Electricity-hydrogen comprehensive energy system scheduling method based on pipe-pipeline cooperative hydrogen transportation

Abstract

The invention discloses an electricity-hydrogen comprehensive energy system scheduling method based on pipe-pipeline cooperative hydrogen transportation. The method comprises the steps that a variable-hydrogen-doping-ratio natural gas hydrogen-doping transportation model based on flow balance is established; establishing a variable-hydrogen-doping-ratio natural gas hydrogen-doping transportation linear model; establishing a long tube trailer hydrogen transporting model; establishing an electricity-hydrogen comprehensive energy system scheduling model based on pipe-pipeline cooperative hydrogen transportation by considering related constraints of an electric power system and coupling constraints between a natural gas system and a road system and between the electric power system and the natural gas system and combining a variable hydrogen-doping-ratio natural gas hydrogen-doping transportation linear model and a long pipe trailer hydrogen transportation model; and on the basis of the electro-hydrogen comprehensive energy system scheduling model, establishing an electro-hydrogen comprehensive energy system distribution robust scheduling model, re-representing chance constraints in the distribution robust scheduling model as linear constraints, and solving the converted electro-hydrogen comprehensive energy system scheduling linear model. According to the scheduling method, the solving difficulty is reduced, the hydrogen transportation is reduced, and the hydrogen transportation accuracy is improved; and peak clipping and valley filling are carried out on the electrical load.

Description

Technical Field

[0001] The present invention relates to the technical field of electric-hydrogen integrated energy dispatching, and in particular to a dispatching method for an electric-hydrogen integrated energy system based on pipeline-road coordinated hydrogen transportation. Background Art

[0002] In recent years, in order to cope with the energy and environmental problems caused by the excessive consumption of traditional fossil energy, the installed capacity of renewable energy, including photovoltaic, wind power and hydropower, has grown rapidly. Due to the volatility and instability of renewable energy, and the fact that most renewable energy is distributed in remote areas, it is difficult to absorb a large amount of renewable energy. Combining the power system with a hydrogen energy system containing a water electrolysis hydrogen production plant is an effective way to solve the problem of surplus renewable energy. Hydrogen can not only be used in fuel cell vehicles, but can also be widely used in power generation, industrial production, heating and other fields.

[0003] However, since it is necessary to build a hydrogen production plant near a renewable energy power plant, most renewable energy hydrogen production plants are located in remote areas, while most hydrogen users are located in economically developed areas. This requires the transportation of hydrogen from the hydrogen production plant to the hydrogen users. Hydrogen can be transported by ships, hydrogen tube trailers, and pipelines. For example, as described in the literature [1]: C. Shao, C. Feng, M. Shahidehpour, Q. Zhou, X. Wang, and X. Wang. Optimal Stochastic Operation of Integrated Electric Power and Renewable Energy With Vehicle-Based Hydrogen Energy System [J]. IEEE TRANSACTIONS ON POWER SYSTEMS, 2021, 36 (5): 4310-4321.

[0004] Ship transportation is mainly applicable to water areas such as river networks or oceans, and is subject to greater geographical restrictions. Hydrogen in trailer transportation can be transported in gaseous, liquid, or solid forms. Reference [2]: Cheng Huan, Ren Zhouyang, Sun Zhiyuan, et al. Electricity-hydrogen coupling system scheduling under cross-regional coordinated transportation of electricity and methanol [J]. Transactions of the Chinese Society of Electrotechnical Engineering, 2024, 39(3): 731-744.

[0005] Document [3]: H. Tan, Y. Wang, Q. Wang, and Z. Lin. Day-ahead dispatch of electricity-hydrogen systems under solid-state transportation mode of hydrogen energy via FV-IGDT approach [J]. ENERGY, 2024, 300.

[0006] However, they are only suitable for short-distance transportation within a city or between adjacent cities. When transporting hydrogen over long distances and on a large scale, the transportation cost will increase significantly. Pipeline transportation includes two types: hydrogen pipelines and natural gas pipelines. Among them, the former is not considered for the time being due to its high construction cost and relatively short existing construction length. The latter can be used for large-scale hydrogen transportation due to the large existing natural gas pipeline network and the gradual maturity of pipeline hydrogen blending technology, but it cannot accurately meet the user's hydrogen load demand due to the hydrogen blending ratio. Summary of the invention

[0007] In order to solve the above technical problems, the present invention provides a method for dispatching an electric-hydrogen integrated energy system based on pipeline-road coordinated hydrogen transportation. First, based on the steady-state model of the natural gas pipeline, by taking into account hydrogen injection and considering pipeline storage, a variable hydrogen blending ratio natural gas hydrogen transportation model based on flow balance is proposed; secondly, a natural gas hydrogen transportation linear model is derived based on piecewise linearization, piecewise McCormick technology and equivalent substitution method; finally, natural gas hydrogen transportation is combined with trailer transportation, a pipeline-road coordinated hydrogen transportation mode is proposed, and the electric-hydrogen integrated energy system is dispatched. This scheduling method reduces the difficulty of solving, reduces the transportation of hydrogen, and increases the accuracy of hydrogen transportation; and it also gives play to the characteristics of hydrogen energy storage and performs peak shaving and valley filling of the electric load.

[0008] The technical solution adopted by the present invention is:

[0009] The method for dispatching an electric-hydrogen integrated energy system based on pipeline-road coordinated hydrogen transportation includes the following steps:

[0010] Step 1: Based on the steady-state model of the natural gas pipeline, taking into account hydrogen injection and pipeline storage, a natural gas hydrogen blending transportation model with variable hydrogen blending ratio based on flow balance is established;

[0011] Step 2: Linearize the variable hydrogen blending ratio natural gas hydrogen blending transportation model in step 1 based on piecewise linearization, piecewise McCormick technology and equivalent substitution method, and establish a variable hydrogen blending ratio natural gas hydrogen blending transportation linear model;

[0012] Step 3: Establish a long-tube trailer hydrogen transportation model by simulating the spatiotemporal dynamics of the long-tube trailer;

[0013] Step 4: Considering the constraints related to the power system and the coupling constraints between the natural gas system and the road system, as well as between the power system and the natural gas system, and combining the variable hydrogen blending ratio natural gas hydrogen blending transportation linear model in step 2 and the long tube trailer hydrogen transportation model in step 3, establish an electric-hydrogen integrated energy system scheduling model based on pipeline-road coordinated hydrogen transportation;

[0014] Step 5: Based on the dispatch model of the electric-hydrogen integrated energy system in step 4, considering the uncertainty of wind power, a distributed robust dispatch model of the electric-hydrogen integrated energy system is established, and the opportunity constraints in the distributed robust dispatch model are re-expressed as linear constraints. Finally, the GUROBI solver is called on the MATLAB platform to solve the converted linear model of the electric-hydrogen integrated energy system dispatch.

[0015] In step 1, the natural gas hydrogen blending transportation model with variable hydrogen blending ratio based on flow balance is established as follows:

[0016]

[0017] In formula (1): Ω SG is the natural gas source point set; Ω SH is the set of hydrogen source points; Ω in is the set of pipes flowing into node m; Ω out is the set of pipes that flow out of node m; are the volume flow rates of natural gas and hydrogen injected at node m at time t, respectively; are the volume flow rates of the mixed gas flowing out of the pipeline and flowing into the node m at time t, respectively; is the volume flow rate of the mixed gas flowing out of node m at time t;

[0018]

[0019] In formula (2): are the volume flow rates of hydrogen flowing out of the pipeline and flowing into node m at time t, respectively; is the volume flow rate of the mixed gas flowing out of node m at time t;

[0020]

[0021] In formula (3): m,t is the volume ratio of hydrogen in the mixed gas at node m at time t;

[0022]

[0023] In formula (4): is the natural gas load at node m at time t;

[0024]

[0025] In formula (5): mn,t is the volume ratio of hydrogen in the mixed gas in the pipeline (m,n) at time t;

[0026]

[0027] ω m,t =ω mn,t (7);

[0028]

[0029] In formula (8): is the average volume flow rate of the pipeline (m,n) at time t; p m,t and p n,t are the pressures at nodes m and n at time t respectively; is the Weymouth coefficient of the pipe (m,n);

[0030]

[0031] In formula (10): and are the upper and lower limits of the hourly volume flow rate of the pipeline (m,n);

[0032]

[0033] In formula (11): is the maximum pressure ratio at both ends of the pipeline; p m,t and p n,t are the pressures at nodes m and n at time t respectively;

[0034]

[0035] In formula (12): The minimum allowable pressure of pipeline nodes; The maximum allowable pressure at a pipeline node.

[0036]

[0037] In formula (13): B mn,t and B mn,t-1 is the pipeline inventory of pipeline (m,n) at time t and t-1;

[0038]

[0039] In formula (14): is the pipe storage coefficient of the pipeline (m,n);

[0040]

[0041] In formula (15): Ω pipe is the collection of all natural gas pipelines; B min Indicates the minimum value of the sum of all pipeline inventories.

[0042]

[0043] In formula (16): and is the maximum and minimum volume flow rate of natural gas that can be injected into node m; is the volume flow rate of natural gas injected at node m at time t.

[0044]

[0045] In formula (17): and is the maximum and minimum volume flow rate of natural gas that can be injected into node m; is the volume flow rate of hydrogen injected at node m at time t.

[0046] They are the Weymouth coefficient and pipe storage coefficient of the pipeline (m,n), and their expressions are as follows:

[0047]

[0048] In formula (18) and formula (19), D mn , l mn 、f mn are the diameter, length and friction coefficient of the pipe (m,n) respectively; π is the circumference of a circle; and are the specific gas constant and density of the mixed gas respectively; The compression factor and the density of the mixed gas when hydrogen is injected into the pipeline (m, n); and are the specific gas constant and density of natural gas respectively; is the compression coefficient of the pipeline (m, n) without taking into account the hydrogen injection; the present invention assumes that the influence of temperature change on gas flow is negligible, that is, T0 is the ambient temperature.

[0049] Considering that the gap between gas molecules is large, the actual pressure has little effect on the compressibility coefficient, so the average value of the upper and lower limits of the pressure at each node is used. Instead of the actual pressure, Since the hydrogen content in pipeline is relatively small, the change of compression coefficient with hydrogen content ratio can be ignored. The absolute temperature T of natural gas is used. c NG and absolute pressure P cNG Instead of the absolute temperature and absolute pressure of the mixed gas, the compression coefficient of the mixed gas and natural gas is equal, that is, Since the amount of hydrogen added is relatively small and has little effect on the friction coefficient of the pipeline, this effect is ignored, that is, the friction coefficient is only related to the pipeline diameter. Then the expression of the mixed gas compressibility factor is:

[0050]

[0051] In formula (20): Indicates the mixed gas compressibility factor; T c NG represents absolute temperature; Represents the average value of the upper and lower limits of pressure at each node; represents the average value of the upper and lower limits of the pressure at node n; P c NG represents absolute pressure;

[0052]

[0053] In formula (21): D mn is the diameter of the pipe (m,n); Indicates D mn 1 / 6 of the power.

[0054] Specific gas constant of a gas mixture It is related to the hydrogen doping ratio and the molar coefficient of each gas component, and its expression is shown in formula (22):

[0055]

[0056] In formula (22), r is the universal gas constant, is the molar coefficient of the mixed gas in the pipeline (m,n); M NG 、M H2 are the molar coefficients of natural gas and hydrogen, respectively.

[0057] Density of the gas mixture is a function of hydrogen doping ratio, and its expression is:

[0058]

[0059] In formula (24), ρ NG , H2 are the standard densities of natural gas and hydrogen respectively.

[0060]

[0061] The natural gas steady-state model includes not only the natural gas flow balance, equation (24), but also equations (8), (10), (11), and (12).

[0062] In step 2, the nonlinear terms of the variable hydrogen blending ratio natural gas hydrogen blending transportation model are formula (3), formula (4), formula (5), formula (8), and formula (17). Among them, formula (3), formula (4), and formula (5) all contain bilinear terms, which can be directly linearized using the piecewise McCormick technology. Taking formula (3) as an example, the node outflow volume flow rate is selected As the variable to be divided, it can be transformed into:

[0063]

[0064] In formula (25): S is the number of segments; and are the minimum and maximum values ​​of hydrogen doping ratio respectively; and are the minimum and maximum values ​​of the mixed gas volume flow rate out of node m, respectively; and are the minimum and maximum values ​​of the volume flow rate of the mixed gas flowing out of node m in the sth segment respectively; is the value of the mixed gas volume flow rate of the sth segment outflowing from node m; ω m,s,t is the value of the hydrogen doping ratio of node m in the sth segment at time t; is the auxiliary 0-1 variable introduced; is the volume flow rate of the mixed gas flowing out of node m at time t;

[0065] Equation (8) and Equation (17) contain highly nonlinear terms, which need to be replaced by equivalent terms first, and then transformed into linear terms using the piecewise linearization method and the piecewise McCormick technique. Taking Equation (8) as an example, the equivalent terms are first replaced as follows:

[0066]

[0067] In the formula, χ mn,t , K m,t , K n,t They are all introduced intermediate auxiliary variables.

[0068] Among them, equation (27) and equation (28) can be linearized using the piecewise McCormick technique. Equations (29), (30) and (31) are obviously monotonically increasing functions and can be linearized using the piecewise linearization method. Taking equation (29) as an example, it can be converted into:

[0069]

[0070] In formula (32), Np is the total number of segments; is the slope of the volume flow rate of the pipeline (m,n) in the kth segment; is the actual volume flow rate of the kth segment of the pipeline (m,n) in time period t; Minimum volume flow rate of pipe (m,n); is the maximum volume flow rate of the pipeline (m,n) in the kth segment.

[0071] Weymouth coefficient of pipe mn can be restated as:

[0072]

[0073] In formula (33), X1, X2, and X3 are introduced intermediate auxiliary variables; X1 is a constant, and X2 and X3 are both about ω m,t A positive increasing function; ρ NG , H2 are the standard densities of natural gas and hydrogen respectively. It is also a positive increasing function and can be linearized using piecewise linearization. From then on, all nonlinear terms are linearized.

[0074] In step 3, the long-tube trailer hydrogen transportation model established includes trailer operation constraints and hydrogen interaction constraints between the trailer and each hydrogen energy user. The trailer operation constraint expression is:

[0075]

[0076] In formula (34), Ω RN Represents the set of nodes in the traffic network; Ω Road Represents a set of traffic network paths. c,m,t represents the parking state of trailer c at time t, μ c,m,t = 1, at time t, trailer c is located at node m of the traffic network; v c,mn,t represents the driving state of trailer c at time t, v c,mn,t = 1, at time t, trailer c is located on the traffic network path mn;

[0077] η c,m,t -κ c,m,t =μ c,m,t -μ c,m,t-1 (35);

[0078] In formula (35), η c,m,t is the state variable of trailer c arriving at node m at time t; κ c,m,t is the state variable of trailer c leaving node m at time t; μ c,m,t-1 represents the parking state of trailer c at time t-1;

[0079]

[0080] In the above formula, η c,n,t is the state variable of trailer c arriving at node n at time t; θ c,n,t is the auxiliary 0-1 variable introduced;

[0081]

[0082] In formula (39), η c,m,t is the state variable of trailer c arriving at node m at time t; θ c,m,t is the auxiliary 0-1 variable introduced;

[0083]

[0084] In formula (40), v c,mn,t-1 represents the driving state of trailer c at time t-1; Indicates tt min The driving status of trailer c at the moment;

[0085] t min =[L mn / V c ] + (41);

[0086] In formula (41), define [i] + is the rounding symbol, t min V is the shortest time required for the trailer to pass through the path mn; c is the average speed of the trailer; L mn Represents the length of the road between node m and node n.

[0087] The hydrogen interaction constraints between the trailer and each hydrogen energy user are:

[0088]

[0089] In the formula, and are the mass of hydrogen transported by trailer c at time t and t-1 respectively; is the maximum mass of hydrogen that trailer c can transport; is the mass of hydrogen loaded by trailer c from hydrogen energy users at node m at time t; is the maximum mass of hydrogen that trailer c can load from hydrogen energy users at node m in one hour.

[0090] In step 4, the power system related constraints include power balance constraints, unit start and stop constraints, active upper and lower limit constraints, and can provide spare capacity constraints, ramp constraints and line transmission capacity constraints, and the expressions are as follows:

[0091]

[0092] In formula (45), N g 、N w 、N d 、N H2 They are a collection of conventional units, wind farms, power loads, and electrolyzers; is the predicted power generation of unit i during period t; Forecast output for wind farms; For the power load; is the power input to the electrolyzer; t represents a time period; N t represents the set of time periods t.

[0093]

[0094] In formula (46), are the start and stop state variables of unit i during period t; u i,t is the operating state variable of unit i during period t;

[0095]

[0096] In formula (47), T i on The minimum start and stop time for conventional units;

[0097]

[0098] In formula (48), T i off is the minimum shutdown time of conventional units;

[0099]

[0100] In the above formula, and are the predicted power generation of unit i at time periods t and t-1 respectively; and are the downward reserve capacity provided by unit i during periods t and t-1 respectively; and are the upward reserve capacity provided by unit i during periods t and t-1 respectively; is the predicted power generation of wind farm i during period t; is the active load of wind farm i; is the input power of electrolytic cell i during period t. i G,min , P i G,max are the minimum and maximum output of conventional units respectively; Ri UP,max , R i DN,max They are the maximum upward and downward reserve capacities that conventional units can provide, respectively; They are respectively the maximum upward and downward climbing powers of conventional units; is the maximum transmission capacity of line ij; matrix M ij is the transfer power matrix between line power flow and node power injection; matrix A G , A W , A L , A H2 They are bus-generator, bus-wind farm, bus load association matrix, and bus-electrolyzer association matrix respectively.

[0101] The power system is coupled to the natural gas system through the electrolyzer and hydrogen storage tank in the hydrogen production plant, with the following constraints:

[0102]

[0103] In the above formula: η p2h is the electricity-to-hydrogen conversion efficiency of the electrolyzer; H v is the lower heating value of hydrogen; and are the hydrogen storage capacity of the gas tank at time periods t and t-1 respectively; and are the minimum and maximum hydrogen storage capacity of the gas tank respectively; is the output hydrogen volume of electrolyzer i during period t. is the volume flow rate of hydrogen injected at node i at time t.

[0104] The coupling between the natural gas system and the road system is achieved through the hydrogen balance, which is expressed as:

[0105]

[0106] In the formula, is the hydrogen load; is the amount of hydrogen load removal; Ω HT for the collection of trailers; represents the amount of hydrogen that the tube trailer c interacts with the hydrogen energy user at the natural gas system node m during period t; represents the hydrogen load of node m during period t.

[0107] The dispatch model of the electric-hydrogen integrated energy system based on pipeline-road coordinated hydrogen transportation aims to minimize the total operating cost, including: the power system operating cost F om , Trailer hydrogen transportation cost F HT , whose expression is:

[0108] minF=F om +F HT +F cut (58);

[0109] Where, F represents the total operating cost of EHIES; F om 、F HT and F cut They are the power system operating cost, the hydrogen transportation cost by long-tube trailer and the hydrogen load removal cost.

[0110]

[0111]

[0112] In the above formula, are the unit power generation cost, startup cost coefficient, shutdown cost coefficient, and no-load cost coefficient of unit i respectively; is the upper and lower reserve capacity cost coefficient of unit i; c fix is the fixed cost of HT per hour; c cha Ω is the driving cost per kilometer of HT; DH2 is the hydrogen load set. N t represents the set of time periods t; Ω Road Represents a set of traffic network paths; represents the cost coefficient of hydrogen load removal at node m; Represents the amount of hydrogen load removal at node m.

[0113] In step 5, first, define the actual distribution P of wind power output and the empirical distribution The Wasserstein distance between:

[0114]

[0115] In the formula, Represents the true distribution P and the empirical distribution The Wasserstein distance between 2 is the support set of the random variable ξ; represents ξ and The first-order norm between ; Denotes ξ and The joint distribution of ; inf is the infimum function.

[0116] Then construct the fuzzy set Y based on Wasserstein distance:

[0117]

[0118] Where: Ο(Ψ) represents the set of all values ​​of the true distribution P in the support set; ρ is the radius of the Wasserstein ball.

[0119] Then, the conventional unit dispatch capacity cost and related constraints are added to the objective function (54):

[0120]

[0121] In formula (64): Y t The fuzzy set representing the wind power error distribution in period t; Cost coefficient for adjusting active output for conventional units; Active power output adjusted for conventional units; express expected value.

[0122]

[0123] In formula (65): express The probability of this event; α is the allowed constraint violation probability.

[0124]

[0125]

[0126] Where: It is the actual active power output of conventional units; The actual output of wind farms;

[0127] Finally, the nonlinear term in the objective function (64) and the opportunity constraint in the constraint condition are transformed: (65)-(68), where the opportunity constraint takes (65) as an example:

[0128] First of all, Make the following equivalent substitutions:

[0129]

[0130] In the formula, ξ i,t represents the prediction error of wind farm i at time t; τ t represents the total wind power prediction error; I is a unit column vector; the superscript T represents the transpose of the matrix; β i,t Indicates the error ratio of each unit regulation.

[0131] The objective function in the distributed robust scheduling model of the electric-hydrogen integrated energy system of the present invention can be re-expressed as:

[0132]

[0133] In formula (72), Y t The fuzzy set representing the wind power error distribution in period t; express expected value.

[0134] make Under the fuzzy set Y, we can is equivalent to:

[0135]

[0136]

[0137] In the above formula, K is the number of sample sets; represents the kth sample; λ, ε k , γ k ,ω 1,i,t , 1,i,t , ε 1,i,t,n , γ 11,i,t,n , γ 12,i,t,n is an auxiliary variable introduced and has no practical significance; G and s are known matrices and vectors related to setting boundary conditions; ρ is the radius of the Wasserstein sphere; is an intermediate auxiliary matrix; is the auxiliary matrix γ k The transpose of N k Represents a collection of samples; Indicates that λ is a real number; represents a positive, real vector of length 2K. represents the kth sample in time period t; G T Represents the transpose of the constant matrix G.

[0138] All nonlinear terms in the distributed robust model of the electric-hydrogen integrated energy system are linearized. The converted mixed integer programming model can be solved directly on the MATLAB platform through the CPLEX commercial solver. The solution includes the following steps:

[0139] Step 1: Program the input parameters of the mixed integer programming model on the MATLAB platform;

[0140] Step 2: Define the decision variables of the mixed integer programming model on the MATLAB platform;

[0141] Step 3: Program the objective function of the mixed integer programming model on the MATLAB platform;

[0142] Step 4: Program the constraints of the mixed integer programming model on the MATLAB platform;

[0143] Step 5: Call the CPLEX solver on the MATLAB platform to solve the mixed integer programming model.

[0144] Step 6: Output relevant parameters of the power system, natural gas system and road system on the MATLAB platform.

[0145] The present invention provides a method for dispatching an electric-hydrogen integrated energy system based on pipeline-road coordinated hydrogen transportation, and the technical effects are as follows:

[0146] 1) The present invention introduces pipeline storage characteristics into the traditional pipeline hydrogen transportation model, so that the hydrogen storage capacity of the pipeline is effectively simulated, thereby more realistically reflecting the operational flexibility of the pipeline under different loads and the changing laws of hydrogen transportation.

[0147] 2) In the scheduling method of the present invention, for problems that are difficult to solve with traditional models, the model is linearized to greatly reduce the complexity of the calculation and effectively improve the solution speed.

[0148] 3) The present invention develops a special long-tube trailer hydrogen transportation model, which can accurately simulate the spatiotemporal dynamic characteristics of the long-tube trailer during transportation.

[0149] 4) The present invention achieves multi-system coordinated operation by coupling the power system, the natural gas system and the road transportation system, thereby improving the renewable energy absorption capacity and the economy of the system.

[0150] 5) The distributed robust model of the present invention ensures that the system can maintain efficient and stable operation in different situations by modeling the uncertain parameters in the system and performing robust optimization, thereby balancing the relationship between economy and reliability. This method not only improves the anti-interference ability of the system, but also ensures that the system can still operate stably under extreme conditions, avoiding potential risks caused by uncertain factors. BRIEF DESCRIPTION OF THE DRAWINGS

[0151] The present invention will be further described below in conjunction with the accompanying drawings and embodiments:

[0152] Figure 1 This is the operational framework diagram of the electric-hydrogen integrated energy system.

[0153] Figure 2 This is the topological structure diagram of the electric-hydrogen integrated energy system.

[0154] Figure 3 This is the scheduling result diagram of hydrogen production, injection and storage capacity of the hydrogen production plant.

[0155] Figure 4 This is the diagram of load hydrogen blending ratio and scheduling result hydrogen blending ratio.

[0156] Figure 5This is the hydrogen doping ratio error diagram.

[0157] Figure 6 Schematic diagram of the effect of the number of segments on model error and solution time.

[0158] Figure 7 Flowchart for solving the mixed integer programming model. DETAILED DESCRIPTION

[0159] The method for dispatching an electric-hydrogen integrated energy system based on pipeline-road coordinated hydrogen transportation includes the following steps:

[0160] Step 1: Based on the steady-state model of the natural gas pipeline, a natural gas hydrogen blending transportation model with variable hydrogen blending ratio based on flow balance is established by taking into account hydrogen injection and pipeline storage.

[0161] Step 2: Linearize the variable hydrogen blending ratio natural gas hydrogen blending transportation model in step 1 based on piecewise linearization, piecewise McCormick technology and equivalent substitution method, and establish a variable hydrogen blending ratio natural gas hydrogen blending transportation linear model;

[0162] Step 3: Establish a long-tube trailer hydrogen transportation model by simulating the spatiotemporal dynamics of the long-tube trailer;

[0163] Step 4: Considering the constraints related to the power system and the coupling constraints between the natural gas system and the road system, as well as between the power system and the natural gas system, and combining the variable hydrogen blending ratio natural gas hydrogen blending transportation linear model in step 2 and the long tube trailer hydrogen transportation model in step 3, establish an electric-hydrogen integrated energy system scheduling model based on pipeline-road coordinated hydrogen transportation;

[0164] Step 5: Based on the dispatch model of the electric-hydrogen integrated energy system in step 4, considering the uncertainty of wind power, a distributed robust dispatch model of the electric-hydrogen integrated energy system is established. Then, the opportunity constraints in the model are re-expressed as linear constraints. Finally, the GUROBI solver is called on the MATLAB platform to solve the converted linear model of the electric-hydrogen integrated energy system dispatch.

[0165] The present invention proposes a pipeline-road coordinated hydrogen transportation mode, which can economically and flexibly transport hydrogen from a hydrogen production plant to various hydrogen energy users, and coordinately optimize the power system, natural gas system and road system, which can effectively improve the renewable energy consumption and system economy.

[0166] In step 1, the natural gas hydrogen blending transportation model with variable hydrogen blending ratio based on flow balance is established as follows:

[0167]

[0168] In formula (1): Ω SG is the natural gas source point set; ΩSH is the set of hydrogen source points; Ω in is the set of pipes flowing into node m; Ω out is the set of pipes that flow out of node m; are the volume flow rates of natural gas and hydrogen injected at node m at time t, respectively; are the volume flow rates of the mixed gas flowing out of the pipeline and flowing into the node m at time t, respectively; is the volume flow rate of the mixed gas flowing out of node m at time t;

[0169]

[0170] In formula (2): are the volume flow rates of hydrogen flowing out of the pipeline and flowing into node m at time t, respectively; is the volume flow rate of the mixed gas flowing out of node m at time t;

[0171]

[0172] In formula (3): m,t is the volume ratio of hydrogen in the mixed gas at node m at time t;

[0173]

[0174] In formula (4): is the natural gas load at node m at time t;

[0175]

[0176] In formula (5): mn,t is the volume ratio of hydrogen in the mixed gas in the pipeline (m,n) at time t;

[0177]

[0178] ω m,t =ω mn,t (7);

[0179]

[0180] In formula (8): is the average volume flow rate of the pipeline (m,n) at time t; p m,t and p n,t are the pressures at nodes m and n at time t respectively; is the Weymouth coefficient of the pipe (m,n);

[0181]

[0182]

[0183] In formula (10): and are the upper and lower limits of the hourly volume flow rate of the pipeline (m,n);

[0184]

[0185] In formula (11): is the maximum pressure ratio at both ends of the pipeline; p m,t and p n,t are the pressures at nodes m and n at time t respectively;

[0186]

[0187] In formula (12): The minimum allowable pressure of pipeline nodes; The maximum allowable pressure at a pipeline node.

[0188]

[0189] In formula (13): B mn,t and B mn,t-1 is the pipeline inventory of pipeline (m,n) at time t and t-1;

[0190]

[0191] In formula (14): is the pipe storage coefficient of the pipeline (m,n);

[0192]

[0193] In formula (15): Ω pipe is the collection of all natural gas pipelines; B min Indicates the minimum value of the sum of all pipeline inventories.

[0194]

[0195] In formula (16): and is the maximum and minimum volume flow rate of natural gas that can be injected into node m; is the volume flow rate of natural gas injected at node m at time t.

[0196]

[0197] In formula (17): and is the maximum and minimum volume flow rate of natural gas that can be injected into node m; is the volume flow rate of hydrogen injected at node m at time t.

[0198] They are the Weymouth coefficient and pipe storage coefficient of the pipeline (m,n), and their expressions are as follows:

[0199]

[0200] In formula (18) and formula (19), D mn , l mn 、f mn are the diameter, length and friction coefficient of the pipe (m,n) respectively; π is the circumference of a circle; and are the specific gas constant and density of the mixed gas respectively; The compression factor and the density of the mixed gas when hydrogen is injected into the pipeline (m, n); and are the specific gas constant and density of natural gas, respectively; is the compression coefficient of the pipeline (m, n) without taking into account the hydrogen injection; the present invention assumes that the influence of temperature change on gas flow is negligible, that is, T0 is the ambient temperature.

[0201] Considering that the gap between gas molecules is large, the actual pressure has little effect on the compressibility coefficient, so the average value of the upper and lower limits of the pressure at each node is used. Instead of the actual pressure, Since the hydrogen content in pipeline is relatively small, the change of compression coefficient with hydrogen content ratio can be ignored. The absolute temperature T of natural gas is used. c NG and absolute pressure P c NG Instead of the absolute temperature and absolute pressure of the mixed gas, the compression coefficient of the mixed gas and natural gas is equal, that is, Since the amount of hydrogen added is relatively small and has little effect on the friction coefficient of the pipeline, this effect is ignored, that is, the friction coefficient is only related to the pipeline diameter. Then the expression of the mixed gas compressibility factor is:

[0202]

[0203] In formula (20): Indicates the mixed gas compressibility factor; T c NG represents absolute temperature; Represents the average value of the upper and lower limits of pressure at each node; represents the average value of the upper and lower limits of the pressure at node n; P c NG represents absolute pressure;

[0204]

[0205] In formula (21): D mn is the diameter of the pipe (m,n); Indicates D mn 1 / 6 of the power.

[0206] Specific gas constant of a gas mixture It is related to the hydrogen doping ratio and the molar coefficient of each gas component, and its expression is shown in formula (22):

[0207]

[0208] In formula (22), r is the universal gas constant, is the molar coefficient of the mixed gas in the pipeline (m,n); M NG 、M H2 are the molar coefficients of natural gas and hydrogen, respectively.

[0209] Density of the gas mixture is a function of hydrogen doping ratio, and its expression is:

[0210]

[0211] In formula (24), ρ NG , H2 are the standard densities of natural gas and hydrogen respectively.

[0212]

[0213] The natural gas steady-state model includes not only the natural gas flow balance, equation (24), but also equations (8), (10), (11), and (12).

[0214] In step 2, the nonlinear terms of the variable hydrogen blending ratio natural gas hydrogen blending transportation model are formula (3), formula (4), formula (5), formula (8), and formula (17). Among them, formula (3), formula (4), and formula (5) all contain bilinear terms, which can be directly linearized using the piecewise McCormick technology. Taking formula (3) as an example, the node outflow volume flow rate is selected As the variable to be divided, it can be transformed into:

[0215]

[0216] In formula (25): S is the number of segments; and are the minimum and maximum values ​​of hydrogen doping ratio respectively; and are the minimum and maximum values ​​of the mixed gas volume flow rate out of node m, respectively; and are the minimum and maximum values ​​of the volume flow rate of the mixed gas flowing out of node m in the sth segment respectively; is the value of the mixed gas volume flow rate of the sth segment outflowing from node m; ω m,s,t is the value of the hydrogen doping ratio of node m in the sth segment at time t; is the auxiliary 0-1 variable introduced; is the volume flow rate of the mixed gas flowing out of node m at time t;

[0217] Equation (8) and Equation (17) contain highly nonlinear terms, which need to be replaced by equivalent terms first, and then transformed into linear terms using the piecewise linearization method and the piecewise McCormick technique. Taking Equation (8) as an example, the equivalent terms are first replaced as follows:

[0218]

[0219]

[0220] In the formula, χ mn,t , K m,t , K n,t They are all introduced intermediate auxiliary variables.

[0221] Among them, equation (27) and equation (28) can be linearized using the piecewise McCormick technique. Equations (29), (30) and (31) are obviously monotonically increasing functions and can be linearized using the piecewise linearization method. Taking equation (29) as an example, it can be converted into:

[0222]

[0223] In formula (32), Np is the total number of segments; is the slope of the volume flow rate of the pipeline (m,n) in the kth segment; is the actual volume flow rate of the kth segment of the pipeline (m,n) in time period t; Minimum volume flow rate of pipe (m,n); is the maximum volume flow rate of the pipeline (m,n) in the kth segment.

[0224] Weymouth coefficient of pipe mn can be restated as:

[0225]

[0226] In formula (33), X1, X2, and X3 are introduced intermediate auxiliary variables; X1 is a constant, and X2 and X3 are both about ω m,t A positive increasing function; ρ NG , H2are the standard densities of natural gas and hydrogen respectively. It is also a positive increasing function and can be linearized using piecewise linearization. From then on, all nonlinear terms are linearized.

[0227] In step 3, the long-tube trailer hydrogen transportation model established includes trailer operation constraints and hydrogen interaction constraints between the trailer and each hydrogen energy user. The trailer operation constraint expression is:

[0228]

[0229] In formula (34), Ω RN Represents the set of nodes in the traffic network; Ω Road Represents a set of traffic network paths. c,m,t represents the parking state of trailer c at time t, μ c,m,t = 1, at time t, trailer c is located at node m of the traffic network; v c,mn,t represents the driving state of trailer c at time t, v c,mn,t = 1, at time t, trailer c is located on the traffic network path mn;

[0230] η c,m,t -κ c,m,t =μ c,m,t -μ c,m,t-1 (35);

[0231] In formula (35), η c,m,t is the state variable of trailer c arriving at node m at time t; κ c,m,t is the state variable of trailer c leaving node m at time t; μ c,m,t-1 represents the parking state of trailer c at time t-1;

[0232]

[0233] In the above formula, η c,n,t is the state variable of trailer c arriving at node n at time t; θ c,n,t is the auxiliary 0-1 variable introduced;

[0234]

[0235] In formula (39), η c,m,t is the state variable of trailer c arriving at node m at time t; θ c,m,t is the auxiliary 0-1 variable introduced;

[0236]

[0237] In formula (40), v c,mn,t-1 represents the driving state of trailer c at time t-1; Indicates ttmin The driving status of trailer c at the moment;

[0238] t min =[L mn / V c ] + (41);

[0239] In formula (41), define [i] + is the rounding symbol, t min V is the shortest time required for the trailer to pass through the path mn; c is the average speed of the trailer; L mn Represents the length of the road between node m and node n.

[0240] The hydrogen interaction constraints between the trailer and each hydrogen energy user are:

[0241]

[0242] In the formula, and are the mass of hydrogen transported by trailer c at time t and t-1 respectively; is the maximum mass of hydrogen that trailer c can transport; is the mass of hydrogen loaded by trailer c from hydrogen energy users at node m at time t; is the maximum mass of hydrogen that trailer c can load from hydrogen energy users at node m in one hour.

[0243] In step 4, the power system related constraints include power balance constraints, unit start and stop constraints, active upper and lower limit constraints, and can provide spare capacity constraints, ramp constraints and line transmission capacity constraints, and the expressions are as follows:

[0244]

[0245] In formula (45), N g 、N w 、N d 、N H2 They are a collection of conventional units, wind farms, power loads, and electrolyzers; is the predicted power generation of unit i during period t; Forecast output for wind farms; For the power load; is the power input to the electrolyzer; t represents a time period; N t represents the set of time periods t.

[0246]

[0247] In formula (46), are the start and stop state variables of unit i during period t; u i,t is the operating state variable of unit i during period t;

[0248]

[0249] In formula (47), T i on The minimum start and stop time for conventional units;

[0250]

[0251] In formula (48), T i off is the minimum shutdown time of conventional units;

[0252]

[0253]

[0254] In the above formula, and are the predicted power generation of unit i at time periods t and t-1 respectively; and are the downward reserve capacity provided by unit i during periods t and t-1 respectively; and are the upward reserve capacity provided by unit i during periods t and t-1 respectively; is the predicted power generation of wind farm i during period t; is the active load of wind farm i; is the input power of electrolytic cell i during period t. i G,min , P i G,max are the minimum and maximum output of conventional units respectively; R i UP,max , R i DN,max They are the maximum upward and downward reserve capacities that conventional units can provide, respectively; They are respectively the maximum upward and downward climbing powers of conventional units; is the maximum transmission capacity of line ij; matrix M ij is the transfer power matrix between line power flow and node power injection; matrix A G , A W , A L , A H2 They are bus-generator, bus-wind farm, bus load association matrix, and bus-electrolyzer association matrix respectively.

[0255] The power system is coupled to the natural gas system through the electrolyzer and hydrogen storage tank in the hydrogen production plant, with the following constraints:

[0256]

[0257] In the above formula: η p2h is the electricity-to-hydrogen conversion efficiency of the electrolyzer; H v is the lower heating value of hydrogen; and are the hydrogen storage capacity of the gas tank at time periods t and t-1 respectively; and are the minimum and maximum hydrogen storage capacity of the gas tank respectively; is the output hydrogen volume of electrolyzer i during period t. is the volume flow rate of hydrogen injected at node i at time t.

[0258] The coupling between the natural gas system and the road system is achieved through the hydrogen balance, which is expressed as:

[0259]

[0260] In the formula, is the hydrogen load; is the amount of hydrogen load removal; Ω HT for the collection of trailers; represents the amount of hydrogen that the tube trailer c interacts with the hydrogen energy user at the natural gas system node m during period t; represents the hydrogen load of node m during period t.

[0261] The dispatch model of the electric-hydrogen integrated energy system based on pipeline-road coordinated hydrogen transportation aims to minimize the total operating cost, including: the power system operating cost F om , Trailer hydrogen transportation cost F HT , whose expression is:

[0262] minF=F om +F HT +F cut (58);

[0263] Where, F represents the total operating cost of EHIES; F om 、F HT and F cut They are the power system operating cost, the hydrogen transportation cost by long-tube trailer and the hydrogen load removal cost.

[0264]

[0265] In the above formula, are the unit power generation cost, startup cost coefficient, shutdown cost coefficient, and no-load cost coefficient of unit i respectively; is the upper and lower reserve capacity cost coefficient of unit i; c fix is the fixed cost of HT per hour; c cha is the driving cost per kilometer of HT; Ω DH2 is the hydrogen load set. N t represents the set of time periods t; Ω Road Represents a set of traffic network paths; represents the cost coefficient of hydrogen load removal at node m; Represents the amount of hydrogen load removal at node m.

[0266] In step 5, first, define the actual distribution P of wind power output and the empirical distribution The Wasserstein distance between:

[0267]

[0268] In the formula, Represents the true distribution P and the empirical distribution The Wasserstein distance between 2 is the support set of the random variable ξ; represents ξ and The first-order norm between ; represents ξ and The joint distribution of ; inf is the infimum function.

[0269] Then construct the fuzzy set Y based on Wasserstein distance:

[0270]

[0271] Where: Ο(Ψ) represents the set of all values ​​of the true distribution P in the support set; ρ is the radius of the Wasserstein ball.

[0272] Then, the conventional unit dispatch capacity cost and related constraints are added to the objective function (54):

[0273]

[0274] In formula (64): Y t The fuzzy set representing the wind power error distribution in period t; Cost coefficient for adjusting active output for conventional units; Active power output adjusted for conventional units; express expected value.

[0275]

[0276] In formula (65): express The probability of this event; α is the allowed constraint violation probability.

[0277]

[0278] Where: It is the actual active power output of conventional units; The actual output of wind farms;

[0279] Finally, the nonlinear term in the objective function (64) and the chance constraint in the constraint condition are transformed into (65)-(68), where the chance constraint is taken as an example.

[0280] First of all, Make the following equivalent substitutions:

[0281]

[0282] In the formula, ξ i,t represents the prediction error of wind farm i at time t; τ t represents the total wind power prediction error; I is a unit column vector; the superscript T represents the transpose of the matrix; β i,t Indicates the error ratio of each unit regulation.

[0283] The objective function in the distributed robust scheduling model of the electric-hydrogen integrated energy system of the present invention can be re-expressed as:

[0284]

[0285] In formula (72), Y t The fuzzy set representing the wind power error distribution in period t; express expected value.

[0286] make Under the fuzzy set Y, we can is equivalent to:

[0287]

[0288] In the above formula, K is the number of sample sets; represents the kth sample; λ, ε k , γ k ,ω 1,i,t , 1,i,t , ε 1,i,t,n , γ 11,i,t,n , γ 12,i,t,n is an auxiliary variable introduced and has no practical significance; G and s are known matrices and vectors related to setting boundary conditions; ρ is the radius of the Wasserstein sphere; is an intermediate auxiliary matrix; is the auxiliary matrix γ k The transpose of N k Represents a collection of samples; Indicates that λ is a real number; represents a positive, real vector of length 2K. represents the kth sample in time period t; G T Represents the transpose of the constant matrix G.

[0289] All nonlinear terms in the distributed robust model of the electric-hydrogen integrated energy system are linearized. The converted mixed integer programming model can be solved directly on the MATLAB platform through the CPLEX commercial solver. The solution includes the following steps:

[0290] Step 1: Program the input parameters of the mixed integer programming model on the MATLAB platform;

[0291] Step 2: Define the decision variables of the mixed integer programming model on the MATLAB platform;

[0292] Step 3: Program the objective function of the mixed integer programming model on the MATLAB platform;

[0293] Step 4: Program the constraints of the mixed integer programming model on the MATLAB platform;

[0294] Step 5: Call the CPLEX solver on the MATLAB platform to solve the mixed integer programming model.

[0295] Step 6: Output relevant parameters of the power system, natural gas system and road system on the MATLAB platform.

[0296] Example:

[0297] The present invention is based on the improved IEEE 5-node power network, 8-node natural gas system and 5-node transportation network, and constructs an EHIES (energy system) based on pipeline-road coordinated hydrogen transportation. In this power system, a hydrogen production plant is set up at node 5, and the produced hydrogen is injected through node 2 of the natural gas network. A wind farm with a capacity of 180MW is established at node 5, and conventional power plant 1, conventional power plant 2 and conventional power plant 3 are set up at nodes 1, 3 and 4 respectively. The natural gas source is connected to the natural gas system node 1, and hydrogen refueling stations are established at nodes 5, 6, 7 and 8 respectively. The topological structure diagram of the entire system is shown as follows: Figure 2 In this system, the daily gas requirement is 5.3830 Mm 3 Natural gas and 0.6355Mm 3Hydrogen. The upper limit of hydrogen blending ratio in natural gas pipeline is 20%. At the same time, the hydrogen load within 24 hours, the predicted output curve of wind farm and the total power load within one day are shown in Appendix C. The system is also equipped with 6 dispatchable hydrogen transport vehicles, each with a maximum hydrogen transport capacity of 200kg, a driving speed of 50km / h, a fixed cost of 600 yuan / vehicle, and a transportation cost coefficient of 8 yuan / km. The number of segments of the improved McCormick technique and piecewise linearization are both set to 10. The simulation process is carried out on the MATLAB2022a platform and solved using the Yalmip toolbox and Gurobi solver.

[0298] 1: Analysis of the dispatch of the electricity-hydrogen integrated energy system:

[0299] First, without considering the uncertainty of wind power output, the scheduling results of the model proposed in this invention are analyzed. Taking the hydrogen production plant as the object, the hydrogen production, hydrogen energy reserves and hydrogen injection amount of the hydrogen production plant are as follows: Figure 3 As shown. At 1:00-6:00, 9:00, 12:00-13:00, and 23:00-24:00, the power load and load are relatively small, the wind power output is large, the electrolyzer input power is large, and the hydrogen energy reserves in the hydrogen production plant increase rapidly. At other times, especially during the peak period of hydrogen load from 18:00 to 19:00, the hydrogen production is much less than the hydrogen injection, and the hydrogen energy reserves in the hydrogen production plant decrease rapidly. This shows that electric hydrogen production can well absorb wind power and can effectively eliminate the power imbalance caused by the anti-peaking characteristics of wind power output and the climbing constraints of conventional units.

[0300] The required gas load hydrogen blending ratio and dispatching results of each gas load node are as follows: Figure 4 As shown, the ratio of hydrogen to natural gas required at each gas load node is different, and the ratio of gas required at the same load node is also different at different times. If hydrogen is transported only by natural gas pipelines, more load shedding will occur. The total hydrogen load shedding per day reaches 12.866t, accounting for 22.68% of the total hydrogen load. After adopting the hydrogen transportation mode proposed by the present invention, the load shedding amount drops to 0. This is because trailers can realize hydrogen transportation between various hydrogen loads and make up for the supply and demand imbalance caused by pipeline transportation. As can be seen from Table 1, the total cost of scenario 1 is 3.50% less than that of scenario 2. This is because the amount of hydrogen load shedding in scenario 2 is large and the load shedding cost is high. The transportation cost and the number of trailers required in scenario 3 are much larger than those in scenario 1, which shows that trailers are not suitable for scenarios with large-scale hydrogen transportation and long transportation distances. Comparing scenario 1 with scenario 4, the transportation cost of scenario 1 is significantly smaller, and its total cost is 5.95% smaller than that of scenario 4, which shows the importance of optimizing scheduling by using the hydrogen blending ratio as a variable.

[0301] Table 1 Comparison of hydrogen load removal percentage and total cost in four scenarios

[0302]

[0303]

[0304] 2: Model error analysis:

[0305] In order to analyze the linearization error of the linear quasi-dynamic model proposed in the present invention, with the goal of minimizing the sum of the hydrogen purchase cost and the hydrogen abandonment load penalty cost, the gurobi solver is used on the MATLAB platform to solve the linear model (model one) proposed in Section 2.2 of the present invention, in which the improved McCormick technique and the number of segments of the piecewise linearization are both set to 10; the IPOPT solver is used to solve the proposed nonlinear programming model (model two). The IPOPT solver uses the interior point method. The interior point method is a type of optimization method that searches for the optimal solution by stepwise iteration within the feasible domain, which is different from traditional methods such as the simplex method (usually starting from the boundary of the feasible domain). The interior point method introduces a "barrier function" or "penalty function" to gradually approach the boundary constraints during the iteration process, and finally converge to the optimal solution.

[0306] The solution time of model 1 is 0.23s, and the solution time of model 2 is 18.52s. The hydrogen mixing ratio and error of model 1 and model 2 are shown in Figure 5 As shown, the left vertical axis represents the pipeline hydrogen blending ratio, and the right vertical axis represents the error percentage. The hydrogen blending ratios of the two models are very close, and the error is almost 0 at some moments. The maximum error is obtained at 23:00, which is about 0.95%.

[0307] Figure 6 It represents the influence of the number of segments of the McCormick technique and the piecewise linearization method on the solution accuracy and solution time. The value of S has a greater influence on the model solution accuracy and solution time, while the value of Np has a smaller influence on the model solution accuracy and solution time. When S and Np continue to increase from 10, the solution accuracy is not significantly lower, but the solution time increases exponentially. Therefore, S=10 and Np=10 are better values ​​for the number of segments.

[0308] In summary, the model of the present invention has a smaller error than the nonlinear model, and the solution time is only 1 / 80 of the latter, which greatly improves the solution efficiency. In particular, when solving larger-scale examples, nonlinear models often cannot be solved.

Claims

1. A dispatching method for an electric-hydrogen integrated energy system based on pipeline-road coordinated hydrogen transportation, characterized in that The following steps are involved: Step 1: Based on the steady-state model of the natural gas pipeline, taking into account hydrogen injection and pipeline storage, a natural gas hydrogen blending transportation model with variable hydrogen blending ratio based on flow balance is established; Step 2: Linearize the variable hydrogen blending ratio natural gas hydrogen blending transportation model in step 1 based on piecewise linearization, piecewise McCormick technology and equivalent substitution method, and establish a variable hydrogen blending ratio natural gas hydrogen blending transportation linear model; Step 3: Establish a long-tube trailer hydrogen transportation model by simulating the spatiotemporal dynamics of the long-tube trailer; Step 4: Considering the constraints related to the power system and the coupling constraints between the natural gas system and the road system, as well as between the power system and the natural gas system, and combining the variable hydrogen blending ratio natural gas hydrogen blending transportation linear model in step 2 and the long tube trailer hydrogen transportation model in step 3, establish an electric-hydrogen integrated energy system scheduling model based on pipeline-road coordinated hydrogen transportation; Step 5: Based on the dispatch model of the electric-hydrogen integrated energy system in step 4, considering the uncertainty of wind power, a distributed robust dispatch model of the electric-hydrogen integrated energy system is established, the opportunity constraints in the distributed robust dispatch model are re-expressed as linear constraints, and the converted linear dispatch model of the electric-hydrogen integrated energy system is solved.

2. The method for dispatching an electric-hydrogen integrated energy system based on pipeline-road coordinated hydrogen transportation according to claim 1 is characterized in that: In step 1, the natural gas hydrogen blending transportation model with variable hydrogen blending ratio based on flow balance is established as follows: In formula (1): Ω SG is the natural gas source point set; Ω SH is the set of hydrogen source points; Ω in is the set of pipes flowing into node m; Ω out is the set of pipes that flow out of node m; are the volume flow rates of natural gas and hydrogen injected at node m at time t, respectively; are the volume flow rates of the mixed gas flowing out of the pipeline and flowing into the node m at time t, respectively; is the volume flow rate of the mixed gas flowing out of node m at time t; In formula (2): are the volume flow rates of hydrogen flowing out of the pipeline and flowing into node m at time t, respectively; is the volume flow rate of the mixed gas flowing out of node m at time t; In formula (3): m,t is the volume ratio of hydrogen in the mixed gas at node m at time t; In formula (4): is the natural gas load at node m at time t; In formula (5): mn,t is the volume ratio of hydrogen in the mixed gas in the pipeline (m,n) at time t; oh m,t =ω mn,t (7); In formula (8): is the average volume flow rate of the pipeline (m,n) at time t; p m,t and p n,t are the pressures at nodes m and n at time t respectively; is the Weymouth coefficient of the pipe (m,n); In formula (10): and are the upper and lower limits of the hourly volume flow rate of the pipeline (m,n); In formula (11): is the maximum pressure ratio at both ends of the pipeline; p m,t and p n,t are the pressures at nodes m and n at time t respectively; In formula (12): The minimum allowable pressure of pipeline nodes; Maximum allowable pressure of pipeline nodes; In formula (13): B mn,t and B mn,t-1 is the pipeline inventory of pipeline (m,n) at time t and t-1; In formula (14): is the pipe storage coefficient of the pipeline (m,n); In formula (15): Ω pipe is the collection of all natural gas pipelines; B min It represents the minimum value of the sum of all pipeline stocks; In formula (16): and is the maximum and minimum volume flow rate of natural gas that can be injected into node m; is the volume flow rate of natural gas injected at node m at time t; In formula (17): and is the maximum and minimum volume flow rate of natural gas that can be injected into node m; is the volume flow rate of hydrogen injected at node m at time t; They are the Weymouth coefficient and pipe storage coefficient of the pipeline (m,n), and their expressions are as follows: In formula (18) and formula (19), D mn , l mn 、f mn are the diameter, length and friction coefficient of the pipe (m,n) respectively; π is the circumference of a circle; and are the specific gas constant and density of the mixed gas respectively; The compression factor and the density of the mixed gas when hydrogen is injected into the pipeline (m, n); and are the specific gas constant and density of natural gas, respectively; is the compression coefficient of the pipeline (m, n) without considering the hydrogen injection; T0 is the ambient temperature; The expression of the mixed gas compressibility factor is: In formula (20): Indicates the mixed gas compressibility factor; represents absolute temperature; Represents the average value of the upper and lower limits of pressure at each node; It represents the average value of the upper and lower limits of the pressure at node n; represents absolute pressure; In formula (21): D mn is the diameter of the pipe (m,n); Indicates D mn 1 / 6 of the power; Specific gas constant of a gas mixture It is related to the hydrogen doping ratio and the molar coefficient of each gas component, and its expression is shown in formula (22): In formula (22), r is the universal gas constant, is the molar coefficient of the mixed gas in the pipeline (m,n); M NG 、M H2 are the molar coefficients of natural gas and hydrogen respectively; Density of the gas mixture is a function of hydrogen doping ratio, and its expression is: In formula (24), ρ NG , H2 are the standard densities of natural gas and hydrogen respectively; The natural gas steady-state model includes the natural gas flow balance of equation (24), and also includes equations (8), (10), (11), and (12).

3. The method for dispatching an electric-hydrogen integrated energy system based on pipeline-road coordinated hydrogen transportation according to claim 2 is characterized in that: In step 2, the nonlinear terms of the variable hydrogen blending ratio natural gas hydrogen blending transportation model are formula (3), formula (4), formula (5), formula (8), and formula (17). Among them, formula (3), formula (4), and formula (5) all contain bilinear terms, which can be directly linearized using the piecewise McCormick technology. Based on formula (3), the node outflow volume flow rate is selected As the variable to be divided, it can be transformed into: In formula (25): S is the number of segments; and are the minimum and maximum values ​​of hydrogen doping ratio respectively; and are the minimum and maximum values ​​of the mixed gas volume flow rate out of node m, respectively; and are the minimum and maximum values ​​of the volume flow rate of the mixed gas flowing out of node m in the sth segment respectively; is the value of the mixed gas volume flow rate of the sth segment outflowing from node m; ω m,s,t is the value of the hydrogen doping ratio of node m in the sth segment at time t; is the auxiliary 0-1 variable introduced; is the volume flow rate of the mixed gas flowing out of the node m at time t.

4. The method for dispatching an electric-hydrogen integrated energy system based on pipeline-road coordinated hydrogen transportation according to claim 3 is characterized in that: Formula (8) and Formula (17) contain highly nonlinear terms. They are first replaced by equivalent terms, and then linearly transformed using the piecewise linearization method and the piecewise McCormick technique. Based on Formula (8), they are first replaced by equivalent terms: In the formula, χ mn,t , K m,t , K n,t They are all introduced intermediate auxiliary variables; Among them, equation (27) and equation (28) can be linearized using the piecewise McCormick technique, equation (29), equation (30) and equation (31) are monotonically increasing functions and can be linearized using the piecewise linearization method, and equation (29) can be converted to: In formula (32), Np is the total number of segments; is the slope of the volume flow rate of the pipeline (m,n) in the kth segment; is the actual volume flow rate of the kth segment of the pipeline (m,n) in time period t; Minimum volume flow rate of pipe (m,n); is the maximum volume flow rate of the pipeline (m,n) in the kth segment; Weymouth coefficient of pipe mn can be restated as: In formula (33), X1, X2, and X3 are introduced intermediate auxiliary variables; X1 is a constant, and X2 and X3 are both about ω m,t A positive increasing function; ρ NG , H2 are the standard densities of natural gas and hydrogen respectively; is a positive increasing function and can be linearized using piecewise linearization methods.

5. The method for dispatching an electric-hydrogen integrated energy system based on pipeline-road coordinated hydrogen transportation according to claim 4 is characterized in that: In step 3, the long-tube trailer hydrogen transportation model established includes trailer operation constraints and hydrogen interaction constraints between the trailer and each hydrogen energy user; The trailer operation constraint expression is: In formula (34), Ω RN Represents the set of nodes in the traffic network; Ω Road represents the set of traffic network paths; μ c,m,t represents the parking state of trailer c at time t, μ c,m,t = 1, at time t, trailer c is located at node m of the traffic network; v c,mn,t represents the driving state of trailer c at time t, v c,mn,t = 1, at time t, trailer c is located on the traffic network path mn; or c,m,t -k c,m,t =μ c,m,t -m c,m,t-1 (35); In formula (35), η c,m,t is the state variable of trailer c arriving at node m at time t; κ c,m,t is the state variable of trailer c leaving node m at time t; μ c,m,t-1 represents the parking state of trailer c at time t-1; In the above formula, η c,n,t is the state variable of trailer c arriving at node n at time t; θ c,n,t is the auxiliary 0-1 variable introduced; In formula (39), η c,m,t is the state variable of trailer c arriving at node m at time t; θ c,m,t is the auxiliary 0-1 variable introduced; In formula (40), v c,mn,t-1 represents the driving state of trailer c at time t-1; Indicates tt min The driving status of trailer c at the moment; t min =[L mn / V c ] + (41); In formula (41), define [i] + is the rounding symbol, t min is the shortest time required for the trailer to pass through the path mn; V c is the average speed of the trailer; L mn represents the road length between node m and node n; The hydrogen interaction constraints between the trailer and each hydrogen energy user are: In the formula, and are the mass of hydrogen transported by trailer c at time t and t-1 respectively; is the maximum mass of hydrogen that trailer c can transport; is the mass of hydrogen loaded by trailer c from hydrogen energy users at node m at time t; is the maximum mass of hydrogen that trailer c can load from hydrogen energy users at node m in one hour.

6. The method for dispatching an electric-hydrogen integrated energy system based on pipeline-road coordinated hydrogen transportation according to claim 5 is characterized in that: In step 4, the power system related constraints include power balance constraints, unit start and stop constraints, active upper and lower limit constraints, and can provide spare capacity constraints, ramp constraints and line transmission capacity constraints, and the expressions are as follows: In formula (45), N g 、N w 、N d 、N H2 They are a collection of conventional units, wind farms, power loads, and electrolyzers; is the predicted power generation of unit i during period t; Forecast output for wind farms; For the power load; is the power input to the electrolyzer; t represents a time period; N t represents the set of time periods t; In formula (46), are the start and stop state variables of unit i during period t; u i,t is the operating state variable of unit i during period t; In formula (47), T i on The minimum start and stop time for conventional units; In formula (48), T i off is the minimum shutdown time of conventional units; In the above formula, and are the predicted power generation of unit i at time periods t and t-1 respectively; and are the downward reserve capacity provided by unit i during periods t and t-1 respectively; and are the upward reserve capacity provided by unit i during periods t and t-1 respectively; is the predicted power generation of wind farm i during period t; is the active load of wind farm i; is the input power of electrolytic cell i during period t; P i G,min , P i G,max They are the minimum and maximum outputs of conventional units respectively; They are the maximum upward and downward reserve capacities that conventional units can provide, respectively; They are respectively the maximum upward and downward climbing powers of conventional units; is the maximum transmission capacity of line ij; matrix M ij is the transfer power matrix between line power flow and node power injection; matrix A G , A W , A L , A H2 They are bus-generator, bus-wind farm, bus load association matrix, and bus-electrolyzer association matrix respectively; The power system is coupled to the natural gas system through the electrolyzer and hydrogen storage tank in the hydrogen production plant, with the following constraints: In the above formula: η p2h is the electricity-to-hydrogen conversion efficiency of the electrolyzer; H v is the lower heating value of hydrogen; and are the hydrogen storage capacity of the gas tank at time periods t and t-1 respectively; and are the minimum and maximum hydrogen storage capacity of the gas tank respectively; is the output hydrogen volume of electrolyzer i during period t; is the volume flow rate of hydrogen injected at node i at time t; The coupling between the natural gas system and the road system is achieved through the hydrogen balance, which is expressed as: In the formula, is the hydrogen load; is the amount of hydrogen load removal; Ω HT for the collection of trailers; represents the amount of hydrogen that the tube trailer c interacts with the hydrogen energy user at the natural gas system node m during period t; represents the hydrogen load of node m during period t.

7. The method for dispatching an electric-hydrogen integrated energy system based on pipeline-road coordinated hydrogen transportation according to claim 6 is characterized in that: The dispatch model of the electric-hydrogen integrated energy system based on pipeline-road coordinated hydrogen transportation aims to minimize the total operating cost, including: the power system operating cost F om , Trailer hydrogen transportation cost F HT , whose expression is: minF=F om +F HT +F cut (58); Where, F represents the total operating cost of EHIES; F om 、F HT and F cut They are the power system operation cost, the hydrogen transportation cost by long tube trailer and the hydrogen load removal cost; In the above formula, are the unit power generation cost, startup cost coefficient, shutdown cost coefficient, and no-load cost coefficient of unit i respectively; is the upper and lower reserve capacity cost coefficient of unit i; c fix is the fixed cost of HT per hour; c cha is the driving cost per kilometer of HT; Ω DH2 is the hydrogen load set; N t represents the set of time periods t; Ω Road Represents a set of traffic network paths; represents the cost coefficient of hydrogen load removal at node m; Represents the amount of hydrogen load removal at node m.

8. The method for dispatching an electric-hydrogen integrated energy system based on pipeline-road coordinated hydrogen transportation according to claim 7 is characterized in that: In step 5, first, define the actual distribution P of wind power output and the empirical distribution The Wasserstein distance between: In the formula, Represents the true distribution P and the empirical distribution The Wasserstein distance between 2 is the support set of the random variable ξ; represents ξ and The first-order norm between ; represents ξ and The joint distribution of ; inf is the infimum function; Then construct the fuzzy set Y based on Wasserstein distance: Where: Ο(Ψ) represents the set of all values ​​of the true distribution P in the support set; ρ is the radius of the Wasserstein ball; Then, the conventional unit dispatch capacity cost and related constraints are added to the objective function (54): In formula (64): Y t The fuzzy set representing the wind power error distribution in period t; Cost coefficient for adjusting active output for conventional units; Active power output adjusted for conventional units; express Expected value; In formula (65): express The probability of this event; α is the allowed constraint violation probability; Where: It is the actual active power output of conventional units; The actual output of wind farms; Finally, the nonlinear terms in the objective function (64) and the chance constraints (65)-(68) in the constraints are transformed.

9. The method for dispatching an electric-hydrogen integrated energy system based on pipeline-road coordinated hydrogen transportation according to claim 8 is characterized in that: In the opportunity constraint of formula (65): First of all, Make the following equivalent substitutions: t t =I T x i,t (69); In the formula, ξ i,t represents the prediction error of wind farm i at time t; τ t represents the total wind power prediction error; I is a unit column vector; the superscript T represents the transpose of the matrix; β i,t Indicates the error ratio of each unit adjustment; The objective function in the distributed robust scheduling model of the electric-hydrogen integrated energy system can be re-expressed as: In formula (72), Y t The fuzzy set representing the wind power error distribution in period t; express Expected value; make Under the fuzzy set Y, we can is equivalent to: In the above formula, K is the number of sample sets; represents the kth sample; λ, ε k , γ k ,ω 1,i,t , 1,i,t , ε 1,i,t,n , γ 11,i,t,n , γ 12,i,t,n is an auxiliary variable introduced and has no practical significance; G and s are known matrices and vectors related to setting boundary conditions; ρ is the radius of the Wasserstein sphere; is an intermediate auxiliary matrix; is the auxiliary matrix γ k The transpose of N k Represents a collection of samples; Indicates that λ is a real number; represents a positive real vector of length 2K; represents the kth sample in time period t; G T Represents the transpose of the constant matrix G.

10. The method for dispatching an electric-hydrogen integrated energy system based on pipeline-road coordinated hydrogen transportation according to claim 9 is characterized in that: All nonlinear terms in the distributed robust model of the electric-hydrogen integrated energy system dispatch are linearized, and the converted mixed integer programming model can be directly solved by the CPLEX commercial solver on the MATLAB platform. The solution includes the following steps: Step 1: Program the input parameters of the mixed integer programming model on the MATLAB platform; Step 2: Define the decision variables of the mixed integer programming model on the MATLAB platform; Step 3: Program the objective function of the mixed integer programming model on the MATLAB platform; Step 4: Program the constraints of the mixed integer programming model on the MATLAB platform; Step 5: Call the CPLEX solver on the MATLAB platform to solve the mixed integer programming model; Step 6: Output relevant parameters of the power system, natural gas system and road system on the MATLAB platform.

Images