Power system scheduling method based on multi-time resolution three-layer robustness

Abstract

The invention relates to the technical field of power systems, and particularly discloses a power system scheduling method based on multi-time resolution three-layer robustness, comprising the following steps: constructing an IPHS collaborative scheduling model under carbon tax constraint, the upper model taking minimization of operating cost and carbon emission of a power system as a target; the constraint conditions of the upper-layer model comprise voltage safety, power balance and power flow constraint, and operation constraint of a lithium battery and an SOFC (Solid Oxide Fuel Cell); the lower layer model aims at maximizing hydrogen production income and minimizing total value chain carbon emission; constructing a collaborative optimization model based on an alternating direction multiplier method; the objective function is to minimize the ADN total operation cost and maximize the economic benefit of the hydrogen station; and constructing a multi-time-resolution three-layer robust scheduling model to solve the IPHS collaborative scheduling model to obtain a scheduling scheme. The method has the advantages that the IPHS carbon emission rebound risk under DRE uncertainty is revealed, and the potential of SOFC grid connection in the aspect of low-carbon economic collaborative development is verified.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power systems, in particular to a multi-time resolution three-layer robust power system scheduling method. BACKGROUND

[0002] In recent years, in order to reduce the dependence on fossil energy and reduce carbon emissions, more and more distributed clean energy (DRE), mainly wind and solar energy, is introduced into the power system. However, the randomness and volatility of DRE require energy storage devices with flexible adjustment capacity to support. Due to the difficulty in adapting to high penetration of clean energy, the hydrogen production and hydrogen power interaction system (IPHS) based on clean energy is considered as a feasible solution strategy, and the carbon emissions are close to zero. However, due to the relatively high energy conversion loss, if the electricity is converted into hydrogen in the IPHS, and then the hydrogen is converted back into electricity, it will be challenging to meet the economic standards of IPHS. In response to this problem, the National Renewable Energy Laboratory pointed out that green hydrogen has great terminal consumption potential, and direct sales of green hydrogen can increase revenue. This is now a successful method to make power-to-hydrogen (P2H) commercially viable. Therefore, the present application proposes an IPHS model containing hydrogen sales, which improves clean energy consumption while improving economic efficiency. However, when P2H is actually implemented, due to the insufficient amount of hydrogen produced by clean energy, coal power transmitted from the power grid is used, which increases the risk of carbon emissions. With the increase of the percentage of coal power generation in the power grid, this risk becomes more obvious.

[0003] The core of the aforementioned issues lies in the utilization and conversion efficiency of clean energy. Among these, the randomness of distributed renewable energy (DRE) presents a significant challenge. Due to the power supply-demand mismatch problem in power systems with a high proportion of DRE, hydrogen production stations (HGS) often tend to draw power from the main transmission grid (TG). Existing research shows that in IPHS, a 1% increase in the electricity-to-hydrogen conversion efficiency can improve DRE utilization by approximately 2.4% to 3.2% and reduce carbon emissions by more than 11 kg. Therefore, hydrogen-blended natural gas systems have attracted considerable attention due to their ability to reduce hydrogen-to-methane conversion losses. However, this technology has an inherent efficiency bottleneck limited by the hydrogen blending concentration. To overcome this limitation, research is shifting towards fuel cells (FCs) to replace traditional gas turbines. According to the U.S. Department of Energy's 2019 "Solid Oxide Fuel Cell Program Progress Report" and the 2024 "Hydrogen and Fuel Cell Technology Office Multi-Year Plan," the development goals for high-temperature fuel cells (such as SOFCs) include: 70% conversion efficiency, a system cost of $900–$1300 / kW, and a lifespan of 40,000–80,000 hours or more. Carbon tax, as a preferred policy tool for managing carbon emissions in energy systems, has proven economic viability. This policy can effectively guide the consumption transition of low-carbon energy sources such as natural gas. Based on the empirical results of carbon pricing mechanisms in a low-carbon economy, this invention incorporates carbon tax into IPHS.

[0004] In addition to the three major issues mentioned above, the time-series coordination problem of renewable energy (DRE), battery storage, hydrogen storage, and electrolyzers (ETs) also needs to be addressed. Because the response times of these four modules differ significantly—the first two (DRE and batteries) require minute-level responses, while the latter three (hydrogen storage and ETs) require hour-level scheduling—forcing a uniform minute-level resolution in IPHS optimization with full-cycle carbon tax constraints would significantly increase the difficulty of solving the problem. Conversely, while using an hour-level resolution could improve model solvability, it would sacrifice adaptability to renewable energy fluctuations and weaken the rapid response advantage of energy storage systems. Adaptive Time Scheduling (TAD) technology offers a feasible solution. Its core lies in dynamically aggregating scheduling periods, rather than the traditional fixed 15-minute or 1-hour intervals, thereby reducing the dimensionality of time-series optimization. Based on this, TAD supports multi-timescale coordinated scheduling: modules with fast responses (such as batteries and DRE) are independently subdivided into finer time granularities, while hydrogen-related models use a coarser resolution. However, given the uncertainties of DRE and cross-departmental data silos, how to effectively integrate multi-resolution modules remains an open question that has not yet been fully explored. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide a three-layer robust power system scheduling method based on multi-time resolution.

[0006] The objective of this invention is achieved through the following technical solution: a three-layer robust power system dispatching method based on multi-time resolution, the method comprising,

[0007] An IPHS coordinated dispatch model under carbon tax constraints is constructed. The upper-level model of the IPHS coordinated dispatch model aims to minimize the operating cost and carbon emissions of the power system, and its objective function is:

[0008] (1)

[0009] (2)

[0010] In the formula, The daily operating cost of an active distribution network; For the cost of natural gas use, The cost of using electricity from the power transmission network; , , , , These are the usage costs of hybrid energy storage systems, lithium batteries, solid oxide fuel cells, electrolyzers, and hydrogen storage tanks, respectively. Carbon penalty costs for active distribution networks; Cost of producing hydrogen by electricity; The price of natural gas; The amount of natural gas injected into the solid oxide fuel cell; The price of electricity in the power transmission network; Power derived from electrical energy transmitted from the power grid; The capacity cost of lithium batteries; The storage capacity of lithium batteries; The power of the lithium battery; This refers to the rated power of the lithium battery. / This refers to the discharge / charge power of the lithium battery. For the lifespan of lithium batteries; This refers to the rated capacity of the lithium battery. Depth of charge and discharge for lithium batteries; , , , These are the parameters of the lithium battery life function constant; Index for electrolytic cell equipment; The power cost of solid oxide fuel cells; The power converted from gas to electricity; For time intervals; This refers to the output power of a solid oxide fuel cell; For the lifespan of solid oxide fuel cells; The electricity price for an active distribution network; The power of the electro-hydrogen production; for; for; for; The carbon penalty cost of gas-to-electricity conversion; Carbon penalty costs for electricity generated by power transmission networks; Carbon penalty costs for distributed clean energy; For carbon tax; Carbon emission factors for gas-to-electricity conversion; This refers to the output power of the gas-to-electric conversion. Carbon emission factors are derived from electricity generated by the upstream power grid. Photovoltaic carbon emission factor; Photovoltaic power; Carbon emission factors for wind power; Wind power output; For the set of optimization variables of active distribution networks; This refers to the power output of the first-stage lithium battery. Energy stored in lithium batteries;

[0011] The upper-level model constraints of the IPHS collaborative scheduling model include voltage safety, power balance and power flow constraints, and operational constraints of lithium batteries and SOFCs.

[0012] The lower-level model of the IPHS collaborative scheduling model aims to maximize hydrogen production revenue and minimize carbon emissions across the entire value chain. Its objective function is:

[0013] (12)

[0014] (13)

[0015] In the formula, The daily operating cost of the hydrogen station; Cost of using the electrolytic cell; Cost of using hydrogen storage tanks; Carbon penalty costs from hydrogen stations; For the revenue generated from the sale of hydrogen; The power cost of the electrolytic cell; The power of the electro-hydrogen production; This refers to the service life of the electrolytic cell; Cost per unit capacity of hydrogen storage tanks; The price of hydrogen; The cost of carbon penalties from the electrolyzer; The amount of hydrogen injected into the hydrogen storage tank; The amount of hydrogen stored in the hydrogen storage tank; Cost of producing hydrogen by electricity; The rated power of the electrolytic cell, ; This refers to the capacity of the hydrogen storage tank. Use carbon emission factors for electrolyzers; For the lifespan of the hydrogen storage tank; The amount of hydrogen sold; For carbon tax; This is the set of optimization variables for HGS;

[0016] A collaborative optimization model based on the alternating direction multiplier method is constructed; its objective function is to minimize the total operating cost of the ADN and maximize the economic benefits of the hydrogen station.

[0017] (16)

[0018] (17)

[0019] (18)

[0020] In the formula, The daily operating cost of an active distribution network; and All are Lagrange multipliers; The daily operating cost of the hydrogen station; These represent the willingness of hydrogen stations and active distribution networks to output hydrogen production power, respectively. , The objective functions are defined as active distribution networks and hydrogen stations with penalty terms;

[0021] A robust IPHS collaborative scheduling model with multiple time resolutions is constructed and solved to obtain the scheduling scheme.

[0022] The constraints of the IPHS cooperative scheduling model are:

[0023] (3)

[0024] (4)

[0025] (5)

[0026] (6)

[0027] (7)

[0028] (8)

[0029] (9)

[0030] (10)

[0031] (11)

[0032] , , ;

[0033] In the formula, The square value of the branch current; Branch current; The square of the node voltage; Node voltage; For load power that does not consider uncertainties; Branch resistance; This is a reactive load; For branch circuit reactance; This refers to the power output of the first-stage lithium battery. This refers to the power adjustment amount for the second-stage lithium battery. This represents the discharge power of the lithium battery in the first stage. The charging power for the first stage of lithium batteries; The storage capacity of lithium batteries; It is at its minimum state of charge; It is at its maximum state of charge; This refers to the rated capacity of the lithium battery. The power converted from gas to electricity; This refers to the collection of nodes in an active distribution network. A collection of runtime segments; DRE output without considering uncertainties; , , These refer to the output power of fuel cells, lithium batteries, and electrolyzers, respectively. To clean up the reactive power output of the power supply; This represents the second-stage output power of the lithium battery. , This represents the maximum output power of lithium batteries and fuel cells. and For the indexes and collections of each device; , , These represent the output power of the J-node fuel cell, lithium battery, and electrolyzer, respectively. and For each device's index and collection; for Node fuel cell output power.

[0034] The constraints of the lower-level model of the IPHS cooperative scheduling model are as follows:

[0035] Process constraints in hydrogen production and storage:

[0036] (14)

[0037] (15)

[0038] In the formula, The power of the electro-hydrogen production; The conversion efficiency for electro-hydrogen production; The power of the electro-hydrogen production; It has the high calorific value of H2; The amount of hydrogen injected into the hydrogen storage tank; The amount of hydrogen stored in the hydrogen storage tank; This refers to the rated power of the electrolytic cell; The amount of hydrogen produced by electricity; The loss rate of hydrogen produced by electricity; This refers to the amount of hydrogen sold.

[0039] The multi-time-resolution three-layer robust scheduling model introduces auxiliary variables. and Couple variables at different time granularities , and :

[0040] (19)

[0041] (20)

[0042] (twenty one)

[0043] in, This is the set of optimization variables with a time resolution of 1 hour, including the first / second / third layer robust optimization variable sets. , , ; This is a set of optimization variables with a time resolution of 5 minutes, including the first / second / third layer robust optimization variable sets. , , ; A collection of runtime segments; Total operating cost; Operating cost of the active distribution network; The operating cost of the hydrogen station.

[0044] By algebraically transforming the nonlinear relations of equations (20)-(21) into the MP problem shown in the following equation:

[0045] (twenty two)

[0046] (twenty three)

[0047] Among them, the (r-1)th coarse time resolution variable solution and Through second / third level fine-grained solution , The frequency domain interpolation is generated, and constraints are added to the MP problem as shown in Equation (22). Then, the 1-hour solution of the MP output in the r-th iteration is obtained. A 5-minute solution needs to be obtained through frequency conversion. ; / / For indexing nodes in an active distribution network; This refers to the collection of nodes in an active distribution network. and For the runtime segment index; Optimize the objective function for the MP problem during the r-th iteration; and For each device's index and collection;

[0048] The SP1 layer, operating at a 5-minute time granularity, optimizes lithium battery regulation strategies. In response to DRE, simultaneously, 5-minute level solutions from the first level. and the solution at the third level This will be incorporated as an additional constraint using the following formula:

[0049] (twenty four)

[0050] Among them, the solution of SP1 layer Solution with the first level This will be jointly transmitted to the third level; and An index and collection for lithium batteries; , These are the power discharge and charging adjustment amounts for the second-stage lithium battery, respectively. Optimize the objective function for the SP1 problem during the r-th iteration;

[0051] SP2 runs at 5-minute intervals to assess the worst-case uncertainty scenarios:

[0052] (25)

[0053] Among them, the solution of SP2 layer The feedback will be sent to the second level. / These represent the load power without considering uncertainties and with consideration of uncertainties, respectively. / These are the DRE outputs with and without considering uncertainty, respectively. and An index and collection for lithium batteries; and Uncertainty fluctuation ranges for load and DRE respectively; The objective function for optimizing the SP2 problem during the r-th iteration is given.

[0054] The present invention has the following advantages:

[0055] This invention, based on real-world datasets and institutional forecasts, reveals the risk of IPHS carbon emission rebound under DRE uncertainty and verifies the potential of SOFC grid connection in achieving coordinated low-carbon economic development. It proposes a carbon tax-aware three-layer IPHS scheduling model and embeds a privacy-preserving ADMM algorithm to support cross-sectoral collaborative optimization, better aligning with real-world scenarios involving multiple stakeholders. For the first time, a multi-time-resolution robust optimizer is proposed to reduce the solution difficulty of the three-layer IPHS scheduling model, improving DRE absorption capacity through fast-slow dynamic coupling while ensuring the effectiveness of carbon constraints throughout the hydrogen energy "production-sales" lifecycle. Attached Figure Description

[0056] Figure 1 This is a schematic diagram of the simulation system structure. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only for explaining the invention and are not intended to limit the invention; that is, the described embodiments are merely some embodiments of the invention, and not all embodiments. The components of the embodiments of the invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0058] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0059] It should be noted that relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0060] The present invention will be further described below with reference to the accompanying drawings, but the scope of protection of the present invention is not limited to the following description.

[0061] like Figure 1 As shown, a robust three-layer power system dispatching method based on multi-time resolution is proposed, which includes:

[0062] An IPHS coordinated dispatch model under carbon tax constraints is constructed. The upper-level model of the IPHS coordinated dispatch model aims to minimize the operating cost and carbon emissions of the power system. The objective function comprehensively considers energy costs, the usage cost of hybrid energy storage systems (HESS), and carbon emission costs based on carbon emission statistics throughout the dispatch cycle. Its objective function is as follows:

[0063] (1)

[0064] (2)

[0065] In the formula, The daily operating cost of an active distribution network; For the cost of natural gas use, The cost of using electricity from the power transmission network; , , , , These are the usage costs of hybrid energy storage systems, lithium batteries, solid oxide fuel cells, electrolyzers, and hydrogen storage tanks, respectively. Carbon penalty costs for active distribution networks; Cost of producing hydrogen by electricity; The price of natural gas; The amount of natural gas injected into the solid oxide fuel cell; The price of electricity in the power transmission network; Power derived from electrical energy transmitted from the power grid; The capacity cost of lithium batteries; The storage capacity of lithium batteries; The power of the lithium battery; This refers to the rated power of the lithium battery. / This refers to the discharge / charge power of the lithium battery. For the lifespan of lithium batteries; This refers to the rated capacity of the lithium battery. Depth of charge and discharge for lithium batteries; , , , These are the parameters of the lithium battery life function constant; Index for electrolytic cell equipment; The power cost of solid oxide fuel cells; The power converted from gas to electricity; For time intervals; This refers to the output power of a solid oxide fuel cell; For the lifespan of solid oxide fuel cells; The electricity price for an active distribution network; The power of the electro-hydrogen production; for; for; for; The carbon penalty cost of gas-to-electricity conversion; Carbon penalty costs for electricity generated by power transmission networks; Carbon penalty costs for distributed clean energy; For carbon tax; Carbon emission factors for gas-to-electricity conversion; This refers to the output power of the gas-to-electric conversion. Carbon emission factors are derived from electricity generated by the upstream power grid. Photovoltaic carbon emission factor; Photovoltaic power; Carbon emission factors for wind power; Wind power output; For the set of optimization variables of active distribution networks; This refers to the power output of the first-stage lithium battery. Energy stored in lithium batteries;

[0066] Energy costs mainly come from natural gas procurement costs. and electricity procurement costs from the upstream power grid The cost of a hybrid energy storage system (HESS) involves the cost of using lithium batteries. Cost of using SOFC It is worth noting that the carbon tax mechanism for the power system covers: the carbon emission costs of the gas-to-electricity conversion process. Carbon emission costs of electricity / distributed renewable energy generation from the upstream grid ( , At the same time, HGS needs to pay the electricity costs for hydrogen production to the power system. .

[0067] The upper-level model constraints of the IPHS collaborative scheduling model include voltage safety, power balance and power flow constraints, and operational constraints of lithium batteries and SOFCs.

[0068] The lower-level model of the IPHS collaborative scheduling model aims to maximize hydrogen production revenue and minimize carbon emissions across the entire value chain. Its objective function is:

[0069] (12)

[0070] (13)

[0071] In the formula, The daily operating cost of the hydrogen station; Cost of using the electrolytic cell; Cost of using hydrogen storage tanks; Carbon penalty costs from hydrogen stations; For the revenue generated from the sale of hydrogen; The power cost of the electrolytic cell; The power of the electro-hydrogen production; This refers to the service life of the electrolytic cell; Cost per unit capacity of hydrogen storage tanks; The price of hydrogen; The cost of carbon penalties from the electrolyzer; The amount of hydrogen injected into the hydrogen storage tank; The amount of hydrogen stored in the hydrogen storage tank; Cost of producing hydrogen by electricity; This refers to the capacity of the hydrogen storage tank. Use carbon emission factors for electrolyzers; For the lifespan of the hydrogen storage tank; The amount of hydrogen sold; For carbon tax; This is the set of optimization variables for HGS;

[0072] Electrolytic cell usage cost Cost of using hydrogen storage tanks This is also taken into consideration. The HGS carbon tax mainly addresses carbon emissions from the use of electrolyzers. ,and Characterizes revenue from hydrogen sales.

[0073] Given the privacy protection requirements of the active distribution network and hydrogen stations in the electric-hydrogen interconnection system, a collaborative optimization model based on the alternating direction multiplier method is constructed; decomposed into: Problem 1: Minimize the total operating cost of the active distribution network under constraints (3)-(11). Problem 2: Maximize the economic benefits of the hydrogen station under constraints (14)-(15), by iteratively exchanging information until the convergence criterion is met. To decouple the exchange power variables between the active distribution network and the hydrogen station. Establish auxiliary variables And impose constraints (18), the objective function of which is to minimize the total operating cost of the active distribution network and maximize the economic benefits of the hydrogen station:

[0074] (16)

[0075] (17)

[0076] (18)

[0077] In the formula, The daily operating cost of an active distribution network; and All are Lagrange multipliers; The daily operating cost of the hydrogen station; These represent the willingness of hydrogen stations and active distribution networks to output hydrogen production power, respectively. , The objective functions are defined as active distribution networks and hydrogen stations with penalty terms;

[0078] A robust three-layer IPHS cooperative scheduling model with multiple time resolutions was constructed and solved to obtain the scheduling scheme. The first layer of the model is the IPHS cooperative scheduling model itself; the second layer solves the worst-case scenario; and the third layer solves the power adjustment problem for the lithium battery with a 5-minute time resolution. This robust three-layer model with multiple time resolutions fully utilizes the responsiveness of the lithium battery while avoiding excessive computational complexity in solving the IPHS scheduling model.

[0079] The constraints of the IPHS cooperative scheduling model are:

[0080] (3)

[0081] (4)

[0082] (5)

[0083] (6)

[0084] (7)

[0085] (8)

[0086] (9)

[0087] (10)

[0088] (11)

[0089] , , ;

[0090] In the formula, The square value of the branch current; Branch current; The square of the node voltage; Node voltage; For load power that does not consider uncertainties; Branch resistance; This is a reactive load; For branch circuit reactance; This refers to the power output of the first-stage lithium battery. This refers to the power adjustment amount for the second-stage lithium battery. This represents the discharge power of the lithium battery in the first stage. The charging power for the first stage of lithium batteries; The storage capacity of lithium batteries; It is at its minimum state of charge; It is at its maximum state of charge; This refers to the rated capacity of the lithium battery. The power converted from gas to electricity; This refers to the collection of nodes in an active distribution network. A collection of runtime segments; DRE output without considering uncertainties; , , These refer to the output power of fuel cells, lithium batteries, and electrolyzers, respectively. To clean up the reactive power output of the power supply; This represents the second-stage output power of the lithium battery. , This represents the maximum output power of lithium batteries and fuel cells. and For each device's index and collection; The active power of the branch between nodes i and j; The active power of the branch between nodes j and h; The reactive power of the branch between nodes i and j; The reactive power of the branch between nodes j and h; for Node fuel cell output power.

[0091] The constraints of the lower-level model of the IPHS cooperative scheduling model are as follows:

[0092] Process constraints in hydrogen production and storage:

[0093] (14)

[0094] (15)

[0095] In the formula, The power of the electro-hydrogen production; The conversion efficiency for electro-hydrogen production; The power of the electro-hydrogen production; It has the high calorific value of H2; The amount of hydrogen injected into the hydrogen storage tank; The amount of hydrogen stored in the hydrogen storage tank; This refers to the rated power of the electrolytic cell; The amount of hydrogen produced by electricity; The loss rate of hydrogen produced by electricity; This refers to the amount of hydrogen sold.

[0096] The three-layer robust scheduling model with multiple time resolutions introduces auxiliary variables to achieve multi-timescale optimization. and Couple variables at different time granularities , and :

[0097] (19)

[0098] (20)

[0099] (twenty one)

[0100] in, This is the set of optimization variables with a time resolution of 1 hour, including the first / second / third layer robust optimization variable sets. , , ; This is a set of optimization variables with a time resolution of 5 minutes, including the first / second / third layer robust optimization variable sets. , , ; A collection of runtime segments; Total operating cost; Operating cost of the active distribution network; The operating cost of the hydrogen station; , .

[0101] First-level optimization problem (MP):

[0102] By algebraically transforming the nonlinear relationships of equations (20)-(21) into the MP problem shown in the following equation, this layer operates with a time resolution of 1 hour, realizing the IPHS collaborative optimization framework described in (1)-(18):

[0103] (twenty two)

[0104] (twenty three)

[0105] Among them, the (r-1)th coarse time resolution variable solution and Through second / third level fine-grained solution , The frequency domain interpolation is generated, and constraints are added to the MP problem as shown in Equation (22). Then, the 1-hour solution of the MP output in the r-th iteration is obtained. A 5-minute solution needs to be obtained through frequency conversion. ; / / For indexing nodes in an active distribution network; This refers to the collection of nodes in an active distribution network. and For the runtime segment index; Optimize the objective function for the MP problem during the r-th iteration; and For each device's index and collection;

[0106] Second-level optimization problem (SP1):

[0107] The SP1 layer, operating at a 5-minute time granularity, optimizes lithium battery regulation strategies. In response to DRE, simultaneously, 5-minute level solutions from the first level. and the solution at the third level This will be incorporated as an additional constraint through the following formula to ensure consistency and coordination between different levels:

[0108] (twenty four)

[0109] Among them, the solution of SP1 layer Solution with the first level This will be jointly transmitted to the third level; and An index and collection for lithium batteries; , These are the power discharge and charging adjustment amounts for the second-stage lithium battery, respectively. Optimize the objective function for the SP1 problem during the r-th iteration;

[0110] Third-level optimization problem (SP2):

[0111] SP2 runs at 5-minute intervals to assess the worst-case uncertainty scenarios:

[0112] (25)

[0113] Among them, the solution of SP2 layer The feedback will be sent to the second level. / These represent the load power without considering uncertainties and with consideration of uncertainties, respectively. / These are the DRE outputs with and without considering uncertainty, respectively. and An index and collection for lithium batteries; and Uncertainty fluctuation ranges for load and DRE respectively; The objective function for optimizing the SP2 problem during the r-th iteration is given.

[0114] As shown in equations (22)-(25), the proposed model is specifically divided into three levels: the first level (optimization problem MP), the second level (optimization problem SP1), and the third level (optimization problem SP2). MP needs to be solved iteratively with SP1 / SP2. The r-th iteration process is shown in equations (22)-(43). There is also an iterative process between SP1 and SP2. As shown in equations (24)-(25), SP1 receives the result of the (s-1)-th iteration of SP2 and performs the s-th iteration, and then passes the result to SP2 until SP1 and SP2 converge. This converged solution will be used for the (r+1)-th iteration between SP1 / SP2 and MP.

[0115] Example system:

[0116] like Figure 1As shown, the simulation system consists of a 110 kV distribution network (24 nodes) and a hydrogen station. In the active distribution network: Distributed renewable energy (DRE) and lithium battery clusters are deployed at nodes E15 / E21 / E23; node E7 is configured with solid oxide fuel cells to realize natural gas-electricity conversion; node E15 is also connected to ET to supply hydrogen to the hydrogen station. All bus voltage limits are set to per unit values ​​of [0.95, 1.05]. 48 typical days (12 representative days per quarter, 5-minute granularity) extracted from the Elia dataset

[35] are used to cover various wind-solar-load scenarios. The output and load uncertainties of distributed renewable energy are simulated in the ranges of ±15% and ±5% of the predicted values, respectively.

[0117] To verify the feasibility of the proposed multi-time-resolution hydrogen energy trading model, four comparison schemes were set up:

[0118] S1: The scheduling model proposed in this paper;

[0119] S2: Based on S1, the objective function removes the full-cycle carbon emission cost. After optimization, the carbon emission and carbon tax are calculated based on the scheduling results and added to the total operating cost.

[0120] S3: Based on S1, remove the natural gas supply for solid oxide fuel cells and active distribution networks;

[0121] S4: Degenerate the multi-time resolution model of S1 into a traditional two-stage robust model, with the scheduling time window for each stage set to 1 hour.

[0122] Simulation results:

[0123] Table 1. Statistical analysis of different control groups

[0124]

[0125] Whether carbon tax is incorporated into the objective function or solid oxide fuel cells are used to support IPHS, the economics of IPHS can be improved and carbon emissions can be reduced, and this improvement effect remains stable in most typical operating scenarios. Quantitative analysis (Table 1) shows that: (1) the carbon tax mechanism brings more significant emission reduction effect (S1 average emission reduction of 20.5% compared to S2); (2) the deployment of solid oxide fuel cells generates better economic benefits (S1 average cost reduction of 7.6% compared to S3).

[0126] Table 2. Statistical results of the effectiveness of the multi-time resolution model

[0127]

[0128] *S1 vs. S4

[0129] This method can simultaneously reduce operating costs and carbon emissions. This is mainly due to the fact that the lithium battery's response to the high-frequency random fluctuations of DRE in the second level effectively compensates for the deviation between actual demand and predicted values ​​caused by the coarse scheduling time resolution in the first level. As a result, the S1 scheme uses less high-cost, high-emission upstream grid power than the S4 scheme.

[0130] The above description is merely a preferred embodiment of the present invention and does not constitute any limitation on the present invention. Any person skilled in the art can make many possible variations and modifications to the technical solution of the present invention, or modify it into equivalent embodiments, without departing from the scope of the present invention. Therefore, any modifications, equivalent changes, and alterations made to the above embodiments based on the technology of the present invention without departing from the scope of the present invention are within the protection scope of the present invention.

Claims

1. A robust three-layer power system dispatching method based on multi-time resolution, characterized in that: The method includes, An IPHS coordinated dispatch model under carbon tax constraints is constructed. The upper-level model of the IPHS coordinated dispatch model aims to minimize the operating cost and carbon emissions of the power system, and its objective function is: （1） （2） In the formula, The daily operating cost of an active distribution network; For the cost of natural gas use, The cost of using electricity from the power transmission network; , , , , These are the usage costs of hybrid energy storage systems, lithium batteries, solid oxide fuel cells, electrolyzers, and hydrogen storage tanks, respectively. Carbon penalty costs for active distribution networks; Cost of producing hydrogen by electricity; The price of natural gas; The amount of natural gas injected into the solid oxide fuel cell; The price of electricity in the power transmission network; Power derived from electrical energy transmitted from the power grid; The capacity cost of lithium batteries; The storage capacity of lithium batteries; The power of the lithium battery; This refers to the rated power of the lithium battery. / This refers to the discharge / charge power of the lithium battery. For lithium battery life; This refers to the rated capacity of the lithium battery. Depth of charge and discharge for lithium batteries; , , , These are the parameters of the lithium battery life function constant; Index for electrolytic cell equipment; The power cost of solid oxide fuel cells; The power converted from gas to electricity; For time intervals; This refers to the output power of a solid oxide fuel cell; For the lifespan of solid oxide fuel cells; The electricity price for an active distribution network; The power of the electro-hydrogen production; The carbon penalty cost of gas-to-electricity conversion; Carbon penalty costs for electricity generated by power transmission networks; Carbon penalty costs for distributed clean energy; For carbon tax; Carbon emission factors for gas-to-electricity conversion; This refers to the output power of the gas-to-electric conversion. Carbon emission factors are derived from electricity generated by the upstream power grid. Photovoltaic carbon emission factor; Photovoltaic power; Carbon emission factors for wind power; Wind power output; For the set of optimization variables of active distribution networks; This refers to the power output of the first-stage lithium battery. Energy stored in lithium batteries; The upper-level model constraints of the IPHS collaborative scheduling model include voltage safety, power balance and power flow constraints, and operational constraints of lithium batteries and solid oxide fuel cells. The lower-level model of the IPHS collaborative scheduling model aims to maximize hydrogen production revenue and minimize carbon emissions across the entire value chain. Its objective function is: （12） （13） In the formula, The daily operating cost of the hydrogen station; Cost of using the electrolytic cell; Cost of using hydrogen storage tanks; Carbon penalty costs from hydrogen stations; For the revenue generated from the sale of hydrogen; The power cost of the electrolytic cell; The power of the electro-hydrogen production; This refers to the service life of the electrolytic cell; Cost per unit capacity of hydrogen storage tanks; The price of hydrogen; The cost of carbon penalties from the electrolyzer; The amount of hydrogen injected into the hydrogen storage tank; The amount of hydrogen stored in the hydrogen storage tank; Cost of producing hydrogen by electricity; The rated power of the electrolytic cell, ; This refers to the capacity of the hydrogen storage tank. Use carbon emission factors for electrolyzers; For the lifespan of the hydrogen storage tank; The amount of hydrogen sold; For carbon tax; This is a set of optimization variables for hydrogen stations. Construct a collaborative optimization model based on the alternating direction multiplier method; Its objective function is to minimize the total operating cost of the active distribution network and maximize the economic benefits of the hydrogen station: （16） （17） （18） In the formula, The daily operating cost of an active distribution network; and All are Lagrange multipliers; The daily operating cost of the hydrogen station; These represent the willingness of hydrogen stations and active distribution networks to output hydrogen production power, respectively. , These are the objective functions for an active distribution network with penalty terms and a hydrogen station, respectively. A robust IPHS collaborative scheduling model with multiple time resolutions is constructed and solved to obtain the scheduling scheme.

2. The power system dispatching method based on a three-layer robust multi-time resolution according to claim 1, characterized in that: The constraints of the IPHS cooperative scheduling model are: （3） （4） （5） （6） （7） （8） （9） （10） （11） , , ； In the formula, The square value of the branch current; Branch current; The square of the node voltage; Node voltage; , These are the minimum and maximum node voltages; For load power that does not consider uncertainties; Branch resistance; This is a reactive load; For branch circuit reactance; This represents the second-stage output power of the lithium battery. This refers to the power output of the first-stage lithium battery. This refers to the power adjustment amount for the second-stage lithium battery. This represents the discharge power of the lithium battery in the first stage. The charging power for the first stage of lithium batteries; The storage capacity of lithium batteries; It is at its minimum state of charge; It is at its maximum state of charge; This refers to the rated capacity of the lithium battery. The power converted from gas to electricity; This refers to the collection of nodes in an active distribution network. A collection of runtime segments; The active power of the branch between nodes i and j; The active power of the branch between nodes j and h; The reactive power of the branch between nodes i and j; The reactive power of the branch between nodes j and h; To clean up the reactive power output of the power supply; , This represents the maximum output power of lithium batteries and fuel cells. DRE output without considering uncertainties; , , These represent the output power of the J-node fuel cell, lithium battery, and electrolyzer, respectively. and For each device's index and collection; for Node fuel cell output power.

3. The power system dispatching method based on a three-layer robust multi-time resolution according to claim 1, characterized in that: The constraints of the lower-level model of the IPHS cooperative scheduling model are as follows: Process constraints in hydrogen production and storage: （14） （15） In the formula, The power of the electro-hydrogen production; The conversion efficiency for electro-hydrogen production; The power of the electro-hydrogen production; It has the high calorific value of H2; The amount of hydrogen injected into the hydrogen storage tank; The amount of hydrogen stored in the hydrogen storage tank; This refers to the rated power of the electrolytic cell; The amount of hydrogen produced by electricity; The loss rate of hydrogen produced by electricity; This refers to the amount of hydrogen sold.

4. The power system dispatching method based on a three-layer robust multi-time resolution according to claim 1, characterized in that: The multi-time-resolution three-layer robust scheduling model introduces auxiliary variables. and Couple variables at different time granularities , and : （19） （20） （21） in, This is a set of optimization variables with a time resolution of 1 hour, including the first / second / third layer robust optimization variable sets. , , ; This is a set of optimization variables with a time resolution of 5 minutes, including the first / second / third layer robust optimization variable sets. , , ; A collection of runtime segments; Total operating cost; For the operating costs of active distribution networks; The operating cost of the hydrogen station; By algebraically transforming the nonlinear relations of equations (20)-(21) into the MP problem shown in the following equation: （22） （23） Among them, the (r-1)th coarse time resolution variable solution and Solving through second / third level fine-grained variables , The frequency domain interpolation is generated, and constraints are added to the MP problem as shown in Equation (22). Then, the 1-hour solution of the MP output in the r-th iteration is obtained. A 5-minute solution needs to be obtained through frequency conversion. ; / / For indexing nodes in an active distribution network; This refers to the collection of nodes in an active distribution network. and For the runtime segment index; Optimize the objective function for the MP problem during the r-th iteration; and For each device's index and collection; The SP1 layer, operating at a 5-minute time granularity, optimizes lithium battery regulation strategies. In response to DRE, simultaneously, 5-minute resolution solution from the first level. and the solution at the third level This will be incorporated as an additional constraint using the following formula: （24） Among them, the solution of SP1 layer Solution with the first level This will be jointly transmitted to the third level; and An index and collection for lithium batteries; , These are the power discharge and charging adjustment amounts for the second-stage lithium battery, respectively. Optimize the objective function for the SP1 problem during the r-th iteration; SP2 runs at 5-minute intervals to assess the worst-case uncertainty scenarios: （25） Among them, the solution of SP2 layer This will be fed back to the second level; / These represent the load power without considering uncertainties and with consideration of uncertainties, respectively. / These are the DRE outputs with and without considering uncertainty, respectively. and An index and collection for lithium batteries; and Uncertainty fluctuation ranges for load and DRE respectively; The objective function for optimizing the SP2 problem during the r-th iteration is given.

Images