Optimized operation method for participation of hydrogen refueling station in day-ahead market and balance market

Abstract

The invention discloses an optimized operation method for a hydrogen refueling station to participate in a day-ahead market and a balance market. The method comprises the following steps: constructing an upper-layer hydrogen refueling station bidding quotation model; a lower-layer day-ahead market clearing model composed of an electric energy market, a standby market and a hydrogen energy market is established; establishing a lower-layer balance market clearing model based on the bidding quotation of each subject and the reserve capacity provided by each subject of the lower-layer day-ahead market clearing model; constructing a lower-layer overall clearing model based on the middle-lower-layer day-ahead market clearing model and the middle-lower-layer balanced market clearing model; converting the lower-layer overall clearing model into a group of constraint conditions by using a KKT condition, adding the constraint conditions into an upper-layer hydrogen refueling station bidding quotation model, and converting the upper-layer model and the lower-layer model into a single-layer model; and respectively linearizing nonlinear terms of the single-layer model through a strong dual theorem and a big-M method so as to convert the nonlinear terms into a mixed integer programming model and directly solve the mixed integer programming model through a solver. According to the invention, the flexible resource regulation potential of the hydrogen refueling station is fully exploited, the flexibility of market transaction is improved, and the income of the hydrogen refueling station is increased.

Description

Technical Field

[0001] The present invention relates to the technical field of power system dispatching, and in particular to an optimized operation method for a hydrogen refueling station to participate in a day-ahead market and a balancing market. Background Art

[0002] Hydrogen-powered vehicles are a promising alternative because they are powered by hydrogen, have zero emissions, and can be recharged in just a few minutes. The superior characteristics of hydrogen-powered vehicles compared to traditional internal combustion engines and electric vehicles have attracted global attention. The construction of hydrogen refueling stations will promote the development of hydrogen-powered vehicles and related hydrogen industries, while supporting the consumption of renewable energy. However, the transportation cost of hydrogen is high. Therefore, hydrogen refueling stations equipped with on-site hydrogen production facilities are seen as the future trend of hydrogen refueling station development.

[0003] At present, some scholars have studied the participation of hydrogen refueling stations in a single energy market or in an energy market combined with auxiliary services, and all of them regard hydrogen refueling stations as price takers. However, in this case, the income of hydrogen refueling stations is low. When they are needed to play the role of hydrogen energy storage to smooth the peak and fill the valley, their income will decrease. In order to increase the income of hydrogen refueling stations, promote the consumption of new energy, increase the competitiveness of water electrolysis hydrogen production units in the hydrogen market, and reduce carbon emissions. Therefore, it is of great practical significance to study the optimal operation of hydrogen refueling stations participating in the day-ahead market and the balancing market as price makers. Summary of the invention

[0004] The present invention provides an optimized operation method for hydrogen refueling stations to participate in the day-ahead market and the balancing market. With hydrogen refueling stations as price makers, a two-layer optimization model for hydrogen refueling stations to participate in the joint market and the balancing market is established. The upper layer is a balancing market bidding and quotation optimization model, and the lower layer is a sequential clearing model for the joint market and the balancing market. At the same time, the two-layer model of hydrogen refueling station participation in the day-ahead market and the balancing market is converted into a single-layer model, which reduces the difficulty of solving the problem, fully taps the flexibility resource regulation potential of hydrogen refueling stations, improves the flexibility of market transactions, and increases the income of hydrogen refueling stations themselves.

[0005] The technical solution adopted by the present invention is:

[0006] A method for optimizing the operation of a hydrogen refueling station participating in a day-ahead market and a balancing market comprises the following steps:

[0007] Step 1: With the goal of maximizing the profit of hydrogen refueling stations, the upper-level hydrogen refueling station bidding model is constructed based on the market clearing price of the lower-level overall clearing model and the market information of hydrogen refueling stations participating in the market;

[0008] Step 2: Based on the bids of hydrogen refueling stations and other market participants, establish a lower-level day-ahead market clearing model consisting of electricity, backup and hydrogen markets;

[0009] Step 3: Considering the uncertainty of load forecast deviation, and based on the bids of each entity and the spare capacity provided by each entity in the lower-level day-ahead market clearing model, establish the lower-level balancing market clearing model;

[0010] Step 4: Based on the lower-level day-ahead market clearing model in step 2 and the lower-level equilibrium market clearing model in step 3, a lower-level overall clearing model is constructed, and a compact form of the lower-level overall clearing model and the upper-level hydrogen station bidding model in step 2 is given;

[0011] Step 5: Use KKT conditions to transform the lower-level overall clearing model in step 4 into a set of constraints and add them into the upper-level hydrogen station bidding model, thereby transforming the upper-lower double-layer model into a single-layer model;

[0012] Step 6: Linearize the nonlinear terms of the single-layer model in step 5 by using the strong duality theorem and the big-M method, thereby converting it into a mixed integer programming model, which is then directly solved by the CPLEX commercial solver on the MATLAB platform.

[0013] In step 1, the upper-level hydrogen station bidding model aims to maximize the profit of hydrogen stations participating in the day-ahead market and the equilibrium market, and its objective function expression is:

[0014]

[0015] Where: C d , C H2 , C re , C BL are the profits obtained by the hydrogen refueling station in the electricity, hydrogen, reserve and balancing markets respectively; t is the time index, is the clearing price of the electricity market at node i at time t, λ t H2 is the hydrogen market clearing price at time t; h is the index of the hydrogen refueling station, is the amount of electricity purchased by the hth hydrogen refueling station in the electricity market at time t, is the electricity sales volume of the hth hydrogen refueling station in the electricity market at time t, is the amount of hydrogen sold by the hth hydrogen refueling station in the electricity market at time t; Ω i is the set of power system nodes; N t is the number of time periods in a cycle; t UP and λ t DN are the upward and downward reserve capacity clearing prices of the reserve market at time t, respectively; The upward reserve capacity provided by the electrolyzer in the hth hydrogen refueling station at time t, is the downward reserve capacity provided by the electrolyzer in the hth hydrogen refueling station at time t, is the upward reserve capacity provided by the fuel cell in the hth hydrogen refueling station at time t, is the downward reserve capacity provided by the fuel cell in the hth hydrogen refueling station at time t; k is the net load deviation scenario index, N k represents the number of net load deviation scenarios, α k is the probability of net load deviation scenario k; is the clearing price of the equilibrium market in the tth hour in scenario k, is the upward reserve capacity of the electrolyzer in the hydrogen refueling station called in the balancing market at hour t under scenario k, is the downward reserve capacity of the electrolyzer in the hydrogen refueling station called in the balancing market at hour t under scenario k; is the upward reserve capacity of the fuel cell in the hydrogen station called in the balancing market at hour t under scenario k, is the downward reserve capacity of the fuel cell in the hydrogen refueling station called in the balancing market at hour t under scenario k.

[0016] The constraints of the above objective function include the bidding quotation constraints of hydrogen refueling stations in the day-ahead market and the constraints related to hydrogen storage tanks:

[0017] The hydrogen in the hydrogen storage tank is characterized by the hydrogen pressure, which is expressed as follows:

[0018]

[0019] Where: is the pressure in the hth hydrogen storage tank at time t; is the pressure in the hth hydrogen storage tank at time t-1; N k Indicates the number of net load deviation scenarios; is the efficiency of the hth electrolyzer, △t is the unit time interval, LHV is the lower heating value of hydrogen, is the efficiency of the hth fuel cell; and are the upper and lower limits of the pressure in the hth hydrogen storage tank, is the volume of the hydrogen storage tank in the hth hydrogen refueling station, R is the universal gas constant, is the temperature inside the hth gas tank, M H2 is the molar mass of hydrogen, is the volume of the hydrogen storage tank in the hth hydrogen refueling station; the third line of formula (2) represents the pressure of the hydrogen storage tank at the end of the scheduling period Equal to the initial pressure

[0020] The bidding constraint expression related to the electrolyzer is as follows:

[0021]

[0022] Where: The amount of electricity bid for hydrogen refueling stations in the electricity market. The bid volume for hydrogen refueling stations to sell hydrogen in the hydrogen energy market; Provide upward reserve capacity bidding in the reserve market for electrolyzers in hydrogen refueling stations; Provide downward reserve capacity bidding for electrolyzers in hydrogen refueling stations in the reserve market; are the maximum and minimum input powers of the h-th electrolyzer respectively; are the maximum upward and downward reserve capacities that can be provided by the h-th electrolyzer, respectively;

[0023] The bidding constraints related to fuel cells are similar to those in equation (3), as shown in equation (4):

[0024]

[0025] Where: The amount of electricity bid for hydrogen refueling stations to sell in the electricity market; Provide upward reserve capacity bidding volume in the reserve market for fuel cells in hydrogen refueling stations; Provide downward reserve capacity bidding for fuel cells in hydrogen refueling stations in the reserve market; are the maximum and minimum output powers of the hth fuel cell respectively; They are respectively the maximum upward and downward reserve capacities that the hth fuel cell can provide.

[0026] In step 2, after receiving the bids from various market participants including hydrogen refueling stations, a lower-level day-ahead market clearing model is established, which is based on the social welfare F DAM Maximization is the goal, and the objective function expression is:

[0027]

[0028] Where: F es 、F hs 、F UP 、F DN are the electricity sales revenue, hydrogen sales revenue, upward reserve capacity cost, and downward reserve capacity cost of all market members respectively; N d 、N g 、N dH2 and N s are the number of electricity users, power plants, hydrogen users and steam methane reforming hydrogen plants in the market; d is the electricity user index; g is the power plant index; d H2 is the hydrogen energy user index, s is the steam methane reforming hydrogen production plant index; and are the energy quotations of the hth hydrogen refueling station and the dth electricity user on the electricity market, respectively; is the energy quotation of the g-th power plant in the electricity market; p d,t is the amount of electricity purchased by the dth electricity user in the electricity market; p g,t is the amount of electricity sold by the g-th power plant in the electricity market; and The d H2 The hydrogen prices and purchase volumes of individual hydrogen users in the hydrogen energy market; The hydrogen price quoted by the hth hydrogen refueling station on the hydrogen energy market; and q s,t are the hydrogen bid and hydrogen sales of the sth steam methane reforming hydrogen production plant in the hydrogen energy market; and Quote the upward reserve capacity in the reserve market for hydrogen stations, electricity users and power plants respectively; and Upward reserve capacity for electricity users and power plants; and Quote downward reserve capacity in the reserve market for hydrogen stations, electricity users and power plants respectively; and Downward reserve capacity provided to electricity users and power plants.

[0029] The day-ahead market constraints include: power, hydrogen and reserve balance constraints, line transmission capacity constraints, the output of each participant, and the physical limitations of providing reserve capacity. Power, hydrogen and reserve balance constraints, and line transmission capacity constraints are shown in the following formula:

[0030]

[0031] Where: R t UP and R t DN is the upward and downward spare capacity requirements at time t; is the maximum transmission capacity of line l; matrix G is the transfer power matrix of line power flow and node power; G (l,:) is the row of matrix G corresponding to line l; matrix A H , A D , A G They are bus-hydrogenation station, bus-load, and bus power plant association matrices respectively.

[0032] The output of the hydrogen refueling station and the physical limitations of providing backup capacity are expressed as:

[0033]

[0034] The output of other participants and the physical constraints on providing backup capacity are expressed as:

[0035]

[0036] Where: is the upper limit of the electric energy load required by power user d at time t; is the output upper limit of power plant g; is the hydrogen energy user d at time t H2 The upper limit of the required hydrogen energy load; is the upper limit of the hydrogen quality that can be sold by the steam methane reforming hydrogen production plant s at time t.

[0037] In step 3, the uncertainty of the power load forecast deviation is processed by using the scenario, and the balancing market is cleared based on the bids of each subject and the spare capacity that each subject can provide in the day-ahead market. The balancing market is adjusted to minimize the cost of the load deviation by adjusting the output of each subject. BLM As the target, the specific expression is:

[0038]

[0039] Where: W load The cost of using the reserve capacity of power users in the balancing market; W gen The cost of using the reserve capacity of power plants in the balancing market; W HFS To balance the cost of calling the spare capacity of hydrogen refueling stations in the market; W Force To balance the cost of load shedding in the market; is the upward reserve capacity of power user d at time t in scenario k, The downward reserve capacity of power user d at time t under scenario k is called; is the upward reserve capacity of power plant g at time t under scenario k, B is the downward reserve capacity of power plant g at time t in scenario k; CL is the unit load shedding cost, is the load shedding amount of power user d at time t under scenario k; is the upward reserve capacity of the electrolyzer in the hydrogen refueling station called in the balancing market at hour t under scenario k, is the downward reserve capacity of the electrolyzer in the hydrogen refueling station called in the balancing market at hour t under scenario k; is the upward reserve capacity of the fuel cell in the hydrogen station called in the balancing market at hour t under scenario k, is the downward reserve capacity of the fuel cell in the hydrogen refueling station called in the balancing market at hour t under scenario k.

[0040] The balance between net load deviation and dispatch reserve is shown in the following equation:

[0041]

[0042] Where: L t,k is the net load deviation at time t under scenario k. In addition to the balance constraint, the lower-level balance market clearing model constraints also include load shedding upper and lower limit constraints and dispatch reserve capacity upper and lower limit constraints, as shown in the following formula;

[0043]

[0044] In step 4, the lower-layer day-ahead market clearing model and the lower-layer equilibrium market clearing model are first combined to establish a lower-layer overall clearing model, and then a compact form of the upper-layer hydrogen station bidding quotation and the lower-layer overall clearing model is given.

[0045] Since the adjustment cost of the balancing market can be regarded as part of the cost of the lower-level overall clearing model, the objective function of the lower-level overall clearing model is shown in formula (11):

[0046] max(F DAM -W BLM )(11);

[0047] The constraints of the objective function of the lower-level overall clearing model include: day-ahead market constraints and equilibrium market constraints.

[0048] Since the uncertainty of the power load in the present invention is handled by the scenario method, the lower overall clearing model is divided into two stages: the first stage is the decision-making that is independent of the scenario, including the output of each market player participating in the electricity-hydrogen market and the reserve provided, that is, here-and-now decision variables; the second stage is the reserve capacity and load shedding called by each player under a specific scenario k, that is, wait-and-see decision variables.

[0049] 1) The compact form of the upper hydrogen station bidding model can be expressed as:

[0050]

[0051] Where: λ is the auxiliary variable introduced in the KKT transformation process of the lower overall clearing model, and is also the set vector of market clearing prices; J1 is the auxiliary coefficient matrix; A1, B1 and C1 are the constraint constant coefficient matrices, and a is the constraint constant coefficient vector; x, y are the optimization variables of the upper hydrogen station bidding model; z and s are the optimization variables of the lower overall clearing model, and their specific expressions are:

[0052]

[0053] 2) The compact form of the lower-level overall clearing model can be expressed as:

[0054]

[0055] Where: J2 is the auxiliary coefficient matrix; T in the upper right corner represents the transpose of the matrix; A2, B2, C2, D2 and E2 are the constant coefficient matrices in the constraints; b, c, d are constant coefficient vectors; u, v and γ are introduced auxiliary variables.

[0056] In step 5, the KKT condition is used to transform the lower overall clearing model into a set of constraint conditions and add them into the upper hydrogen station bidding model, thereby transforming the upper and lower double-layer models into a single-layer model:

[0057] Step 1: Construct the Lagrangian function of the lower-level overall clearing model, as shown in formula (15).

[0058]

[0059] Step 2: Derivative the variables z and s to obtain the corresponding gradient conditions, and finally complete the addition of complementary relaxation conditions and non-negativity conditions of Lagrange multipliers, as shown in formula (16).

[0060]

[0061] In formula (16), the second and third lines are gradient conditions, the fourth, fifth and sixth lines are complementary relaxation conditions, and the eighth line is the non-negativity condition of the Lagrange multiplier.

[0062] In step 6, the nonlinear terms in the single-layer model are linearized respectively by using the strong duality theorem and the big-M method, thereby converting it into a mixed integer programming model.

[0063] For the quadratic terms in the constraints, taking u·(E2s-d)=0 as an example, linearization is performed using the big-M method:

[0064]

[0065] Wherein: M is a very large constant, and in the present invention, M=10000; n is an introduced auxiliary 0-1 variable.

[0066] For the quadratic term in the objective function, the strong duality theory is used for transformation, and the transformation process is as follows:

[0067] First, write the strong dual form of the objective function of the lower-level overall clearing model:

[0068] x T J2z+b T s = λc + vD2y + γd (18);

[0069] Multiply the second row of each term in equation (16) by z to obtain equation (19):

[0070] -x T J2z+λJ1z+uB2z+vC2z=0 (19);

[0071] Finally, the complementary relaxation condition of formula (16) and formula (19) are used to replace formula (18) equivalently to obtain:

[0072] λJ1z=-b T s+λc+γd (20);

[0073] Since then, all nonlinear terms in the single-layer model have been linearized, and the converted mixed integer programming model can be directly solved by the CPLEX commercial solver on the MATLAB platform.

[0074] Solution steps:

[0075] Step 1: Program the input parameters of the mixed integer programming model on the MATLAB platform. The input parameters include: the power upper limit and efficiency of the electrolyzer and fuel cell in the hydrogen refueling station, the upper and lower pressure limits of the hydrogen storage tank, the quotation of the power user, the upper limit of the power load demand and the upper limit of the backup capacity provided, the quotation of the power plant, the output upper limit and the upper limit of the backup capacity provided, the quotation of the hydrogen energy user and the upper limit of the hydrogen energy load demand, the quotation of the steam methane reforming hydrogen production plant and the upper limit of hydrogen sales, the topology of the power system, etc.

[0076] Step 2: Define the decision variables of the mixed integer programming model on the MATLAB platform;

[0077] Step 3: Program the compact form matrix and vector of the mixed integer programming model on the MATLAB platform according to equations (12) to (14);

[0078] Step 4: Program the objective function of the mixed integer programming model according to formula (20) on the MATLAB platform.

[0079] Step 5: Program the constraints of the mixed integer programming model on the MATLAB platform according to equations (12), (16) and (17).

[0080] Step 6: Call the CPLEX solver on the MATLAB platform to solve the mixed integer programming model.

[0081] Step 7: Output the quotation of hydrogen refueling station, the purchase and sale of electricity in the electricity market, the sales of hydrogen in the hydrogen market, the reserve capacity provided in the reserve market and the reserve capacity called in the balancing market on the MATLAB platform.

[0082] The present invention provides an optimized operation method for a hydrogen refueling station to participate in the day-ahead market and the balancing market, and the technical effects are as follows:

[0083] 1) In the present invention, the hydrogen refueling station participates in the electricity, hydrogen energy and backup markets as a price maker, which can effectively increase the revenue of the hydrogen refueling station.

[0084] 2) In the present invention, the joint clearing of the electricity, hydrogen and reserve markets optimizes resource allocation and improves system reliability, promotes the integration of renewable energy and the improvement of economic benefits, and reduces carbon emissions.

[0085] 3) In the present invention, balanced market regulation brings about the improvement of economic efficiency and consumer welfare through efficient resource allocation, promotion of competition and transmission of price signals, and increases the self-healing ability of the market, ensuring the real-time balance of electricity supply and demand, and ensuring the safe and stable operation of the power system.

[0086] 4) In the present invention, the lower-layer day-ahead market clearing model and the lower-layer equilibrium market clearing model are combined to construct a lower-layer overall clearing model, which facilitates the transformation of the two-layer model through KKT. The two-layer model is represented in a compact form to reduce redundant information, making the model more concise, easy to understand and communicate, reducing unnecessary variables and constraints, and improving the solution rate.

[0087] 5) In the present invention, the double-layer model is converted into a single-layer model through the KKT condition, which can not only simplify problem solving and improve computational efficiency, but also enhance the flexibility and scalability of the model.

[0088] 6) In the present invention, when linearizing the bilinear terms in the objective function of the upper hydrogen station bidding model, compared with other linearization methods such as the Boolean expansion method, the use of strong duality theory for linearization has the advantages of small error and short solution time.

[0089] 7) In the present invention, participation in the reserve and balancing market can significantly increase the revenue of hydrogen refueling stations and encourage more users to invest in the construction of hydrogen refueling stations. The production and consumption of hydrogen can play an important role in the balance of the power system, and can effectively absorb new energy and perform peak load shifting and valley filling. By providing reserve capacity, the competitiveness of hydrogen electrolysis in hydrogen refueling stations against steam methane reforming hydrogen plants in the hydrogen energy market is improved, and carbon emissions are reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0090] The present invention will be further described below in conjunction with the accompanying drawings and embodiments:

[0091] Figure 1 Schematic diagram of the framework of the two-layer optimization model.

[0092] Figure 2 Schematic diagram of the simulation network topology.

[0093] Figure 3 This is a schematic diagram of the predicted wind and solar power output and electricity and hydrogen loads.

[0094] Figure 4 Schematic diagram of the market clearing price.

[0095] Figure 5 Schematic diagram of the electrical power of the electrolyzer and fuel cell.

[0096] Figure 6 This is a trading chart of the hydrogen energy market.

[0097] Figure 7 This is a graph showing the proportion of various benefits of hydrogen refueling stations in the market.

[0098] Figure 8 This is the sensitivity analysis result diagram.

[0099] Fig. 9 Solve the flow chart for the mixed integer programming model. DETAILED DESCRIPTION

[0100] A method for optimizing the operation of a hydrogen fueling station (Hydrogen fueling station) participating in a combined electricity-reserve-hydrogen energy market, comprising the following steps:

[0101] S1. With the goal of maximizing the profit of hydrogen refueling stations, the upper-level hydrogen refueling station bidding model is constructed based on the market clearing price of the lower-level model and the market information of hydrogen refueling stations;

[0102] S2. Based on the bids and quotations of hydrogen refueling stations and other market participants in S1, a lower-level day-ahead market clearing model consisting of electricity, backup and hydrogen energy markets is established;

[0103] S3, considering the uncertainty of load forecast deviation, and based on the bids of various entities and the spare capacity provided by various entities in the day-ahead market in S2, establish a lower-level balancing market clearing model;

[0104] S4. Based on the day-ahead market clearing model in S2 and the equilibrium market clearing model in S3, a lower-level overall clearing model is constructed, and its compact form with the upper-level model in S1 is given;

[0105] S5: Use KKT (Karush-Kuhn-Tucker) conditions to transform the lower-level overall clearing model in S4 into a set of constraints and add them to the upper-level model, thereby transforming the upper-lower double-level model into a single-level model;

[0106] S6. The nonlinear terms of the single-layer model in S5 are linearized by the strong duality theorem and the big-M method respectively, thereby converting it into a mixed integer programming model, which is then directly solved by the CPLEX commercial solver on the MATLAB platform.

[0107] The present invention transforms the two-layer model of hydrogen refueling station participation from the day-ahead market and the equilibrium market into a single-layer model, which reduces the difficulty of solving the problem, fully taps the flexibility resource regulation potential of hydrogen refueling stations, improves the flexibility of market transactions, and increases the revenue of hydrogen refueling stations themselves.

[0108] Example:

[0109] This section adopts Figure 3 The IEEE 14-bus bus system shown in the figure verifies the effectiveness of the proposed two-stage model. Figure 4 Shows solar power, wind power generation and total power load. Reserve demand R t UP and R t DN The uncertainty in the balancing market is represented by two scenarios, namely N k =2, their probabilities are equal, α k =0.5. The grid consists of 6 power loads and 4 power plants. The load bid is 95€ / MWh, and the load at node 6 and node 8 offers a reserve bid of 90€ / MWh. Conventional power plant 1, conventional power plant 2, photovoltaic power plant, and wind farm are established at nodes 1, 2, 4, and 12, with electricity sales bids of 55, 80, 13, and 13€ / MWh respectively, and reserve bids of 10, 40, 6, and 6€ / MWh. The maximum output power of the two conventional power plants is 30MW and 94MW respectively. And the reserve that the power plant and the load can provide should not exceed 10% of the maximum output. The hydrogen refueling station is located at node 9, the minimum tank pressure of the hydrogen storage tank is 10bar, the maximum tank pressure is 100bar, the maximum power of the electrolyzer and fuel cell is 1MW and 0.5MW respectively, the minimum power is 0.1MW and 0.05MW, and the efficiency is 0.7. Other relevant data are shown in Table 1.

[0110] Table 1 Hydrogen station related parameters

[0111]

[0112] According to S1, an upper-level optimization model with the goal of maximizing the profit of hydrogen refueling stations is established;

[0113] According to S2, a day-ahead market clearing model is established;

[0114] According to S3, establish the equilibrium market clearing model;

[0115] According to S4, the lower-level overall clearing model is established, and its compact form with the upper-level model is given;

[0116] According to S5, the overall clearing model of the lower layer in S4 is transformed using the KKT condition;

[0117] According to S6, the nonlinear terms in the single-layer model in S5 are linearized by the strong duality theorem and the big-M method, respectively, and the final model is directly solved by the CPLEX commercial solver on the MATLAB platform.

[0118] 1: Analysis of hydrogen refueling stations participating in the joint market operation:

[0119] (1) Energy Market

[0120] like Figure 3 , Figure 4 , Figure 5 As shown, there is a positive correlation between the transaction price in the electricity market and the net load curve, which indicates that the transaction price of the hydrogen refueling station in the electricity market can accurately reflect the supply and demand of electricity. During the peak period of net load, as the load increases, the transaction price in the electricity market also rises significantly. The electrolyzer consumes electricity to produce hydrogen at every moment of the day, and it does not operate at rated capacity. This is because some capacity needs to be reserved for the backup market in order to profit from the backup market. At t=1-5 and 23, 24, the electricity price is low, and the electrolyzer purchases more electricity to produce hydrogen and stores the excess hydrogen in the hydrogen storage tank; at t=10, 11, 17, 18, the electricity price is high, and the electrolyzer input power is small, thereby reducing costs. It can be seen from Figure 6 It can be seen that the output power of the fuel cell is very small. This is because the energy conversion efficiency through electricity-hydrogen-electricity is low, which makes fuel cell power generation uneconomical. However, at t=1-8 and t=17-20, because the clearing price of the backup market is high, in order to obtain benefits in the backup market, the output power of the fuel cell is not 0.

[0121] (2) Hydrogen energy market:

[0122] like Figure 6 As shown in the figure, at t=1-5, after the hydrogen generated by the hydrogen refueling station meets the hydrogen load, the excess hydrogen is stored in the hydrogen storage tank. At t=6-11, the hydrogen load is large, and the hydrogen generated by the hydrogen refueling station alone cannot meet the demand. At this time, the hydrogen storage tank releases hydrogen to meet the hydrogen balance in the market. At t=17 and 18, affected by the hydrogen release rate of the hydrogen storage tank and the high electricity price, a steam methane reforming hydrogen production plant is needed to generate part of the hydrogen to meet the hydrogen load.

[0123] (3) Revenue analysis:

[0124] Figure 7The figure shows the cost and revenue structure of hydrogen refueling stations in the integrated market, with the inner circle representing cost, the outer circle representing revenue, and the ring representing profit. Figure 7 It can be seen that for hydrogen refueling stations, electricity costs and downstream balancing costs account for 61.1% and 38.9% of total costs respectively. H2 sales account for 32.5% of revenue, upstream balancing revenue accounts for 26.1% of revenue, upstream reserve accounts for 23.2% of revenue, and downstream reserve accounts for 32.5% of revenue.

[0125] 2: Sensitivity analysis:

[0126] like Figure 8 As shown in the neutron diagram a, the revenue of hydrogen refueling stations will increase with the increase of electricity load. This is because the increase in load may lead to insufficient power supply capacity during peak load periods, which will increase the difference in electricity prices between peak load and trough load periods. Therefore, hydrogen refueling stations can play the role of hydrogen energy storage to suppress the peak-to-valley difference.

[0127] like Figure 8 As shown in the neutron diagram b, if the power plant quotation at node 2 increases, the revenue of the hydrogen refueling station will decrease. This is because it will lead to an increase in the energy market clearing price, resulting in an increase in the electricity purchase cost of the hydrogen refueling station. The electrolyzer's quotation in the hydrogen energy market will increase, and its market will be squeezed out by the hydrogen generated by the steam methane reforming hydrogen plant, resulting in a decrease in profits in the hydrogen energy market, which has little impact on fuel cells, thereby resulting in a decrease in the revenue of the hydrogen refueling station.

[0128] like Figure 8 As shown in neutron diagram c, as the hydrogen load increases, the price of hydrogen increases, and the hydrogen production of the hydrogen refueling station will also increase, resulting in greater revenue from hydrogen sales.

[0129] picture Figure 8 As shown in neutron diagram d, the quotation of steam methane reforming hydrogen production plant will affect its competitiveness in the hydrogen energy market. As its quotation increases, its market share will decrease, and the market share of hydrogen refueling stations will increase, resulting in increased profits for hydrogen refueling stations.

Claims

1. A method for optimizing the operation of a hydrogen refueling station participating in the day-ahead market and the balancing market, characterized in that The following steps are involved: Step 1: With the goal of maximizing the profit of hydrogen refueling stations, the upper-level hydrogen refueling station bidding model is constructed based on the market clearing price of the lower-level overall clearing model and the market information of hydrogen refueling stations participating in the market; Step 2: Based on the bids of hydrogen refueling stations and other market participants, establish a lower-level day-ahead market clearing model consisting of electricity, backup and hydrogen markets; Step 3: Considering the uncertainty of load forecast deviation, and based on the bids of each entity and the spare capacity provided by each entity in the lower-level day-ahead market clearing model, establish the lower-level balancing market clearing model; Step 4: Based on the lower-layer day-ahead market clearing model in step 2 and the lower-layer equilibrium market clearing model in step 3, construct a lower-layer overall clearing model; Step 5: Use KKT conditions to transform the lower-level overall clearing model in step 4 into a set of constraints and add them into the upper-level hydrogen station bidding model, thereby transforming the upper-lower double-layer model into a single-layer model; Step 6: Linearize the nonlinear terms of the single-layer model in step 5 by using the strong duality theorem and the big-M method, thereby converting it into a mixed integer programming model and solving it directly through the solver.

2. According to claim 1, a method for optimizing the operation of a hydrogen refueling station participating in the day-ahead market and the balancing market, characterized in that: In step 1, the upper-level hydrogen station bidding model aims to maximize the profit of hydrogen stations participating in the day-ahead market and the equilibrium market, and its objective function expression is: Where: C d , C H2 , C re , C BL They are the profits obtained by hydrogen refueling stations in the electricity, hydrogen, reserve and balancing markets; t is the time index, is the clearing price of electricity market at node i at time t, is the hydrogen energy market clearing price at time t; h is the index of the hydrogen refueling station, is the amount of electricity purchased by the hth hydrogen refueling station in the electricity market at time t, is the electricity sales volume of the hth hydrogen refueling station in the electricity market at time t, is the amount of hydrogen sold by the hth hydrogen refueling station in the electricity market at time t; Ω i is the set of power system nodes; N t is the number of periods in a cycle; and are the upward and downward reserve capacity clearing prices of the reserve market at time t, respectively; The upward reserve capacity provided by the electrolyzer in the hth hydrogen refueling station at time t, The downward reserve capacity provided by the electrolyzer in the hth hydrogen refueling station at time t, The upward reserve capacity provided by the fuel cell in the hth hydrogen refueling station at time t, is the downward reserve capacity provided by the fuel cell in the hth hydrogen refueling station at time t; k is the net load deviation scenario index, N k represents the number of net load deviation scenarios, α k is the probability of net load deviation scenario k; is the clearing price of the equilibrium market in the tth hour in scenario k, is the upward reserve capacity of the electrolyzer in the hydrogen refueling station called in the balancing market at hour t under scenario k, is the downward reserve capacity of the electrolyzer in the hydrogen refueling station called in the balancing market at hour t under scenario k; is the upward reserve capacity of the fuel cell in the hydrogen station called in the balancing market at hour t under scenario k, is the downward reserve capacity of the fuel cell in the hydrogen refueling station called in the balancing market at hour t under scenario k.

3. According to claim 2, a method for optimizing the operation of a hydrogen refueling station participating in the day-ahead market and the balancing market, characterized in that: The constraints of the objective function include: the day-ahead market bidding quotation constraints of hydrogen refueling stations and the constraints related to hydrogen storage tanks: The hydrogen in the hydrogen storage tank is characterized by the hydrogen pressure, which is expressed as follows: Where: is the pressure in the hth hydrogen storage tank at time t; is the pressure in the hth hydrogen storage tank at time t-1; N k Indicates the number of net load deviation scenarios; is the efficiency of the hth electrolyzer, △t is the unit time interval, LHV is the lower heating value of hydrogen, is the efficiency of the hth fuel cell; and are the upper and lower limits of the pressure in the hth hydrogen storage tank, is the volume of the hydrogen storage tank in the hth hydrogen refueling station, R is the universal gas constant, is the temperature inside the hth gas tank, M H2 is the molar mass of hydrogen, is the volume of the hydrogen storage tank in the hth hydrogen refueling station; The bidding constraint expression related to the electrolyzer is as follows: Where: The amount of electricity bid for hydrogen refueling stations in the electricity market. The bid volume for hydrogen refueling stations to sell hydrogen in the hydrogen energy market; Provide upward reserve capacity bidding in the reserve market for electrolyzers in hydrogen refueling stations; Provide downward reserve capacity bidding for electrolyzers in hydrogen refueling stations in the reserve market; are the maximum and minimum input powers of the hth electrolyzer respectively; are the maximum upward and downward reserve capacities that the h-th electrolyzer can provide, respectively; The bidding constraints related to fuel cells are similar to those in equation (3), as shown in equation (4): Where: The amount of electricity bid for hydrogen refueling stations to sell in the electricity market; Provide upward reserve capacity bidding volume in the reserve market for fuel cells in hydrogen refueling stations; Provide downward reserve capacity bidding for fuel cells in hydrogen refueling stations in the reserve market; are the maximum and minimum output powers of the hth fuel cell respectively; They are respectively the maximum upward and downward reserve capacities that the hth fuel cell can provide.

4. According to claim 1, a method for optimizing the operation of a hydrogen refueling station participating in the day-ahead market and the balancing market, characterized in that: In step 2, after receiving the bids from various market participants including hydrogen refueling stations, a lower-level day-ahead market clearing model is established, which is based on the social welfare F DAM Maximization is the goal, and the objective function expression is: Where: F es 、F hs 、F UP 、F DN are the electricity sales revenue, hydrogen sales revenue, upward reserve capacity cost, and downward reserve capacity cost of all market members respectively; N d 、N g 、N dH2 and N s are the number of electricity users, power plants, hydrogen users and steam methane reforming hydrogen plants in the market; d is the electricity user index; g is the power plant index; d H2 is the hydrogen energy user index, s is the steam methane reforming hydrogen production plant index; and are the energy quotations of the hth hydrogen refueling station and the dth electricity user on the electricity market, respectively; is the energy quotation of the g-th power plant in the electricity market; p d,t is the amount of electricity purchased by the dth electricity user in the electricity market; p g,t is the amount of electricity sold by the g-th power plant in the electricity market; and The d H2 The hydrogen prices and purchase volumes of individual hydrogen users in the hydrogen energy market; The hydrogen price quoted by the hth hydrogen refueling station on the hydrogen energy market; and q s,t are the hydrogen bid and hydrogen sales of the sth steam methane reforming hydrogen production plant in the hydrogen energy market; and Quote the upward reserve capacity in the reserve market for hydrogen stations, electricity users and power plants respectively; and Upward reserve capacity for electricity users and power plants; and Quote downward reserve capacity in the reserve market for hydrogen stations, electricity users and power plants respectively; and Downward reserve capacity provided to electricity users and power plants.

5. The method for optimizing the operation of a hydrogen refueling station participating in the day-ahead market and the balancing market according to claim 4, characterized in that: The day-ahead market constraints include: power, hydrogen and reserve balance constraints, line transmission capacity constraints, the output of each participant, and the physical limitations of providing reserve capacity; power, hydrogen and reserve balance constraints, and line transmission capacity constraints are shown in the following formula: Where: and is the upward and downward spare capacity requirements at time t; is the maximum transmission capacity of line l; matrix G is the transfer power matrix of line power flow and node power; G (l,:) is the row of matrix G corresponding to line l; matrix A H , A D , A G They are bus-hydrogenation station, bus-load, and bus power plant association matrices; The output of the hydrogen refueling station and the physical limitations of providing backup capacity are expressed as: The output of other participants and the physical constraints on providing backup capacity are expressed as: Where: is the upper limit of the electric energy load required by power user d at time t; is the upper limit of the power plant g’s output; is the hydrogen energy user d at time t H2 The upper limit of the required hydrogen energy load; is the upper limit of the hydrogen quality that can be sold by the steam methane reforming hydrogen production plant s at time t.

6. The method for optimizing the operation of a hydrogen refueling station participating in the day-ahead market and the balancing market according to claim 1, characterized in that: In step 3, the uncertainty of the power load forecast deviation is processed by using the scenario, and the balancing market is cleared based on the bids of each subject and the spare capacity that each subject can provide in the day-ahead market. The balancing market is adjusted to minimize the cost of the load deviation by adjusting the output of each subject. BLM As the target, the specific expression is: Where: W load The cost of using the reserve capacity of power users in the balancing market; W gen The cost of using the reserve capacity of power plants in the balancing market; W HFS To balance the cost of calling the spare capacity of hydrogen refueling stations in the market; W Force To balance the cost of load shedding in the market; is the upward reserve capacity of power user d at time t in scenario k, The downward reserve capacity of power user d at time t under scenario k; is the upward reserve capacity of power plant g at time t under scenario k, B is the downward reserve capacity of power plant g at time t in scenario k; CL is the unit load shedding cost, is the load shedding amount of power user d at time t under scenario k; is the upward reserve capacity of the electrolyzer in the hydrogen refueling station called in the balancing market at hour t under scenario k, is the downward reserve capacity of the electrolyzer in the hydrogen refueling station called in the balancing market at hour t under scenario k; is the upward reserve capacity of the fuel cell in the hydrogen station called in the balancing market at hour t under scenario k, is the downward reserve capacity of the fuel cell in the hydrogen refueling station called in the balancing market at hour t under scenario k; The balance between net load deviation and dispatch reserve is shown in the following equation: Where: L t,k is the net load deviation at time t under scenario k. In addition to the balance constraint, the lower-level balance market clearing model constraints also include load shedding upper and lower limit constraints and dispatch reserve capacity upper and lower limit constraints, as shown in the following formula; 7. The method for optimizing the operation of a hydrogen refueling station participating in the day-ahead market and the balancing market according to claim 1, characterized in that: In step 4, the lower-layer day-ahead market clearing model and the lower-layer equilibrium market clearing model are first combined to establish a lower-layer overall clearing model, and then a compact form of the upper-layer hydrogen station bidding quotation and the lower-layer overall clearing model is given; Since the adjustment cost of the balancing market can be regarded as part of the cost of the lower-level overall clearing model, the objective function of the lower-level overall clearing model is shown in formula (11): max(F DAM -W BLM ) (11)。 8. The method for optimizing the operation of a hydrogen refueling station participating in the day-ahead market and the balancing market according to claim 7, characterized in that: The lower-level overall clearing model is divided into two stages: the first stage is the decision-making that is independent of the scenario, including the output and reserve provided by each market player in the electricity-hydrogen market, that is, here-and-now decision variables; the second stage is the reserve capacity and load shedding called by each player under a specific scenario k, that is, wait-and-see decision variables; 1) The compact form of the upper hydrogen station bidding model is expressed as: Where: λ is the auxiliary variable introduced in the KKT transformation process of the lower-level overall clearing model, and is also the set vector of market clearing prices; J1 is the auxiliary coefficient matrix; A1, B1 and C1 are the constraint constant coefficient matrices, and a is the constraint constant coefficient vector; x, y are the optimization variables of the upper-level hydrogen station bidding model; z and s are the optimization variables of the lower-level overall clearing model, and their specific expressions are: 2) The compact form of the lower-level overall clearing model can be expressed as: Where: J2 is the auxiliary coefficient matrix; T in the upper right corner represents the transpose of the matrix; A2, B2, C2, D2 and E2 are the constant coefficient matrices in the constraints; b, c, d are constant coefficient vectors; u, v and γ are introduced auxiliary variables.

9. The method for optimizing the operation of a hydrogen refueling station participating in the day-ahead market and the balancing market according to claim 1, characterized in that: In step 5, the KKT condition is used to transform the lower overall clearing model into a set of constraint conditions and add them into the upper hydrogen station bidding model, thereby transforming the upper and lower double-layer models into a single-layer model: Step 1: Construct the Lagrangian function of the lower-level overall clearing model, as shown in formula (15); Step 2: Derivative the variables z and s to obtain the corresponding gradient conditions, and finally complete the addition of complementary relaxation conditions and non-negativity conditions of Lagrange multipliers, as shown in formula (16); 10. The method for optimizing the operation of a hydrogen refueling station participating in the day-ahead market and the balancing market according to claim 1, characterized in that: In step 6, the nonlinear terms in the single-layer model are linearized respectively by the strong duality theorem and the big-M method, thereby converting it into a mixed integer programming model; For the quadratic terms in the constraints, for u·(E2s-d)=0, linearization is performed using the big-M method: Where: M is a very large constant; n is an auxiliary 0-1 variable introduced; For the quadratic term in the objective function, the strong duality theory is used for transformation. The transformation process is as follows: First, write the strong dual form of the objective function of the lower-level overall clearing model: x T J2z+b T s=λc+vD2y+γd (18); Multiply the second row of each term in equation (16) by z to obtain equation (19): -x T J2z+λJ1z+uB2z+vC2z=0 (19); Finally, the complementary relaxation condition of formula (16) and formula (19) are used to replace formula (18) equivalently to obtain: λJ1z=-b T s+λc+γd (20); Since then, all nonlinear terms in the single-layer model have been linearized, and the converted mixed integer programming model can be directly solved by the CPLEX commercial solver on the MATLAB platform.

Images