Electricity-hydrogen comprehensive energy system planning method considering multi-time scale uncertainty

Abstract

The invention relates to an electricity-hydrogen comprehensive energy system planning method considering multi-time scale uncertainty. The method comprises the following steps: establishing an electricity-hydrogen comprehensive energy system model; based on the electricity-hydrogen comprehensive energy system model, considering the uncertainty of wind and light output and load requirements, and constructing a planning mathematical model of the electricity-hydrogen comprehensive energy system by using a multi-time scale uncertainty modeling method fusing opportunity constraint and stochastic optimization; based on the electricity-hydrogen comprehensive energy system model, establishing an electricity market model as a lower-layer optimization model of a planning mathematical model; and solving the planning mathematical model by adopting a two-stage optimization method to obtain a planning scheme. Compared with the prior art, a multi-time-scale planning mathematical model of the electricity-hydrogen comprehensive energy system is established, and planning of the electricity-hydrogen comprehensive energy system is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of integrated energy system planning, and in particular to a method for planning an electric-hydrogen integrated energy system taking into account multi-time scale uncertainties. Background Art

[0002] The global energy crisis and environmental pollution are becoming increasingly severe. Against this backdrop, promoting a low-carbon transition in the energy system has become a core strategy for achieving sustainable development. Integrated energy systems couple multiple energy subsystems and comprehensively consider the interactions between these energy sources. The uncertainty of wind and solar power output and load demand in an electric-hydrogen integrated energy system creates complex information exchange and coordinated operations among various entities, making planning for this system challenging.

[0003] Chinese patent application CN119004752A proposes a comprehensive energy system planning method that considers multi-timescale uncertainty. This method employs stochastic optimization and fuzzy optimization methods for models that consider short-timescale uncertainty and long-timescale uncertainty, respectively, to construct a low-carbon planning model for a hydrogen-containing comprehensive energy system that considers multi-timescale uncertainty. The method transforms the multi-objective optimization problem into a single-objective optimization problem using a weighted satisfaction index method and solves the planning results. The configuration of the hydrogen-containing comprehensive energy system is adjusted based on the planning results, achieving efficient thermal energy utilization and economically optimal operation of the hydrogen-containing comprehensive energy system. However, this method fails to consider the dynamic characteristics of energy transmission and storage in detail, which may result in inaccurate simulations of energy flows in practical applications, affecting the scientific nature and accuracy of system planning. Furthermore, the method employs stochastic optimization and fuzzy optimization methods to handle uncertainty, but fails to fully consider the synergistic effect between the two methods and their combined impact with other uncertainties. This may fail to fully capture the uncertainty in the system, affecting the reliability of the planning results. Summary of the Invention

[0004] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and provide a method for planning an electric-hydrogen integrated energy system taking into account multi-time scale uncertainties.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] A method for planning an electric-hydrogen integrated energy system considering multi-timescale uncertainties, the method comprising the following steps:

[0007] Establish an electricity-hydrogen integrated energy system model;

[0008] Based on the above-mentioned electric-hydrogen integrated energy system model, considering the uncertainty of wind and solar power output and load demand, a multi-time-scale uncertainty modeling method integrating chance constraints and stochastic optimization is used to construct a planning mathematical model for the electric-hydrogen integrated energy system;

[0009] Based on the electricity-hydrogen integrated energy system model, an electricity market model is established as a lower-level optimization model of the planning mathematical model;

[0010] A two-stage optimization method is used to solve the planning mathematical model to obtain a planning solution.

[0011] Furthermore, the electric-hydrogen integrated energy system model includes a photovoltaic unit, a wind turbine unit, a traditional generator unit, a battery, a hydrogen storage system, an electrolyzer, a hydrogen storage tank and a fuel cell, wherein the battery and the hydrogen storage system constitute a hybrid energy storage device, and the hydrogen storage system includes an electrolyzer, a fuel cell and a hydrogen storage tank.

[0012] Furthermore, the planning mathematical model includes an objective function and corresponding constraints. The objective function aims to minimize the annual investment cost of the electricity-hydrogen integrated energy system. The expression of the objective function is:

[0013] minC=C inv +C m +C fuel +C Q +C E

[0014] Among them, C is the objective function, C inv is the investment cost, C fuel is the fuel cost, C m is the maintenance cost, C Q is the penalty cost for curtailment of electricity, C E The cost of purchasing electricity for the system.

[0015] Furthermore, the expression of the investment cost is:

[0016]

[0017] Among them, r represents the discount rate, β represents the life of the equipment, α represents the maintenance cost coefficient, and c SHS,e 、c SHS,p They represent the SHS unit capacity investment cost and unit power investment cost respectively, c Bat,e 、c Bat,p They represent the battery unit capacity investment cost and unit power investment cost respectively, E SHS,max Indicates the upper limit of hydrogen storage system capacity, P el,max Indicates the upper limit of electrolytic cell power, P fc,max Indicates the upper limit of electrolytic cell power, EBat,max Indicates the upper limit of battery capacity, P Bat,max Indicates the upper limit of battery power, P CU,max Indicates the upper limit of the output of traditional units, c CU Represents the unit power generation cost of traditional generator sets;

[0018] The expression of the maintenance cost is:

[0019] C m =αC inv

[0020] Among them, c inv represents the investment cost;

[0021] The fuel cost is expressed as:

[0022]

[0023] Where D represents the total number of typical days, D p is the total number of days in a year, W d Indicates the weights of different typical days, W s Indicates the probability of different scenarios, c fuel Indicates the fuel cost per unit output of traditional units; represents the total power generation of traditional units at time t on day d under scenario s;

[0024] The expression of the curtailment penalty cost is:

[0025]

[0026] Among them, c Q represents the penalty cost corresponding to unit curtailed power, represents the total amount of power abandoned by the system at time t on day d under scenario s;

[0027] The expressions of the electricity purchase cost are:

[0028]

[0029] Among them, c LS represents the marginal price of electricity in the electricity market, It represents the total power purchased by the system at time t on day d under scenario s.

[0030] Furthermore, the constraints include short-term power balance constraints, equipment constraints, long-term power balance constraints, conversion determinism constraints, investment constraints and two-stage coordinated output constraints.

[0031] Furthermore, the expression of the short-term power balance constraint is:

[0032]

[0033] in, represents the short-term photovoltaic output fluctuation at time t on day d under scenario s, represents the short-term wind power output fluctuation at time t on day d under scenario s, represents the short-term electricity demand fluctuation at time t on day d under scenario s, represents the short-term traditional unit output at time t on day d under scenario s, represents the short-term electricity purchase amount at time t on day d under scenario s, represents the discharge power of the short-term battery at time t on day d under scenario s, represents the charging power of the short-term battery at time t on day d under scenario s, represents the power curtailment during the short-term power fluctuation phase at time t on day d under scenario s;

[0034] The expression of the device constraint is:

[0035]

[0036]

[0037] in, represents the capacity of the battery at time t on day d under scenario s, Represents the battery's self-consumption rate, η bat,char Indicates the battery charging efficiency, η bat,dis Indicates the discharge efficiency of the battery, Indicates the charging mark of the battery at time t on day d in scenario s. Indicates the discharge mark of the battery at time t on day d under scenario s, E bat,min Indicates the lower limit of battery capacity, E bat,max Indicates the upper limit of battery capacity;

[0038] The expression of long-term power balance constraint is:

[0039]

[0040] in, Indicates the long-term photovoltaic unit output, Indicates the long-term wind turbine output, Indicates the output of traditional generator sets when meeting long-term power demand. It represents the amount of electricity purchased by the system to meet long-term electricity demand. Indicates the amount of power wasted during long-term fluctuations. represents the output of the electrolytic cell at time t on day d during the long-term energy fluctuation stage, represents the output of the fuel cell at time t on day d during the long-term energy fluctuation phase, represents the long-term load demand, and ε represents the risk factor of the system;

[0041] The expression of the transformed deterministic constraint is:

[0042]

[0043] Among them, Z ε represents the quantile with a confidence level of 1-ε under the normal distribution, represents the average value of long-term photovoltaic output, represents the average value of long-term wind power output, represents the average value of long-term load demand, σ Δp represents the standard deviation, represents the variance of long-term photovoltaic output, represents the variance of long-term wind power output, represents the variance of long-term load demand;

[0044] The investment constraint is expressed as:

[0045]

[0046] in, is the total power generation of traditional units at time t on day d under scenario s, represents the total amount of power abandoned by the system at time t on day d under scenario s;

[0047] The expression of the two-stage coordinated output constraint is:

[0048]

[0049] Furthermore, the electricity market model is led by an independent power system operator, which optimizes the electricity purchase amount of the electricity-hydrogen integrated energy system through market clearing, and feeds the market clearing results into the planning mathematical model.

[0050] Furthermore, the expression of the electricity market model is:

[0051]

[0052]

[0053] Among them, c cg represents the bid price of CG, represents the output power of CG at time t on day d under scenario s, represents the total system power purchase at time t on day d under scenario s, C e represents the bidding price in the electricity market, Ceu represents the bid price of EU, represents the electricity consumption of EU at time t on day d under scenario s, D p is the total number of days in a year, represents the climbing limit of CG, λ it Lagrange multipliers representing equality constraints, and Lagrange multipliers representing inequality constraints.

[0054] Furthermore, the process of solving the planning mathematical model includes:

[0055] Constructing a Lagrangian function of the electricity market model, and based on the KKT constraints, converting the electricity market model into constraints of a planning mathematical model to form a single-layer nonlinear model with equilibrium constraints;

[0056] The nonlinear terms in the single-layer nonlinear model are linearized using the Big-M method, and the single-layer nonlinear model is converted into a single-layer mixed integer linear programming problem and solved to obtain a planning solution.

[0057] Furthermore, the KKT constraints include:

[0058]

[0059]

[0060]

[0061]

[0062] Among them, c cg represents the bidding price of CG, C e represents the bidding price in the electricity market, C eu represents the bid price of EU, λ it Lagrange multipliers representing equality constraints, and Lagrange multipliers representing inequality constraints, represents the output power of CG at time t on day d under scenario s, represents the total system power purchase at time t on day d under scenario s, represents the electricity consumption of EU at time t on day d under scenario s, D p is the total number of days in a year, Indicates the climbing limit of CG, P cg,max Indicates the maximum value of CG output power, P cg,min Indicates the minimum value of CG output power, P buy,maxIndicates the maximum amount of electricity purchased, P buy ,min Indicates the minimum value of electricity purchase, P eu,min Indicates the minimum value of EU power consumption, P eu,max Indicates the maximum value of EU electric power consumption.

[0063] Compared with the prior art, the present invention has the following beneficial effects:

[0064] 1. In view of the uncertainty of renewable energy output and load demand, the present invention integrates chance constraints and random optimization methods to establish a multi-time-scale planning mathematical model for the electric-hydrogen integrated energy system, thereby realizing the planning of the electric-hydrogen integrated energy system. In the first stage of the planning mathematical model, the present invention optimizes the long-term trend with chance constraints, ensuring the reliability and economy of the long-term operation of the system. In the second stage, random optimization processes short-term fluctuations, enhancing the short-time-scale regulation capability of the system. As a whole, the system optimizes planning at the long-term scheduling level, while enhancing the flexibility of short-term operation, and can more effectively coordinate the matching relationship between renewable energy output and load demand, thereby improving the economy and operational stability of the system. The present invention considers the power market model as the lower-level optimization model, effectively allocates resources through the market mechanism, and improves the economy, flexibility and operational efficiency of the overall system.

[0065] 2. The present invention utilizes a hybrid energy storage device consisting of a battery and a hydrogen storage system to address both long-term and short-term power imbalances, effectively coordinating the matching relationship between renewable energy output and load demand.

[0066] 3. This invention constructs a planning mathematical model with the goal of minimizing annual investment costs. By optimizing energy storage device configuration and scheduling strategies, it reduces unnecessary redundant energy storage configurations. For example, chance constraints allow for defaults within a certain probability range, reducing the need for capacity over-allocation in extreme situations and lowering system investment costs. Stochastic optimization dynamically coordinates system timing operations, reducing the risk of fuel procurement fluctuations, reducing energy waste, and ultimately lowering overall system costs.

[0067] 4. This paper analyzes the impact of the chance constraint confidence level and the number of random optimization scenarios on modeling results, providing a decision-making basis for practical applications. By adjusting the confidence level, a balance can be found between economy and reliability. Although increasing the number of random optimization scenarios will increase costs and computational complexity, it can more comprehensively describe the uncertainty of short-term fluctuations, enhance the system's operational adaptability, and enable the system to better cope with varying degrees of risk and uncertainty. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 is a flow chart of the method of the present invention;

[0069] Figure 2This is a schematic diagram of the framework of the electricity-hydrogen integrated energy system model in an embodiment of the present invention;

[0070] Figure 3 It is a two-stage uncertainty optimization framework in an embodiment of the present invention;

[0071] Figure 4 Schematic diagram of load curve decomposition in an embodiment of the present invention;

[0072] Figure 5 A schematic diagram of solving a model in an embodiment of the present invention;

[0073] Figure 6 The differential configuration of electrolyzer, fuel cell and battery power in three scenarios in the embodiment of the present invention;

[0074] Figure 7 is the difference between the hydrogen storage capacity and the battery capacity under the three scenarios in the embodiment of the present invention;

[0075] Figure 8 1 is the power configuration of system equipment at different confidence levels in the embodiment of the present invention.

[0076] Figure 9 1 shows the system equipment capacity configuration at different confidence levels in the embodiment of the present invention.

[0077] Figure 10 This is the charging and discharging condition of the hydrogen storage system in the chance constraint with a confidence level of 85% in the embodiment of the present invention.

[0078] Figure 11 This is the charging and discharging condition of the hydrogen storage system in the chance constraint with a confidence level of 90% in the embodiment of the present invention.

[0079] Figure 12 This is the charging and discharging condition of the hydrogen storage system in the chance constraint with a confidence level of 100% in the embodiment of the present invention.

[0080] Figure 13 Generates system device power configurations for different numbers of scenarios in an embodiment of the present invention.

[0081] Figure 14 Generates system equipment capacity configuration under different number of scenarios in the embodiment of the present invention. DETAILED DESCRIPTION

[0082] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0083] Example 1

[0084] This embodiment discloses a method for planning an electric-hydrogen integrated energy system taking into account multi-time scale uncertainties, aiming to optimize the configuration and scheduling strategy of energy storage equipment and improve the flexibility and reliability of the system. Figure 1 As shown, the method includes the following steps:

[0085] Step S1, establishing an electricity-hydrogen integrated energy system model;

[0086] Step S2: Based on the electric-hydrogen integrated energy system model, taking into account the uncertainty of wind and solar power output and load demand, a multi-time-scale uncertainty modeling method integrating chance constraints and stochastic optimization is used to construct a planning mathematical model for the electric-hydrogen integrated energy system;

[0087] Step S3: Based on the electricity-hydrogen integrated energy system model, establish an electricity market model as a lower-level optimization model of the planning mathematical model;

[0088] Step S4: Solve the planning mathematical model using a two-stage optimization method to obtain a planning solution.

[0089] In step S1, the electricity-hydrogen integrated energy system model framework is as follows Figure 2 As shown, it includes photovoltaic units, wind turbines, traditional generator sets, batteries, hydrogen storage systems, electrolyzers, hydrogen storage tanks and fuel cells. Among them, the batteries and hydrogen storage systems constitute a hybrid energy storage device, and the hydrogen storage system includes electrolyzers, fuel cells and hydrogen storage tanks.

[0090] The system's photovoltaic, wind, and conventional turbines are used to meet conventional electricity load demand. Hybrid energy storage, comprised of batteries and hydrogen storage systems, primarily addresses power imbalances on long and short timescales, respectively. The hydrogen storage system, primarily composed of an electrolyzer, fuel cell, and hydrogen storage tank, converts excess electricity into hydrogen, stores it, and then converts the hydrogen back into electricity when needed. This characteristic makes the hydrogen storage system particularly suitable for addressing power imbalances on long timescales and, in conjunction with batteries, can address fluctuations in electricity supply and demand over multiple timescales. As a market participant, the Electric-Hydrogen Integrated Energy System (EH-IES) optimizes its power purchasing strategy by participating in the electricity market and adjusts its operating mode to adapt to market price fluctuations and resource allocation. The system leverages market mechanisms to rationally dispatch electricity and hydrogen resources, improving overall economic efficiency, flexibility, and operational efficiency. It also collaborates with the power system to achieve efficient energy utilization and low-carbon operation. Under this framework, EH-IES optimizes its own charging and discharging strategies based on market clearing results to better respond to changes in electricity demand and reduce system operating costs.

[0091] Taking into account the uncertainty of wind and solar power output and load demand, this method proposes a multi-timescale uncertainty modeling approach that integrates chance constraints and stochastic optimization for the planning of integrated electricity-hydrogen energy systems. Based on the differences in energy cycle characteristics between hydrogen storage systems and batteries, this model utilizes hydrogen storage systems to address long-term imbalances in electricity demand and batteries to smooth short-term power fluctuations, thereby improving system stability and scheduling flexibility. Figure 3 The proposed two-stage uncertainty optimization framework is presented. In the first stage, a chance-constrained optimization approach is used to dispatch traditional power generation units and hydrogen storage systems to address the uncertainty of long-term power demand. This stage establishes a chance-constrained model by setting uncertainty confidence levels to ensure the system's reliability and economic efficiency in the long term. In the second stage, a stochastic optimization approach is used to construct a scenario-based optimal dispatch model to address the uncertainties of short-term load demand and wind and solar power output fluctuations. This model dynamically adjusts the power dispatch of batteries and traditional power generation units, thereby enhancing the system's short-term regulation capabilities and improving its flexibility and stability.

[0092] In step S2, if Figure 4 The power demand shown is divided into long-term power demand and short-term power fluctuations. The optimization model is based on the multi-time scale fluctuation characteristics of wind and solar power output. The specific expression of power demand includes:

[0093]

[0094] in, represents the long-term load demand; represents the original load demand; represents short-term electricity demand fluctuations; Indicates the long-term photovoltaic unit output; Indicates the raw photovoltaic output; Indicates short-term PV output fluctuations; Indicates the long-term wind turbine output; represents the original wind power output; Indicates short-term wind power output fluctuations.

[0095] The planning mathematical model includes an objective function and corresponding constraints. In this embodiment, the goal is to minimize the annual investment cost and the expected operating cost. The specific objective function is expressed as follows:

[0096] minC=C inv +C m +C fuel +C Q +C E

[0097] Among them, C is the objective function, C inv is the investment cost, C fuel is the fuel cost, C m is the maintenance cost, C Q is the penalty cost for curtailment of electricity, C E The cost of purchasing electricity for the system.

[0098] The investment cost expression is:

[0099]

[0100] Among them, r represents the discount rate, β represents the life of the equipment, α represents the maintenance cost coefficient, and c SHS,e 、c SHS,p They represent the SHS unit capacity investment cost and unit power investment cost respectively, c Bat,e 、c Bat,p They represent the battery unit capacity investment cost and unit power investment cost respectively, E SHS,max Indicates the upper limit of hydrogen storage system capacity, P el,max Indicates the upper limit of electrolytic cell power, P fc,max Indicates the upper limit of electrolytic cell power, E Bat,max Indicates the upper limit of battery capacity, P Bat,max Indicates the upper limit of battery power, P CU,max Indicates the upper limit of the output of traditional units, c CU Represents the unit power generation cost of traditional generator sets;

[0101] The expression of maintenance cost is:

[0102] C m =αC inv

[0103] Among them, c inv represents the investment cost;

[0104] The expression for fuel cost is:

[0105]

[0106] Where D represents the total number of typical days, D p is the total number of days in a year, W d Indicates the weights of different typical days, W s Indicates the probability of different scenarios, c fuel Indicates the fuel cost per unit output of traditional units; represents the total power generation of traditional units at time t on day d under scenario s;

[0107] The expression of curtailment penalty cost is:

[0108]

[0109] Among them, c Q represents the penalty cost corresponding to unit curtailed power, represents the total amount of power abandoned by the system at time t on day d under scenario s;

[0110] The expressions for electricity purchase costs are:

[0111]

[0112] Among them, c LS represents the marginal price of electricity in the electricity market, It represents the total power purchased by the system at time t on day d under scenario s.

[0113] In this embodiment, the constraints of the planning mathematical model include short-term power balance constraints, equipment constraints, long-term power balance constraints, conversion determinism constraints, investment constraints, and two-stage coordinated output constraints.

[0114] On a short time scale, the operation of the electric-hydrogen integrated energy system is affected by the uncertainty of wind and solar power output fluctuations and load demand changes. To this end, this embodiment constructs a short-term uncertainty constraint through a random optimization method. This constraint generates possible wind and solar power output and load demand situations based on multiple random scenarios, and requires the system to meet the power balance conditions in these scenarios. Specifically, the scheduling of short-term energy storage equipment (such as batteries) and traditional generator sets must meet the power constraints, energy storage charging and discharging constraints and equipment operation constraints under all random scenarios to ensure that the system can operate stably under different short-term uncertainty scenarios. Through this uncertainty constraint, the system can optimize energy scheduling within a short time scale, improve regulation capabilities, reduce operational deviations caused by short-term fluctuations, and enhance the stability and adaptability of the system.

[0115] In this embodiment, the expression of the short-term power balance constraint is:

[0116]

[0117] in, represents the short-term photovoltaic output fluctuation at time t on day d under scenario s, represents the short-term wind power output fluctuation at time t on day d under scenario s, represents the short-term electricity demand fluctuation at time t on day d under scenario s, represents the short-term traditional unit output at time t on day d under scenario s, represents the short-term electricity purchase amount at time t on day d under scenario s, represents the discharge power of the short-term battery at time t on day d under scenario s, represents the charging power of the short-term battery at time t on day d under scenario s, represents the power curtailment during the short-term power fluctuation phase at time t on day d under scenario s;

[0118] The expression of the device constraint is:

[0119]

[0120] in, represents the capacity of the battery at time t on day d under scenario s, Represents the battery's self-consumption rate, η bat,char Indicates the battery charging efficiency, η bat,dis Indicates the discharge efficiency of the battery, Indicates the charging mark of the battery at time t on day d in scenario s. Indicates the discharge mark of the battery at time t on day d under scenario s, E bat,min Indicates the lower limit of battery capacity, E bat,max Indicates the upper limit of battery capacity.

[0121] Over long timescales, the electric-hydrogen integrated energy system faces uncertainty in renewable energy output and load demand. Traditional deterministic optimization methods struggle to guarantee system reliability in long-term operation. To this end, a chance constraint approach is employed to construct long-term uncertainty constraints. By setting a confidence level, the system is able to meet long-term energy balance conditions within a certain probability range. Specifically, this constraint requires that the scheduling of hydrogen storage systems and traditional power generation units be able to meet long-term electricity demand in most cases, allowing only minor constraint violations in extreme cases to reduce unnecessary redundant energy storage configurations.

[0122] In this embodiment, the expression of the long-term power balance constraint is:

[0123]

[0124] in, Indicates the long-term photovoltaic unit output, Indicates the long-term wind turbine output, Indicates the output of traditional generator sets when meeting long-term power demand. It represents the amount of electricity purchased by the system to meet long-term electricity demand. Indicates the amount of power wasted during long-term fluctuations. represents the output of the electrolytic cell at time t on day d during the long-term energy fluctuation stage, represents the output of the fuel cell at time t on day d during the long-term energy fluctuation phase, represents the long-term load demand, and ε represents the risk factor of the system.

[0125] According to the Central Limit Theorem (CLT), when the number of samples of a group of independent random variables from the same distribution is large enough, the distribution of their means will tend to be normally distributed, even if the original data itself does not obey the normal distribution. In the calculation of the daily average value of wind and solar power output and load demand, if the daily data consists of multiple hourly sampling points and the output data at each moment are statistically relatively independent or weakly correlated, it can be approximately considered that the daily average value obeys the normal distribution. The expectation of the distribution is The variance of the distribution is In order to transform the chance constraint into a deterministic constraint, the probability quantile method is used to find the corresponding quantile using the confidence level 1-ε, thereby replacing the random variable with a linear combination of its mean and standard deviation. According to the properties of the normal distribution, the standardized quantile z ε To approximately replace the random variable, so as to express it as a deterministic equivalent form, thus, the above chance constraint can be transformed into a deterministic constraint by adopting the quantile of the probability distribution.

[0126] In this embodiment, the expression of the transformed deterministic constraint is:

[0127]

[0128]

[0129] Among them, Z ε represents the quantile with a confidence level of 1-ε under the normal distribution, represents the average value of long-term photovoltaic output, represents the average value of long-term wind power output, represents the average value of long-term load demand, σ Δp represents the standard deviation, represents the variance of long-term photovoltaic output, represents the variance of long-term wind power output, Represents the variance of long-term load demand.

[0130] In this embodiment, other constraints, such as investment constraints, are expressed as follows:

[0131]

[0132] in, is the total power generation of traditional units at time t on day d under scenario s, represents the total amount of power abandoned by the system at time t on day d under scenario s;

[0133] The expression of the two-stage coordinated output constraint is:

[0134]

[0135] In step S3, the electricity market model is led by an independent power system operator, which optimizes the electricity purchase amount of the electricity-hydrogen integrated energy system through market clearing, and feeds the market clearing results back to the planning mathematical model.

[0136] The electricity market model, as a lower-level optimization model, is led by an independent power system operator. It optimizes the system's electricity purchase volume through market clearing and feeds the market clearing results back to the upper-level model. Through market mechanisms, resources are effectively allocated, thereby improving the economy, flexibility, and operational efficiency of the overall system. Specifically, the participants in the day-ahead electricity market are composed of coal-fired generation units (CG), large-scale electricity users (EU), and the Electric-Hydrogen Integrated Energy System (EH-IES) as consumers. The structure diagram is shown in Figure 5. In market operation, the CG group and EU provide electricity production and consumption needs, while the EH-IES, as a consumer, further optimizes its charging and discharging strategies to cope with fluctuations in electricity demand and price changes.

[0137] In the electricity market model, the independent system operator (ISO) is responsible for clearing electricity prices, with the goal of maximizing social welfare. The main participants in the model are the CG, EU, and the EH-IES, which act as consumers. The CG provides electricity, aiming to maximize profits through market transactions while ensuring power balance. The EH-IES can utilize long- and short-term energy storage to respond to market price fluctuations. This electricity market model uses formulas to constrain the purchase volumes of the CG, EU, and EH-IES, ensuring stable operation and efficient dispatch of the electricity market.

[0138] In this embodiment, the expression of the electricity market model is:

[0139]

[0140]

[0141] Among them, c cg represents the bid price of CG, represents the output power of CG at time t on day d under scenario s, represents the total system power purchase at time t on day d under scenario s, C e represents the bidding price in the electricity market, C eu represents the bid price of EU, D represents the electricity consumption of EU at time t on day d under scenario s. p is the total number of days in a year, represents the climbing limit of CG, λ it Lagrange multipliers representing equality constraints, and Lagrange multipliers representing inequality constraints.

[0142] The above formulas respectively minimize the operating costs of CG, EU and the electric-hydrogen integrated energy system in the system (equivalently representing the maximization of social welfare); ensure the power balance of the system; impose actual output and ramp limits on CG; limit the maximum amount of electricity purchased by the system; and limit the maximum amount of electricity purchased by EU.

[0143] In this embodiment, the solution is as follows Figure 5 As shown in the figure, the two-stage uncertainty programming model contains integer variables and continuous variables, accompanied by nonlinear constraints; the electricity market model is a mixed integer linear programming problem. Due to the coupling relationship between the upper layer (planning mathematical model) and the lower layer model (electricity market model), it is difficult for commercial solvers to solve it directly. By constructing the Lagrangian function of the lower layer electricity market model and based on the KKT complementary relaxation condition of the lower layer model, the lower layer model is equivalently converted into the constraints of the upper layer model. The Big-M method is then used to linearize the nonlinear terms in the converted single-layer nonlinear model, ultimately forming a single-layer mixed integer linear programming problem.

[0144] In step S4, the process of solving the planning mathematical model includes:

[0145] Construct the Lagrangian function of the power market model and, based on the KKT constraints, convert the power market model into the constraints of the planning mathematical model to form a single-layer nonlinear model with equilibrium constraints.

[0146] The nonlinear terms in the single-layer nonlinear model are linearized using the Big-M method. The single-layer nonlinear model is transformed into a single-layer mixed integer linear programming problem and solved to obtain the planning scheme.

[0147] The mathematical planning model proposed in this paper aims to optimize the energy storage configuration within the system while also considering the interactions between the power market and the system. The upper-level model primarily focuses on energy storage system configuration, while the power market model determines power purchases through power clearing. The difficulty lies in coupling the system with the power market. Because the power market clearing model exhibits convexity, the optimality conditions for the power market model can be formulated using KKT conditions, effectively transforming the original two-level optimization model into a mathematical programming model with equilibrium constraints (MPEC).

[0148] KKT constraints include:

[0149]

[0150]

[0151]

[0152] Among them, c cg represents the bidding price of CG, C e represents the bidding price in the electricity market, C eu represents the bid price of EU, λ it Lagrange multipliers representing equality constraints, and Lagrange multipliers representing inequality constraints, represents the output power of CG at time t on day d under scenario s, represents the total system power purchase at time t on day d under scenario s, represents the electricity consumption of EU at time t on day d under scenario s, D p is the total number of days in a year, Indicates the climbing limit of CG, P cg,max Indicates the maximum value of CG output power, P cg,min Indicates the minimum value of CG output power, P buy,max Indicates the maximum amount of electricity purchased, P buy,min Indicates the minimum value of electricity purchase, P eu,min Indicates the minimum value of EU power consumption, P eu,max Indicates the maximum value of EU electric power consumption.

[0153] The Big-M method is used to linearize the nonlinear constraints in the model. The Big-M method transforms the nonlinear constraints into linear form by introducing binary decision variables, thus transforming the problem into a mixed integer linear programming (MILP) problem.

[0154] The constraints on the charging and discharging of hydrogen storage systems and battery equipment are nonlinear. By introducing several binary variables, the original nonlinear constraints are equivalently converted into mixed integer linear constraints.

[0155] In this embodiment, the Big-M method is used to linearize the nonlinear constraints in the model. The solution expression is:

[0156]

[0157] Example 2

[0158] This embodiment is based on the above-mentioned embodiment 1, and selects a certain industrial park as a research object to perform case analysis to verify the effectiveness of the obtained planning scheme.

[0159] The annual wind and solar power generation forecast data and load forecast data used in the example are from existing datasets. The relevant parameters of each device are shown in Tables 1 and 2.

[0160] Table 1 Parameters related to electricity-hydrogen comprehensive energy

[0161]

[0162] Table 2 Energy conversion equipment parameters

[0163]

[0164] In order to verify the effectiveness of the two-stage uncertainty modeling method proposed in Example 1, this embodiment designed three comparative scenarios for analysis. The first scenario adopts a deterministic modeling method, without considering the uncertainty of renewable energy output and load fluctuations, to evaluate the system operation under uncertainty-free modeling conditions. The second scenario only considers the opportunity constraint modeling of the first stage, and describes the uncertainty impact of the system on the long-term scale through constraints under a certain confidence level, so as to analyze its role in system investment and operation optimization. The third scenario adopts a complete multi-time scale uncertainty modeling method, comprehensively considering the opportunity constraint optimization of the first stage and the random optimization strategy of the second stage, to comprehensively evaluate the uncertainty impact of the system on different time scales and its optimization effect, as shown in Table 3. By comparing and analyzing the above three scenarios, the advantages of the proposed method in improving the economy, reliability and robustness of the system are explored.

[0165] Table 3 Settings for scenarios 1 to 3

[0166]

[0167] Next, we analyze the planning costs and system configuration results under different scenarios to illustrate the reliability and economy of the two-stage uncertainty modeling. The specific results are shown in Table 4 below.

[0168] Table 4 Cost of electricity-hydrogen integrated energy system under different scenarios

[0169]

[0170] The comparative results in Table 4 show that different uncertainty handling strategies significantly impact the annual cost of the electric-hydrogen integrated energy system. Scenario 1 employs a deterministic constraint model without incorporating any stochastic or probabilistic conditions. This results in a conservative system design with a safety margin, requiring the reservation of excess redundant capacity to mitigate potential risks. This in turn increases system investment, fuel, and electricity purchase costs. In contrast, Scenario 2 introduces chance constraints in the first stage, allowing key constraints to be satisfied with a preset confidence level. Using probabilistic thresholds, random variables are converted into deterministic equivalents. This optimizes resource allocation while ensuring system robustness, reduces redundancy requirements, and reduces total costs by approximately 17.3% compared to Scenario 1. Scenario 3 builds on Scenario 2 by applying stochastic optimization. This not only minimizes expected costs under probabilistic constraints but also dynamically coordinates the sequential operation of the electric-hydrogen integrated energy system through stochastic programming techniques. This leverages the output of renewable energy, the flexibility of hydrogen storage, and the stochastic nature of market prices, improving energy efficiency while mitigating the risk of fuel procurement fluctuations. Although the total cost of Scenario 3 increases compared to Scenario 2, it remains lower than Scenario 1 and achieves more efficient renewable energy utilization. Therefore, this strategy can more effectively reduce energy waste and improve the overall low-carbon nature and sustainability of the system while taking into account both economic efficiency and system operation safety.

[0171] Figure 6 The differential configuration of electrolyzer, fuel cell, and battery power under the three scenarios is presented. The electrolyzer and fuel cell power in Scenario 2 and Scenario 3 are both lower than that in Scenario 1. This is because under the opportunity-constrained optimization framework, the system allows a certain degree of supply and demand imbalance, thereby reducing the need for capacity over-allocation in extreme situations and thus reducing the scale of investment in electrolyzers. The battery power configuration is the highest in Scenario 3 and relatively low in Scenario 1 and Scenario 2. This result shows that under the stochastic optimization framework, the role of batteries is more prominent, mainly used to smooth short-term fluctuations and reduce the system's dependence on traditional units. In contrast, in Scenario 1 and Scenario 2, due to the limited ability of the system to adjust to short-term fluctuations, the battery power configuration is lower to avoid excessive investment costs.

[0172] Figure 7The differences between hydrogen storage capacity and battery capacity under the three scenarios are presented. Hydrogen storage capacity remains stable at approximately 8,000 kWh across all scenarios, primarily due to its high energy density and long-term storage characteristics, effectively balancing seasonal fluctuations in renewable energy output with long-term load demand. In contrast, hydrogen storage capacity in Scenario 2 is slightly lower than in Scenario 1. This is primarily because the chance constraint allows for defaults within a certain probability range, thereby reducing the system's stringent requirements for extreme conditions and the need for hydrogen storage capacity. Battery capacity, on the other hand, increases with the evolution of the optimization approach, with Scenario 3 achieving the highest battery capacity. This trend indicates that, when accounting for short-term fluctuations, the system relies on battery storage to provide short-term regulation to cope with the uncertainty of wind and solar power output and rapid load fluctuations. In contrast, in Scenario 1, the system's demand for batteries is lower because output uncertainty is not accounted for. In Scenario 2, the battery capacity increases somewhat because some uncertainty is absorbed by the probabilistic constraint, but remains lower than in Scenario 3.

[0173] In the two-stage modeling approach that accounts for uncertainty, the first stage employs chance constraints, eliminating the need to configure redundant equipment for all possible extreme scenarios. Instead, defaults are allowed within a certain probability range. This strategy reduces overinvestment in equipment such as hydrogen energy storage, thereby lowering the overall system investment cost. The second stage further implements stochastic optimization to address short-term fluctuations. By dynamically adjusting energy storage charging and discharging, fuel cell operation, and external power purchase strategies, the system can more efficiently adapt to short-term random fluctuations. This flexible scheduling approach avoids the response lag caused by overreliance on long-term energy storage in deterministic scenarios, enabling the system to more accurately match real-time load demand. Two-stage optimization ensures system reliability while reducing unnecessary energy waste. Compared to deterministic modeling, accounting for uncertainty allows the system to more effectively coordinate wind and solar power output with load demand. Furthermore, by combining chance constraints with stochastic optimization, the system can more efficiently utilize renewable energy, reduce the output of traditional units, and improve overall energy efficiency. In summary, compared to deterministic optimization, two-stage uncertainty modeling can reduce investment costs, increase energy scheduling flexibility, reduce renewable energy waste, and ensure system operation is more aligned with actual needs, all while ensuring system reliability.

[0174] To further validate the effectiveness of the first-stage chance-constrained uncertainty modeling, this section sets up chance-constrained planning scenarios with different confidence levels—85%, 90%, and 98%—to analyze the impact of varying risk tolerances on system optimization results. By comparing system planning costs and energy storage configurations at different confidence levels, we evaluate the adaptability and optimization effectiveness of the chance-constrained approach under varying reliability requirements.

[0175] Table 5 Economic analysis of different confidence levels

[0176]

[0177] The results in Table 5 show that the total cost of the system increases with increasing confidence levels. This is primarily because higher confidence levels correspond to stricter reliability requirements, meaning the system must meet constraints with a lower probability of default. Therefore, to reduce the risk of supply-demand imbalance, the system tends to increase the capacity of energy storage equipment and the output of traditional units at high confidence levels to provide more robust dispatch capabilities, resulting in higher system investment and fuel costs. This result shows that increasing the confidence level can enhance system reliability, but it also comes at a higher economic cost. Therefore, in actual planning, it is necessary to find a reasonable balance between economy and reliability to optimize the overall benefits of system operation.

[0178] As the confidence level increases, the system's robustness requirements for uncertainty during the optimization process also increase, that is, it is necessary to ensure that the system can meet the balance of power supply and demand under stricter constraints. Figure 8 and Figure 9 The results show that at higher confidence levels, the system requires greater energy storage capacity and power to mitigate the risks associated with renewable energy output fluctuations and load uncertainty. This trend suggests that while lower confidence levels can reduce investment costs, they may lead to unreliable system operation under high volatility conditions. Higher confidence levels, on the other hand, can improve system safety and feasibility but increase equipment costs. This further demonstrates the need to balance economic efficiency with reliability during system planning, selecting an appropriate confidence level to optimize system performance.

[0179] Figures 10 to 12Figures 1 and 2 show the charging and discharging of the hydrogen storage system under chance constraints at confidence levels of 85%, 90%, and 98%, respectively. At higher confidence levels, the system needs to ensure it can meet load demand under more extreme uncertainties. Therefore, to mitigate potential fluctuations in renewable energy, the hydrogen storage charging power increases compared to lower confidence levels, providing a more stable dispatch margin in the future. At lower confidence levels, the system's tolerance for uncertainty is higher, resulting in smaller hydrogen storage charging and discharging amplitudes, indicating a greater reliance on external power supply rather than deep regulation through hydrogen storage. When the confidence level increases to 90%, the hydrogen storage charging and discharging power increases, reflecting a greater preference for using hydrogen storage to buffer against uncertainties in wind and solar power output and load when faced with tighter supply-demand balance constraints. When the confidence level is further increased to 98%, the hydrogen storage charging and discharging power reaches its maximum, and the range of charging and discharging fluctuations increases significantly, indicating that the system needs to more actively utilize hydrogen storage to ensure supply-demand balance under extreme conditions. Therefore, a higher confidence level requires the system to operate under stricter safety constraints, prompting hydrogen storage to assume greater regulatory responsibility, thereby improving the utilization rate of energy storage equipment, but at the same time may also lead to higher energy storage investment and operating costs.

[0180] To further validate the effectiveness of the second-stage stochastic optimization, this example constructs a system model by generating different numbers of scenarios and evaluating their impact on the optimization results. Specifically, 5, 10, and 15 random scenarios are generated to characterize the uncertainty of wind and solar power output and load. By comparing system planning costs and energy storage configurations under different numbers of scenarios, the adaptability and optimization effectiveness of the stochastic optimization method are evaluated with varying degrees of uncertainty characterization accuracy.

[0181] Table 6 Economic analysis under different number of scenarios

[0182]

[0183] The results in Table 6 show that as the number of random optimization scenarios increases, the total system planning cost shows an upward trend. With a larger number of scenarios, the model can more comprehensively characterize the uncertainty of renewable energy output and load demand, requiring higher investment costs to enhance system robustness. In addition, maintenance and fuel costs also increase accordingly, reflecting that under stricter uncertainty constraints, the system needs to more frequently call on flexible adjustment resources, such as traditional units and energy storage equipment, to ensure that supply and demand balance can be met in different scenarios. In summary, although increasing the number of scenarios can improve the system's adaptability to uncertainty, it also leads to higher investment and operating costs. Therefore, in practical applications, a trade-off needs to be made between computational complexity, economy, and system reliability.

[0184] Figures 13 to 14The results show that as the number of scenarios generated by stochastic optimization increases, the power and capacity configurations of system equipment generally show an upward trend, especially the battery capacity and power configurations, which increase significantly. This is mainly because the second-stage stochastic optimization models the uncertainty of short-term energy storage. As the number of scenarios increases, the optimization model can more accurately characterize the fluctuation characteristics of wind power, photovoltaic power, and load, thereby making the compensation requirements for short-term power gaps more stringent. In this case, in order to improve the flexibility and regulation capabilities of the system, the power and capacity configuration of batteries, as the main equipment for short-term energy storage, are significantly increased to better cope with short-term power fluctuations. In addition, as the number of scenarios increases, the system optimization results tend to be conservative. That is, in order to adapt to more possible supply and demand fluctuations, the system tends to increase the battery configuration scale to reduce the scheduling risk caused by insufficient short-term energy storage. Therefore, compared with other equipment, the increase in battery power and capacity is the most obvious.

[0185] In summary, this example verifies the effectiveness of the electric-hydrogen integrated energy system planning method considering multi-time-scale uncertainties in Example 1 in improving system economy and renewable energy utilization by comparing different uncertainty modeling methods, and explores the influence of key modeling parameters on the optimization results.

[0186] Example 3

[0187] This embodiment, based on the above-mentioned embodiment 1, discloses a method for planning an electric-hydrogen integrated energy system considering multi-time-scale uncertainties, including:

[0188] Electricity-hydrogen integrated energy system module: establishes an electricity-hydrogen integrated energy system model;

[0189] Planning mathematical model construction module: Based on the electric-hydrogen integrated energy system model, taking into account the uncertainty of wind and solar power output and load demand, a multi-time-scale uncertainty modeling method that integrates chance constraints and stochastic optimization is used to construct a planning mathematical model for the electric-hydrogen integrated energy system. Based on the electric-hydrogen integrated energy system model, an electricity market model is established as the underlying optimization model of the planning mathematical model;

[0190] Solution module: Use a two-stage optimization method to solve the planning mathematical model and obtain the planning solution.

[0191] The specific details of the above modules can be understood by referring to the relevant descriptions and effects in Example 1.

[0192] If the above functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0193] It will be understood by those skilled in the art that the embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of the present invention may be implemented in various computer languages, for example, the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0194] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0195] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0196] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0197] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

[0198] The foregoing description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art will readily conceive of various equivalent modifications or substitutions within the technical scope disclosed herein, and such modifications or substitutions are intended to be encompassed within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be subject to the scope of protection of the claims.

Claims

1. A method for planning an integrated electricity-hydrogen energy system considering multi-time scale uncertainties, characterized in that: The method comprises the following steps: Establish an electricity-hydrogen integrated energy system model; Based on the above-mentioned electric-hydrogen integrated energy system model, considering the uncertainty of wind and solar power output and load demand, a multi-time-scale uncertainty modeling method integrating chance constraints and stochastic optimization is used to construct a planning mathematical model for the electric-hydrogen integrated energy system; Based on the electricity-hydrogen integrated energy system model, an electricity market model is established as a lower-level optimization model of the planning mathematical model; A two-stage optimization method is used to solve the planning mathematical model to obtain a planning solution.

2. The method for planning an electric-hydrogen integrated energy system considering multi-time scale uncertainties according to claim 1, characterized in that: The electric-hydrogen integrated energy system model includes a photovoltaic unit, a wind turbine unit, a traditional generator unit, a battery, a hydrogen storage system, an electrolyzer, a hydrogen storage tank and a fuel cell, wherein the battery and the hydrogen storage system constitute a hybrid energy storage device, and the hydrogen storage system includes an electrolyzer, a fuel cell and a hydrogen storage tank.

3. The method for planning an electric-hydrogen integrated energy system considering multi-time scale uncertainties according to claim 1, characterized in that: The planning mathematical model includes an objective function and corresponding constraints. The objective function aims to minimize the annual investment cost of the electricity-hydrogen integrated energy system. The expression of the objective function is: minC=C inv +C m +C fuel +C Q +C E Among them, C is the objective function, C inv is the investment cost, C fuel is the fuel cost, C m is the maintenance cost, C Q is the penalty cost for curtailment of electricity, C E The cost of purchasing electricity for the system.

4. The method for planning an electric-hydrogen integrated energy system considering multi-time scale uncertainty according to claim 3, characterized in that: The expression of the investment cost is: Among them, r represents the discount rate, β represents the life of the equipment, α represents the maintenance cost coefficient, and c SHS,e 、c SHS,p They represent the SHS unit capacity investment cost and unit power investment cost respectively, c Bat,e 、c Bat,p They represent the battery unit capacity investment cost and unit power investment cost respectively, E SHS,max Indicates the upper limit of hydrogen storage system capacity, P el,max Indicates the upper limit of electrolytic cell power, P fc ,max Indicates the upper limit of electrolytic cell power, E Bat,max Indicates the upper limit of battery capacity, P Bat,max Indicates the upper limit of battery power, P CU,max Indicates the upper limit of the output of traditional units, c CU Represents the unit power generation cost of traditional generator sets; The expression of the maintenance cost is: C m =αC inv Among them, c inv represents the investment cost; The fuel cost is expressed as: Where D represents the total number of typical days, D p is the total number of days in a year, W d Indicates the weights of different typical days, W s Indicates the probability of different scenarios, c fuel Indicates the fuel cost per unit output of traditional units; represents the total power generation of traditional units at time t on day d under scenario s; The expression of the curtailment penalty cost is: Among them, c Q represents the penalty cost corresponding to unit curtailed power, represents the total amount of power abandoned by the system at time t on day d under scenario s; The expressions of the electricity purchase cost are: Among them, c LS represents the marginal price of electricity in the electricity market, It represents the total power purchased by the system at time t on day d under scenario s.

5. The method for planning an electric-hydrogen integrated energy system considering multi-time scale uncertainties according to claim 3, characterized in that: The constraints include short-term power balance constraints, equipment constraints, long-term power balance constraints, conversion determinism constraints, investment constraints and two-stage coordinated output constraints.

6. The method for planning an electric-hydrogen integrated energy system considering multi-time scale uncertainties according to claim 5, characterized in that: The expression of the short-term power balance constraint is: in, represents the short-term photovoltaic output fluctuation at time t on day d under scenario s, represents the short-term wind power output fluctuation at time t on day d under scenario s, represents the short-term electricity demand fluctuation at time t on day d under scenario s, represents the short-term traditional unit output at time t on day d under scenario s, represents the short-term electricity purchase amount at time t on day d under scenario s, represents the discharge power of the short-term battery at time t on day d under scenario s, represents the charging power of the short-term battery at time t on day d under scenario s, represents the power curtailment during the short-term power fluctuation phase at time t on day d under scenario s; The expression of the device constraint is: in, represents the capacity of the battery at time t on day d under scenario s, Represents the battery's self-consumption rate, η bat,char Indicates the battery charging efficiency, η bat,dis Indicates the discharge efficiency of the battery, Indicates the charging mark of the battery at time t on day d in scenario s. Indicates the discharge mark of the battery at time t on day d under scenario s, E bat,min Indicates the lower limit of battery capacity, E bat,max Indicates the upper limit of battery capacity; The expression of long-term power balance constraint is: in, Indicates the long-term photovoltaic unit output, Indicates the long-term wind turbine output, Indicates the output of traditional generator sets when meeting long-term power demand. It represents the amount of electricity purchased by the system to meet long-term electricity demand. Indicates the amount of power wasted during long-term fluctuations. It represents the output of the electrolytic cell at time t on day d during the long-term energy fluctuation stage. represents the output of the fuel cell at time t on day d during the long-term energy fluctuation phase, represents the long-term load demand, and ε represents the risk factor of the system; The expression of the transformed deterministic constraint is: Among them, Z ε represents the quantile with a confidence level of 1-ε under the normal distribution, represents the average value of long-term photovoltaic output, represents the average value of long-term wind power output, represents the average value of long-term load demand, σ Δp represents the standard deviation, represents the variance of long-term photovoltaic output, represents the variance of long-term wind power output, represents the variance of long-term load demand; The investment constraint is expressed as: in, is the total power generation of traditional units at time t on day d under scenario s, represents the total amount of power abandoned by the system at time t on day d under scenario s; The expression of the two-stage coordinated output constraint is:

7. The method for planning an electric-hydrogen integrated energy system considering multi-time scale uncertainties according to claim 1, characterized in that: The electricity market model is led by an independent power system operator, which optimizes the electricity purchase volume of the electricity-hydrogen integrated energy system through market clearing, and feeds the market clearing results into the planning mathematical model.

8. The method for planning an electric-hydrogen integrated energy system considering multi-time scale uncertainties according to claim 1, characterized in that: The expression of the electricity market model is: Among them, c cg represents the bid price of CG, represents the output power of CG at time t on day d under scenario s, represents the total system power purchase at time t on day d under scenario s, C e represents the bidding price in the electricity market, C eu represents the bid price of EU, represents the electricity consumption of EU at time t on day d under scenario s, D p is the total number of days in a year, represents the climbing limit of CG, λ it Lagrange multipliers representing equality constraints, and Lagrange multipliers representing inequality constraints.

9. The method for planning an electric-hydrogen integrated energy system considering multi-time scale uncertainties according to claim 1, characterized in that: The process of solving the planning mathematical model includes: Constructing a Lagrangian function of the electricity market model, and based on the KKT constraints, converting the electricity market model into constraints of a planning mathematical model to form a single-layer nonlinear model with equilibrium constraints; The nonlinear terms in the single-layer nonlinear model are linearized using the Big-M method, and the single-layer nonlinear model is converted into a single-layer mixed integer linear programming problem and solved to obtain a planning solution.

10. The method for planning an electric-hydrogen integrated energy system considering multi-time scale uncertainties according to claim 9, characterized in that: The KKT constraints include: Among them, c cg represents the bidding price of CG, C e represents the bidding price in the electricity market, C eu represents the bid price of EU, λ it Lagrange multipliers representing equality constraints, and Lagrange multipliers representing inequality constraints, represents the output power of CG at time t on day d under scenario s, represents the total system power purchase at time t on day d under scenario s, represents the electricity consumption of EU at time t on day d under scenario s, D p is the total number of days in a year, Indicates the climbing limit of CG, P cg,max Indicates the maximum value of CG output power, P cg,min Indicates the minimum value of CG output power, P buy,max Indicates the maximum amount of electricity purchased, P buy ,min Indicates the minimum value of electricity purchase, P eu,min Indicates the minimum value of EU power consumption, P eu,max Indicates the maximum value of EU electric power consumption.

Images