced to achieve coordinated and efficient operation of multiple PMs. Each PM contains both electricity and hydrogen energy, primarily including wind turbines, photovoltaic modules, a self-owned power plant, electricity-to-hydrogen equipment, hydrogen recovery equipment, and polysilicon production equipment. Multiple PMs are connected to the EHSES via hydrogen pipelines and cables, enabling indirect electricity-hydrogen interaction and sharing between multiple microgrids. The EHSES includes operators, lithium iron phosphate batteries, and hydrogen storage tanks. Driven by internal pricing incentives, multiple microgrids are more inclined to engage in electricity-hydrogen trading with the EHSES. The operator acts as both an agent for electricity-hydrogen trading between multiple microgrids and the EHSES, and is also responsible for electricity trading between the EHSES and external power grids and hydrogen refueling stations, serving as the EHSES's dispatch center. Step 1.2 Component Model

[0014] Step 2: Two-level optimization scheduling model based on Stackelberg game theory

[0015] Step 2.1 Decision-making process

[0016] The upper-level EHSES sets the internal electricity price, internal hydrogen price, and EHSES charging / discharging power and hydrogen charging / discharging rate curves. The lower-level PMi optimizes its own energy consumption and energy trading strategies. The energy trading strategies of the lower-level PMi are uploaded to the upper-level EHSES. After calculating its revenue, the EHSES redefines the internal price accordingly until the optimal internal price is found. Furthermore, the decisions of both parties are inherently sequential, which is consistent with the dynamic game scenario of the Stackelberg hierarchy. Therefore, the energy trading process can be viewed as a Stackelberg game model.

[0017]

[0018] In the formula, and These represent the prices at which EHSES buys and sells electrical power from PMs, respectively. and These represent the prices at which EHSES buys and sells HV from PMs, respectively. and These represent the electrical power that EHSES purchases from PMs and the electrical power that it sells, respectively. and These represent the HV purchased and sold by EHSES from PMs, respectively; L eco L env L srf and L i Let represent the economic objective function of EHSES, the environmental objective function of EHSES, the sustainability objective of EHSES, and the objective function of PM i, respectively; where EHSES is the leader in the upper-level model and PMs are the followers in the lower-level model. and These represent the leader's strategy set and the follower's strategy set, respectively.

[0019] Step 2.2 Upper-level EHSES multi-objective optimization scheduling model

[0020] Step 2.2.1 Objective Function

[0021] The economic objective of EHSES's multi-objective optimization scheduling is to maximize the daily profit of EHSES, the environmental objective is to minimize the system's carbon emissions, and the sustainability objective is to maximize the system's self-reliance factor.

[0022] Step 2.2.2 Constraints

[0023] (1) Power balance constraint

[0024]

[0025] (2) Hydrogen energy balance constraints

[0026]

[0027] (3) Electricity trading constraints

[0028]

[0029] In the formula, This is a binary variable that indicates that EHSES purchases electricity from the grid; a value of 1 indicates that electricity is purchased from the grid. This is a binary variable that indicates that PM i purchases electricity from EHSES; a value of 1 indicates that electricity is purchased from EHSES.

[0030] (4) Constraints on hydrogen energy trading

[0031]

[0032] In the formula, This is a binary variable that indicates that EHSES purchases hydrogen from a hydrogen refueling station; a value of 1 indicates that hydrogen is purchased from a hydrogen refueling station. This is a binary variable that indicates that PMi purchases hydrogen from EHSES; a value of 1 indicates that hydrogen is purchased from EHSES.

[0033] Step 2.3 Lower-layer multi-micronet optimization scheduling model

[0034] Step 2.3.1 Objective Function

[0035] Each microgrid optimizes its scheduling with the goal of minimizing its own operating cost. The objective function can be expressed as:

[0036]

[0037] Step 2.3.2 Constraints

[0038] (1) Power balance constraint

[0039]

[0040] In the formula, Indicates wind power output; Indicates photovoltaic power; Indicates conventional electrical load;

[0041] (2) Hydrogen energy balance constraints

[0042]

[0043] In the formula, This indicates the conventional hydrogen loading in PMi;

[0044] (3) Electricity trading constraints

[0045]

[0046] (4) Constraints on hydrogen energy trading

[0047]

[0048] The constraints of other equipment will not be elaborated here;

[0049] Step 3 Solution Method

[0050] Two-level optimization is an effective method for solving the two-level optimization scheduling problem based on Stackelberg game. Therefore, this invention uses a two-level optimization algorithm that combines an improved multi-objective slime mold algorithm based on back learning and reference points with a CPLEX solver to solve the above model.

[0051] Furthermore, the component model in step 1.2 specifically includes:

[0052] Step 1.2.1 Model of the electro-hydrogen production equipment

[0053] Because alkaline electrolyzers are low-cost, highly commercialized, and more suitable for large-scale hydrogen production, they are chosen as the electrochemical hydrogen production equipment. Their model can be represented as follows:

[0054]

[0055] In the formula, The rate at which hydrogen gas is produced by EL; The electrical power consumed by the EL; For the efficiency of EL; The high calorific value of hydrogen; This refers to the rated electrical power of EL.

[0056] Step 1.2.2 Hydrogen recovery equipment model

[0057] This invention introduces a hydrogen recovery device to recover and reuse hydrogen during the polysilicon production process, thereby improving the utilization rate of hydrogen. The mathematical model is as follows:

[0058]

[0059] In the formula, w re η is the energy consumption coefficient for hydrogen recovery. re Hydrogen recovery efficiency; φ re The proportion of recyclable hydrogen; Let t be the hydrogen supply rate of the system at time t; The density of hydrogen gas;

[0060] Step 1.2.3 Electric-hydrogen shared energy storage model.

[0061] Furthermore, step 1.2.3, the electric-hydrogen shared energy storage model, specifically comprises:

[0062] (1) Energy storage model

[0063] Lithium iron phosphate batteries were selected as SEES;

[0064]

[0065] In the formula, and ξ represents the amount of electricity stored in SEES at time t-1 and time t, respectively; SD This represents the energy dissipation rate of the SEES. In this invention, the self-discharge phenomenon of the SEES is ignored, so its value is 0; P t BC and P t BD Let represent the charging power and discharging power of SEES at time t, respectively; and Let these represent the charging efficiency and discharging efficiency of SEES, respectively. To simplify the model, this invention ignores the effects of temperature and depth of discharge on the charging and discharging efficiency. and Set to a constant value; Δt is the time step; This is the discharge flag for SEES; a value of 1 indicates charging, and a value of 0 indicates discharging. and These are the rated capacity and rated power of SEES, respectively; This is the initial capacity value for SEES; and These are the minimum and maximum values ​​of the SOC of SEES, respectively;

[0066] (2) Hydrogen storage tank model

[0067]

[0068] In the formula, and These represent the hydrogen rate at which the HST is charged and the hydrogen rate at which it is discharged (Hydrogenvelocity, HV), respectively. and These represent the minimum, maximum, and initial pressure values ​​in the HST, respectively; R0 represents the gas constant; T HS Indicates the internal temperature of the HST; V HS Indicates the volume of HST; Indicates the molar mass of hydrogen; This is a flag indicating whether HST is being charged or discharged with hydrogen; a value of 1 indicates that it is being charged with hydrogen.

[0069] Furthermore, in the objective function of step 2.2.1, the decision variables include the electricity price, hydrogen price, and charging / discharging plan of EHSES, as shown in the following formula:

[0070] (1) Economic objective: Maximize daily revenue of EHSES

[0071]

[0072] In the formula, On-grid electricity price; EHSES's power output sold to the grid; The price at which electricity is purchased from the grid; Power purchased from the grid for EHSES; The price at which hydrogen is sold to hydrogen refueling stations; HV sold by SES to hydrogen refueling stations; The price of purchasing hydrogen from a hydrogen refueling station; HV purchased by EHSES from hydrogen refueling stations; r is the rate of return on capital; y sδ represents the lifespan of the EHSES; δ represents the number of days it runs in a year; T represents the simulation cycle; C p C is the unit price of battery power; e Price per unit of battery capacity; C HS This refers to the unit price of the hydrogen storage tank; ζ ope For operation and maintenance coefficient; ζ re This is the residual value recovery factor;

[0073] (2) Environmental protection objective: Minimize system carbon emissions

[0074]

[0075] In the formula, τ e Carbon emission factor for traditional power generation units; η e Carbon allowance coefficient for traditional power generation units; The power output of the self-owned power plant; η re Carbon quota coefficient for renewable energy;

[0076] (3) Sustainability Goal: Maximizing the System's Self-Reliance Factor

[0077]

[0078] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0079] This invention addresses the energy supply characteristics and low-carbon requirements of polycrystalline silicon microgrids by introducing shared energy storage for PMs and constructing a two-layer optimal scheduling model based on Stackelberg game theory. The following conclusions were obtained through policy verification: (1) The introduction of shared energy storage and hydrogen recovery equipment reduced the daily operating cost of PMs by 61.17%, effectively improving the economic efficiency of PMs; (2) The proposed model reduced carbon emissions by 2856.32 kg and increased the self-reliance factor to 98.21%, effectively improving the environmental friendliness and sustainability of the system. Attached Figure Description

[0080] Figure 1 Polysilicon production process;

[0081] Figure 2 System architecture diagram;

[0082] Figure 3 Typical daily data from a polycrystalline silicon microgrid in Xinjiang: Figure 3 (a) Electrical load; Figure 3 (b) Hydrogen loading; Figure 3 (c) Wind power output; Figure 3 (d) Photovoltaic power; Figure 3 (e) Price data;

[0083] Figure 4The electrical balance and hydrogen balance in Case 3: Figure 4 (a) Electrical balance diagram of PM1; Figure 4 (b) Electrical balance diagram of EHSES; Figure 4 (c) Hydrogen balance diagram of PM1; Figure 4 (d) Photovoltaic power;

[0084] Figure 5 EHSES optimization results: Figure 5 (a) State of charge and charge / discharge power of SEES; Figure 5 (b) The pressure and hydrogen charge / discharge rate of the SHES;

[0085] Figure 6 Optimal internal electricity price and hydrogen price: Figure 6 (a) The optimal internal electricity price; Figure 6 (b) Optimal internal hydrogen valence. Detailed Implementation

[0086] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments:

[0087] like Figure 1-6 As shown,

[0088] A two-layer optimal scheduling method for industrial microgrid groups with shared electricity and hydrogen energy storage consists of the following steps:

[0089] 1. System architecture and component model

[0090] 1.1 System Architecture

[0091] In the production of polycrystalline silicon, the Siemens process is often used, and its production process is as follows: Figure 1 As shown, the production process will consume a large amount of hydrogen. Traditional polysilicon companies mainly use the power supply model of "self-owned power plant + purchased electricity" to produce hydrogen, which will generate a large amount of carbon emissions.

[0092] To enhance the low-carbon nature, reliability, and sustainability of PMs (Power Generation Systems), new energy sources are used to replace self-owned power plants for hydrogen production, and an EHSES (Environmentally Friendly Hydrogen Storage System) is introduced to achieve coordinated and efficient operation of multiple PMs. A single PM contains both electricity and hydrogen energy, primarily including wind turbines, photovoltaic modules, self-owned power plants, hydrogen production equipment, hydrogen recovery equipment, and polysilicon production equipment. Multiple PMs are connected to the EHSES via hydrogen pipelines and cables, enabling indirect interaction and sharing of electricity and hydrogen between multiple microgrids. The EHSES includes operators, lithium iron phosphate batteries, and hydrogen storage tanks (HSTs). Driven by internal pricing incentives, microgrids are more inclined to engage in electricity-hydrogen trading with the EHSES. The operator acts as both an agent for electricity-hydrogen trading between the microgrid and the EHSES and is responsible for electricity trading between the EHSES and external power grids and hydrogen refueling stations. It serves as the dispatch center for the EHSES. The system architecture diagram is shown below. Figure 2 As shown:

[0093] 1.2 Component Model

[0094] 1.2.1 Model of an electro-hydrogen production equipment

[0095] Because alkaline electrolyzers are low-cost, highly commercialized, and more suitable for large-scale hydrogen production, they are chosen as the electrochemical hydrogen production equipment. Their model can be represented as follows:

[0096]

[0097] In the formula, The rate at which hydrogen gas is produced by EL; The electrical power consumed by the EL; For the efficiency of EL; The high calorific value of hydrogen; This is the rated power of EL.

[0098] 1.2.2 Hydrogen recovery equipment model

[0099] This invention introduces a hydrogen recovery device to recover and reuse hydrogen during the polysilicon production process, thereby improving the utilization rate of hydrogen. The mathematical model is as follows:

[0100]

[0101] In the formula, w re η is the energy consumption coefficient for hydrogen recovery. re Hydrogen recovery efficiency; φ re The proportion of recyclable hydrogen; Let t be the hydrogen supply rate of the system at time t; The density of hydrogen gas is given.

[0102] 1.2.3 Electricity-Hydrogen Shared Energy Storage Model

[0103] (1) Energy storage model

[0104] Lithium iron phosphate batteries were selected as SEES.

[0105]

[0106] In the formula, and ξ represents the amount of electricity stored in SEES at time t-1 and time t, respectively; SD This represents the energy dissipation rate of the SEES. In this invention, the self-discharge phenomenon of the SEES is ignored, so its value is 0; P t BC and P t BD Let represent the charging power and discharging power of SEES at time t, respectively; and Let these represent the charging efficiency and discharging efficiency of SEES, respectively. To simplify the model, this invention ignores the effects of temperature and depth of discharge on the charging and discharging efficiency. and Set to a constant value; Δt is the time step; This is the discharge flag for SEES; a value of 1 indicates charging, and a value of 0 indicates discharging. and These are the rated capacity and rated power of SEES, respectively; This is the initial capacity value for SEES; and These are the minimum and maximum values ​​of the SOC of SEES, respectively.

[0107] (2) Hydrogen storage tank model

[0108]

[0109] In the formula, and These represent the hydrogen rate at which the HST is charged and the hydrogen rate at which it is discharged (Hydrogenvelocity, HV), respectively. and These represent the minimum, maximum, and initial pressure values ​​in the HST, respectively; R0 represents the gas constant; T HS Indicates the internal temperature of the HST; V HS Indicates the volume of HST; Indicates the molar mass of hydrogen; This is a flag indicating whether HST is being charged or discharged with hydrogen; a value of 1 indicates that it is being charged with hydrogen.

[0110] 2. A Two-Level Optimization Scheduling Model Based on Stackelberg Game Theory

[0111] 2.1 Decision-making process

[0112] The upper-level EHSES sets the internal electricity price, internal hydrogen price, and EHSES charging / discharging power and hydrogen charging / discharging rate curves. The lower-level PMi optimizes its own energy consumption and energy trading strategies. The energy trading strategies of the lower-level PMi are uploaded to the upper-level EHSES. After calculating its revenue, the EHSES redefines the internal price accordingly until the optimal internal price is found. Furthermore, the decision-making of both parties is inherently sequential, which is consistent with the dynamic game scenario of the Stackelberg hierarchy. Therefore, the energy trading process can be viewed as a Stackelberg game model:

[0113]

[0114] In the formula, and These represent the prices at which EHSES buys and sells electrical power from PMs, respectively. and These represent the prices at which EHSES buys and sells HV from PMs, respectively. and These represent the electrical power that EHSES purchases from PMs and the electrical power that it sells, respectively. and These represent the HV purchased and sold by EHSES from PMs, respectively; L eco L env L srf and L i Let represent the economic objective function, environmental objective function, sustainability objective, and PM i objective function of EHSES, respectively. In this model, EHSES is the leader, and PMs are the followers in the lower-level model. and These represent the leader's strategy set and the follower's strategy set, respectively.

[0115] 2.2 Upper-level EHSES multi-objective optimization scheduling model

[0116] 2.2.1 Objective Function

[0117] The economic objectives of EHSES's multi-objective optimal scheduling are maximizing daily profits, minimizing system carbon emissions, and maximizing the system's self-sufficiency factor. Its decision variables mainly include the electricity price, hydrogen price, and charging / discharging schedule of EHSES, as shown in the following equation:

[0118] (1) Economic objective: Maximize daily revenue of EHSES

[0119]

[0120] In the formula, On-grid electricity price; EHSES's power output sold to the grid; The price at which electricity is purchased from the grid; Power purchased from the grid for EHSES; The price at which hydrogen is sold to hydrogen refueling stations; HV sold by SES to hydrogen refueling stations; The price of purchasing hydrogen from a hydrogen refueling station; HV purchased by EHSES from hydrogen refueling stations; r is the rate of return on capital; y s δ represents the lifespan of the EHSES; δ represents the number of days it runs in a year; T represents the simulation cycle; C p C is the unit price of battery power; e Price per unit of battery capacity; C HS This refers to the unit price of the hydrogen storage tank; ζ ope For operation and maintenance coefficient; ζ re This is the residual value recovery factor.

[0121] (2) Environmental protection objective: Minimize system carbon emissions

[0122]

[0123] In the formula, τ e Carbon emission factor for traditional power generation units; η e Carbon allowance coefficient for traditional power generation units; The power output of the self-owned power plant; η re This is the carbon quota factor for renewable energy.

[0124] (3) Sustainability Goal: Maximizing the System's Self-Reliance Factor

[0125]

[0126] 2.2.2 Constraints

[0127] (1) Power balance constraint

[0128]

[0129] (2) Hydrogen energy balance constraints

[0130]

[0131] (3) Electricity trading constraints

[0132]

[0133] In the formula, This is a binary variable that indicates that EHSES purchases electricity from the grid; a value of 1 indicates that electricity is purchased from the grid. The flag indicating whether PM i purchases electricity from EHSES is a binary variable; a value of 1 indicates that electricity is purchased from EHSES.

[0134] (4) Constraints on hydrogen energy trading

[0135]

[0136] In the formula, This is a binary variable that indicates that EHSES purchases hydrogen from a hydrogen refueling station; a value of 1 indicates that hydrogen is purchased from a hydrogen refueling station. This is a binary variable that indicates whether PMi purchases hydrogen from EHSES. A value of 1 indicates that hydrogen is purchased from EHSES.

[0137] 2.3 Lower-layer multi-micronet optimized scheduling model

[0138] 2.3.1 Objective Function

[0139] Each microgrid optimizes its scheduling with the goal of minimizing its own operating cost. The objective function can be expressed as:

[0140]

[0141] 2.3.2 Constraints

[0142] (1) Power balance constraint

[0143]

[0144] In the formula, Indicates wind power output; Indicates photovoltaic power; This indicates a conventional electrical load.

[0145] (2) Hydrogen energy balance constraints

[0146]

[0147] In the formula, This indicates the conventional hydrogen load in PM i.

[0148] (3) Electricity trading constraints

[0149]

[0150] (4) Constraints on hydrogen energy trading

[0151]

[0152] The constraints of the remaining equipment will not be elaborated here.

[0153] 3. Solution Method

[0154] Two-level optimization is an effective method for solving the two-level optimization scheduling problem based on Stackelberg game. Therefore, this invention uses a two-level optimization algorithm that combines an improved multi-objective slime mold algorithm based on back learning and reference points with a CPLEX solver to solve the above model.

[0155] 4. Case Analysis

[0156] To verify the effectiveness of the proposed model, typical daily data from a polycrystalline silicon microgrid in Xinjiang was selected for simulation in this invention. Conventional electrical load, hydrogen load, wind power, photovoltaic power, and price data are as follows: Figure 3 As shown. The simulation period is 24 hours, the simulation step size is 1 hour, and the remaining parameters are shown in Table 1:

[0157] Table 1 System Parameter Settings

[0158]

[0159] To fully analyze the effectiveness of the proposed model, three comparative cases were designed as shown in Table 2.

[0160] Table 2 Comparison Case Settings

[0161]

[0162] 4.1 Analysis of Scheduling Results

[0163] To verify the effectiveness of the proposed model, the optimization scheduling results for the three cases are shown in Table 3. Case 1 serves as the baseline scenario, with daily operating costs of PMs reaching RMB 2723.90, RMB 2638.74, and RMB 2943.94 respectively, significantly higher than the scheduling results of Cases 2 and 3. Compared to Case 1, the total daily operating cost of PMs in Case 2 is reduced by 39.43%, and the total daily operating cost of PMs in Case 3 is reduced by 61.17%, indicating that the addition of the electric-hydrogen shared energy storage and hydrogen recovery device can significantly improve the economics of PMs. The daily revenue of SES in Case 3 reaches RMB 2597.98, indicating that the proposed model can achieve a win-win situation for SES and PMs. The carbon emissions in Case 3 are 4493.6 kg less than in Case 1 and 2856.32 kg less than in Case 2, demonstrating that the proposed model can effectively improve the environmental friendliness of the system. The self-sufficiency factor in Case 3 is as high as 98.21%, exhibiting a high degree of self-sufficiency, indicating that the proposed model can effectively improve the sustainability of the system.

[0164] Table 3 shows the optimized scheduling results for the three cases.

[0165]

[0166] To further analyze the optimized scheduling in Case 3, Figure 4 The electrical balance and hydrogen balance in Case 3 are shown. Figure 4 (a) is the power balance diagram of PM1. It can be seen that the electrolyzer and the self-owned power plant are almost always in full-power operation, and the state of buying and selling electricity from EHSES does not occur at the same time. Figure 4 (b) shows the power balance diagram of EHSES. It can be seen that the charging and discharging behavior of EHSES does not occur simultaneously, nor does it occur simultaneously with the power grid's power purchase and sale behavior. However, EHSES can sell and purchase electricity from PMs simultaneously because different PMs have different power demands. Figure 4 (c) shows the hydrogen balance diagram for PM1. It can be seen that PM1's hydrogen purchase rate from the SES is 0, indicating that the total EL of PM1 and the hydrogen produced by the hydrogen recovery unit can meet PM1's hydrogen demand. It is worth noting that the hydrogen obtained from the hydrogen recovery system accounts for approximately 2.46% of the hydrogen load, which can improve the utilization rate of hydrogen energy. Figure 4 (d) is the hydrogen balance diagram of EHSES. It can be seen that the hydrogen purchase rate of EHSES from the hydrogen refueling station is always 0, indicating that the hydrogen energy in PMs can achieve self-sufficiency.

[0167] 4.2 Optimization Results of EHSES

[0168] Figure 5 The presentation of Case 3 shows the state of charge and charge / discharge power of the SEES, and the pressure and hydrogen charge / discharge rate of the SHES. Analysis shows that the state of charge and pressure are both within the set limits, and no overcharging or over-discharging occurs. The electrical energy storage system is operational for 9 periods, and the hydrogen storage tank is operational for 12 periods, indicating that the hydrogen storage tank operates relatively frequently.

[0169] 4.3 Internal Price Analysis

[0170] Through continuous iteration, the optimal internal electricity price and hydrogen price were obtained as follows: Figure 6 As shown. Figure 6 (a) shows the electricity purchase and sale price of EHSES. In order to incentivize PMs to trade with EHSES rather than directly with the grid, the electricity purchase and sale price of SES is strictly controlled between the grid electricity purchase and sale price. Figure 6 (b) illustrates the hydrogen purchase and sale price of EHSES. Similarly, the hydrogen purchase and sale price of EHSES is strictly controlled within the range of hydrogen purchase and sale prices at hydrogen refueling stations. Analysis shows that the model proposed in this invention can incentivize PMs to trade with EHSES, thereby improving the system's sustainability and self-sufficiency.

[0171] This invention addresses the energy supply characteristics and low-carbon requirements of polycrystalline silicon microgrids by introducing shared energy storage for PMs and constructing a two-layer optimal scheduling model based on Stackelberg game theory. The following conclusions were obtained through policy verification: (1) The introduction of shared energy storage and hydrogen recovery equipment reduced the daily operating cost of PMs by 61.17%, effectively improving the economic efficiency of PMs; (2) The proposed model reduced carbon emissions by 2856.32 kg and increased the self-reliance factor to 98.21%, effectively improving the environmental friendliness and sustainability of the system.

[0172] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions conceived without inventive effort should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope defined in the claims.

Claims

1. An industrial micro-grid group double-layer optimization scheduling method for electricity-hydrogen sharing energy storage, comprising the following steps: Step 1, system architecture and component model Step 1.1 System architecture In the production of polycrystalline silicon, Siemens method is mostly used, which will consume a large amount of hydrogen during the production process. Traditional polycrystalline silicon enterprises mainly use the power supply mode of "self-provided power plant + purchased power" to produce hydrogen, which will produce a large amount of carbon emissions; In order to improve the low-carbon, reliability and sustainability of PMs, new energy is used to replace self-provided power plant to produce hydrogen, and EHSES is introduced to realize the collaborative and efficient operation of multiple PMs; Single PM contains two energy forms of electricity and hydrogen, mainly including wind turbine, photovoltaic module, self-provided power plant, electric hydrogen production equipment, hydrogen recovery equipment and polycrystalline silicon production equipment; Multiple PMs are connected with EHSES through hydrogen pipeline and cable, and indirect interaction and sharing between electricity and hydrogen among multiple microgrids can be realized through EHSES; EHSES contains operator, lithium iron phosphate battery and hydrogen storage tank; Under the incentive of internal price, multiple microgrids are more inclined to trade electricity and hydrogen with EHSES, the operator is not only the agent for electricity and hydrogen transaction between multiple microgrids and EHSES, but also responsible for electricity transaction between EHSES and external power grid and external hydrogen filling station, and is the scheduling center of EHSES: Step 1.2 Component model Step 2, double-layer optimization scheduling model based on Stackelberg game Step 2.1 Decision process The upper EHSES formulates internal electricity price, internal hydrogen price and charging and discharging power and hydrogen charging rate curve of EHSES, and the lower PM i optimizes its own energy consumption and energy transaction strategy; The energy transaction strategy of the lower PM i will be uploaded to the upper EHSES; After EHSES calculates its income, it will redefine the internal price accordingly until the optimal internal price is found; In addition, the decision of both parties is essentially sequential, which is consistent with the dynamic game situation of Stackelberg hierarchy; Therefore, the energy transaction process can be regarded as a Stackelberg game model: ； wherein, and respectively represent the price of the electrical power bought and sold by the EHSES from the PMs; and respectively represent the price of the HV bought and sold by the EHSES from the PMs; and respectively represent the electrical power bought and sold by the EHSES from the PMs; and respectively represent the HV bought and sold by the EHSES from the PMs; , , and respectively represent the economic objective function of the EHSES, the environmental objective function of the EHSES, the sustainability objective of the EHSES and the objective function of the PM i; wherein, EHSES are leaders in the upper model, and PMs are followers in the lower model; and denote the strategy set of leaders and followers, respectively. Step 2.2 Upper EHSES multi-objective optimization scheduling model Step 2.2.1 Objective function The economic objective of multi-objective optimization scheduling of EHSES is to maximize the daily profit of EHSES, the environmental protection objective is to minimize the carbon emissions of the system, and the sustainability objective is to maximize the self-sufficiency factor of the system; Step 2.2.2 Constraint conditions (1) Electricity balance constraint ； (2) Hydrogen energy balance constraint ； (3) Electricity transaction constraint ； wherein EHSES_grid_flag represents a flag for EHSES to purchase power from the grid, and is a binary variable with a value of 1 indicating purchase from the grid; PM_i_EHSES_flag represents a flag for PM i to purchase power from the EHSES, and is a binary variable with a value of 1 indicating purchase from the EHSES; (4) Hydrogen energy transaction constraint ； wherein EHSES i, PM i, and PM j are binary variables that take the value of 1 if the EHSES purchases hydrogen from the hydrogen station, and 0 otherwise; EHSES i, PM i, and PM j are binary variables that take the value of 1 if the EHSES purchases hydrogen from the hydrogen station, and 0 otherwise; Step 2.3 Lower multi-microgrid optimization scheduling model Step 2.3.1 Objective function Each microgrid optimizes its scheduling with the objective of minimizing its own operation cost, and the objective function can be expressed as: ； Step 2.3.2 Constraint conditions (1) Electricity balance constraint ； wherein represents the wind power; represents the photovoltaic power; represents the conventional electrical load; (2) Hydrogen energy balance constraint ； wherein represents the regular hydrogen loading in PMi; (3) Electricity transaction constraint ； (4) Hydrogen energy transaction constraint ； The constraints of the remaining devices are not described here; Step 3, solution method The two-stage optimization method is an effective method for solving the double-layer optimization scheduling problem based on Stackelberg game, therefore, the two-stage optimization algorithm based on the improved multi-objective slime mold algorithm and CPLEX solver is adopted to solve the above model.

2. The method of claim 1, wherein: The step 1.2 component model is specifically: Step 1.2.1 Electrolytic hydrogen production equipment model Because the alkaline electrolytic cell is low in cost, high in commercialization and more suitable for large-scale hydrogen production, the alkaline electrolytic cell is selected as the electrolytic hydrogen production equipment, and the model can be expressed as: ； wherein is the rate of hydrogen production by the EL; is the electrical power consumed by the EL; is the efficiency of the EL; is the higher heating value of the hydrogen; is the rated electrical power of the EL; Step 1.2.2 Hydrogen recovery equipment model The hydrogen recovery equipment is introduced to recover and utilize hydrogen in the polysilicon production process, so as to improve the utilization rate of hydrogen, and the mathematical model is as follows: ； In the formula, is the energy consumption coefficient of hydrogen recovery; is the hydrogen recovery efficiency; is the proportion of recoverable hydrogen; is the hydrogen supply rate of the system at time t; is the density of hydrogen; Step 1.2.3 Electric-hydrogen shared energy storage model.

3. The method of claim 2, wherein: The step 1.2.3 electric-hydrogen shared energy storage model is specifically: (1) Electric energy storage model The lithium iron phosphate battery is selected as the SEES; ； In the formula, and respectively represent the amount of electricity stored by the SEES at time t-1 and time t; represents the power dissipation rate of the SEES, in the present application, the self-discharge phenomenon of the SEES is ignored, so its value is 0; and respectively represent the charging power and discharging power of the SEES at time t; and respectively represent the charging efficiency and discharging efficiency of the SEES, for the purpose of simplifying the model, the present application ignores the influence of temperature and discharge depth on the charging and discharging efficiency, and and are set as constant values; is the time step; is the discharge flag of the SEES, with a value of 1 indicating charging and a value of 0 indicating discharging; and respectively are the rated capacity and rated power of the SEES; is the initial value of the capacity of the SEES; and respectively are the minimum value and maximum value of the SOC of the SEES; (2) Hydrogen storage tank model ； wherein, and respectively represent the hydrogen velocity (Hydrogen velocity, HV) of the hydrogen charged and discharged in the HST; , and respectively represent the minimum value, the maximum value and the initial value of the pressure in the HST; represents the gas constant; represents the temperature inside the HST; represents the volume of the HST; represents the molar mass of hydrogen energy; represents the flag of hydrogen charging and discharging of the HST, and the value of 1 represents hydrogen charging.

4. The method of claim 3, wherein: In the step 2.2.1 target function, the decision variables include the electricity price, hydrogen price and charging and discharging plan of the EHSES, and are as shown in the formula: (1) Economic target: maximize the daily income of the EHSES ； wherein, the on-grid electricity price; the electricity power sold to the grid by the EHSES; the electricity price purchased from the grid; the electricity power purchased from the grid by the EHSES; the hydrogen price sold to the hydrogen station; the HV sold to the hydrogen station by the SES; the hydrogen price purchased from the hydrogen station; the HV purchased from the hydrogen station by the EHSES; the capital return rate; the life span of the EHSES; the number of operation days in a year; the simulation period; the unit price of the battery power; the unit price of the battery capacity; the unit price of the hydrogen storage tank; the operation and maintenance coefficient; the residual value recovery coefficient; (2) Environmental protection target: minimize the carbon emissions of the system ； In the formula, is the carbon emission factor of traditional unit power generation; is the carbon quota coefficient of traditional unit power generation; is the power generation of self-provided power plant; is the carbon quota coefficient of renewable energy; (3) Sustainability target: maximize the self-sufficiency factor of the system 。

Images