Off-grid hydrogen electricity storage system capacity configuration method considering seasonal hydrogen storage

Abstract

The invention belongs to the technical field of photovoltaic power generation hydrogen production, and particularly relates to an off-grid hydrogen electricity storage system capacity configuration method considering seasonal hydrogen storage. The method comprises the following steps: firstly, acquiring photovoltaic power generation data of historical years, dividing according to seasons and performing data enhancement; clustering photovoltaic power generation data of each season, dividing photovoltaic power generation into a high state, a middle state and a low state, and counting the proportion of each state in each season as the initial probability of each state in each season; constructing a photovoltaic power generation state space based on a Markov model to obtain a state transition probability matrix of each season; obtaining a state probability vector updating formula according to the state transition matrix, and calculating the weight of photovoltaic power generation in various states in each day in each season; then, considering the weights of photovoltaic power generation in various states in various seasons, and constructing an objective function and constraint conditions of capacity configuration of the hydrogen electricity storage system; and finally, considering constraint conditions, solving the target function of capacity configuration of the hydrogen electricity storage system to obtain a Pareto front solution set, obtaining an optimal solution from the Pareto front solution set, and further obtaining the optimal planning capacity of each device. According to the method, the uncertainty of photovoltaic output is accurately described, the restriction of photovoltaic power generation on other equipment is considered, and the accuracy of capacity configuration is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of photovoltaic power generation and hydrogen production, and specifically is a capacity configuration method for an off-grid hydrogen storage system taking seasonal hydrogen storage into consideration. Background Art

[0002] Photovoltaic power generation is considered to be one of the main sources of renewable energy generation due to its clean, large-scale, and noise-free characteristics. However, photovoltaic power generation has obvious seasonality and intermittency. For example, light resources fluctuate seasonally, and in extremely bad weather, it may not be able to output electricity for many consecutive days, making it difficult to ensure the balance between supply and demand. In order to increase the penetration rate of photovoltaic power generation, the use of energy storage strategies can solve the intermittent and seasonal problems of photovoltaic power generation. Hydrogen energy can also be used as an energy storage carrier. Hydrogen energy storage has the advantages of cross-seasonal, cross-regional and large-scale long-term storage. Its storage scale ranges from hundreds of kilowatts to gigawatts. At the same time, it has a certain rapid response capability. It perfectly complements electrochemical energy storage in terms of time and scale, and can effectively make up for the shortcomings of electrochemical energy storage. It has strong application value.

[0003] At present, the production and operation research of hydrogen-electricity-storage coupling systems mainly focuses on the consumption and peak regulation of hydrogen energy for renewable energy such as wind and solar power at a certain time scale, and does not consider hydrogen energy as a commodity that can be sold in large quantities. The production and operation research of large-scale hydrogen-electricity-storage coupling systems under off-grid conditions needs to consider many factors, including the uncertainty of photovoltaic output, the uncertainty of external load demand of the system, and the reasonable and efficient arrangement of energy production plans. Conducting multi-time scale capacity optimization and production planning research under many uncertain factors will help to improve energy penetration and increase net present value benefits on the basis of adjusting and reducing the total cost of the system. To this end, the present invention proposes a capacity configuration method for an off-grid hydrogen-electricity storage system considering seasonal hydrogen storage. Summary of the invention

[0004] In view of the deficiencies in the prior art, the technical problem that the present invention intends to solve is to provide a capacity configuration method for an off-grid hydrogen-electric storage system taking into account seasonal hydrogen storage.

[0005] The present invention solves the technical problem by adopting the following technical solution:

[0006] A method for configuring the capacity of an off-grid hydrogen-electricity storage system considering seasonal hydrogen storage, characterized in that the method comprises the following steps:

[0007] Step 1: Obtain the photovoltaic power generation data of historical years, divide it by season and perform data enhancement; cluster the photovoltaic power generation data of each season, divide photovoltaic power generation into three states: high, medium and low, and count the proportion of each state in each season as the initial probability of each state in each season; construct the photovoltaic power generation state space based on the Markov model, and obtain the state transition probability matrix of each season, where the state transition probability matrix P of season s is (s) It is expressed as:

[0008]

[0009] In the formula, represents the probability of photovoltaic power generation transferring from state i to state o in season s, s = 1, 2, 3, 4 represent spring, summer, autumn, and winter respectively;

[0010] The state probability vector is used to describe the probability of photovoltaic power generation being in various states on each day of the season. The state probability vector of season s is It is expressed as:

[0011]

[0012] In the formula, represents the probability that the photovoltaic power generation is in state o on the tth day in season s, Representing high, medium and low states respectively;

[0013] According to the state transition probability matrix of season s, the state probability vector of season s is obtained The update formula is:

[0014]

[0015] In the formula, represents the state probability vector of day t in season s;

[0016] According to the state probability vector of season s, the weights of photovoltaic power generation in various states on each day in season s are calculated, and the weight λ of photovoltaic power generation in state o on the tth day in season s is s,o (t) is calculated by the following formula:

[0017]

[0018] In the formula, λ s,o (0) is the initial probability of state o in season s;

[0019] The weighted average of the weights of photovoltaic power generation in the same state on each day of the season is obtained to obtain the average weight of each state in the season. The average weight of state o in season s is Calculated by the following formula:

[0020]

[0021] Where, T s Indicates the number of days included in season s;

[0022] Step 2: Construct the objective function of hydrogen storage system capacity configuration, including planning and design cost function, operation and maintenance cost function and net present value benefit function;

[0023] The planning and design cost function is:

[0024] minF1=C inv +C rep (6)

[0025] Where F1 is the planning and design cost of the system, C inv is the initial construction cost of the system, C rep It is the cost of updating the system after it is put into use;

[0026] The operating cost function is:

[0027] minF2=C op =∑ m C op,m ; m∈{pv,ele,fc,h,ba} (10)

[0028] In the formula, C op,m is the operation and maintenance cost of equipment m, where pv,ele,fc,h,ba represent the photovoltaic generator set, electrolyzer, fuel cell, hydrogen storage system and battery respectively;

[0029] The operation and maintenance cost C of device m op,m Calculated by the following formula:

[0030]

[0031] O m,s,o =N m ·θ m,s,o ; m∈{pv,ele,fc,h,ba} (12)

[0032] Where, L plan represents the design service life of the system, T y Indicates the number of days in a year, O m,s,o represents the daily production capacity of equipment m in state o in season s, φ m Represents the operation and maintenance cost of equipment m per unit output, N m is the planned capacity of device m, θ m,s,o is the efficiency of equipment m in state o in season s;

[0033] The net present value benefit function is:

[0034]

[0035] In the formula, R l is the total system revenue in year l, r is the discount rate, C op It is the system operation and maintenance cost;

[0036] Total system revenue R in year l l Calculated by the following formula:

[0037] R l =R H2,l +R elec,l +R carbon,l (14)

[0038] In the formula, R H2,l , R elec,l and R carbon,l They are the hydrogen sales revenue, electricity sales revenue and carbon emission rights trading revenue in the first year respectively;

[0039] The income from hydrogen sales in the first year is R H2,l Calculated by the following formula:

[0040]

[0041] In the formula, π H2,l is the hydrogen sales coefficient in the first year, and its value range is (0,1); H2,s,o is the hydrogen production of the electrolyzer in state o in season s, V H2,l is the hydrogen selling price in year l;

[0042] The electricity sales revenue in the first year is R elec,l Calculated by the following formula:

[0043]

[0044] In the formula, π elec,l is the electricity sales coefficient in the first year, and its value range is (0,1); pv,s,o is the photovoltaic power generation in state o in season s, V elec,l represents the electricity selling price in year l;

[0045] The carbon emission trading income R in the first year carbon,l Calculated by the following formula:

[0046]

[0047] In the formula, π carbon,l Represents the electricity-to-carbon conversion coefficient, V carbon,l represents the carbon emission rights trading price in year l;

[0048] Step 3: Taking into account the constraints, solve the objective function of the capacity configuration of the hydrogen storage system to obtain the Pareto frontier solution set, obtain the optimal solution from the Pareto frontier solution set, and then obtain the optimal planned capacity of each equipment.

[0049] Furthermore, the initial construction cost of the system C inv Calculated by the following formula:

[0050] C inv =∑ m N m β m ;m∈{pv,ele,fc,h,ba} (7)

[0051] In the formula, β m is the construction cost per unit capacity of equipment m;

[0052] Update cost after the system is put into use C rep Calculated by the following formula:

[0053] C rep =∑ m N m η m ;m∈{pv,ele,fc,h,ba} (8)

[0054]

[0055] Where η m is the update cost per unit capacity of device m, L m Represents the expected service life of device m.

[0056] Furthermore, the constraints include constraints on photovoltaic power generation units, electrolyzers, fuel cells, hydrogen storage systems, batteries, and supply and demand balance constraints on electricity and hydrogen energy;

[0057] PV generator constraints include:

[0058]

[0059] 0≤O pv,s,o ≤N pv (19)

[0060] O pv,s,o ≤OPV actual (20)

[0061] Where N pv is the planned capacity of the photovoltaic generator set, and Respectively represent the upper and lower boundaries of the planned capacity of the photovoltaic power generation unit, OPV actual It indicates the available photovoltaic power generation;

[0062] Electrolyser constraints include:

[0063]

[0064] Where N ele is the planned capacity of the electrolyzer, and Respectively represent the upper and lower boundaries of the planned capacity of the electrolyzer, θ ele,s,o is the efficiency of the electrolyzer in state o in season s, and They represent the upper and lower bounds of the efficiency of the electrolyzer in state o in season s, respectively;

[0065] Fuel cell constraints include:

[0066]

[0067] Where N fc is the planned capacity of the fuel cell, and Respectively represent the upper and lower boundaries of the planned capacity of the electrolyzer, θ fc,s,o is the efficiency of the fuel cell in state o in season s, and They represent the upper and lower bounds of the efficiency of the fuel cell in state o in season s, respectively;

[0068] The constraints of the hydrogen storage system are:

[0069]

[0070] Where N h is the planned capacity of the hydrogen storage system, and They represent the upper and lower boundaries of the planned capacity of the hydrogen storage system respectively;

[0071] The battery constraints are:

[0072]

[0073] Where N ba is the planned capacity of the battery, and Respectively represent the upper and lower boundaries of the planned battery capacity;

[0074] The supply and demand balance constraint of electric energy is:

[0075] G ele (Δt) = N pv ·θ pv,s,o ·Δt+N fc ·θ fc(ele-generation),s,o·Δt+N ba ·θ ba,discharge,s,o ·Δt (27)

[0076] D ele (Δt) = E ele (Δt)+N ele ·θ ele,s,o ·Δt+N ba ·θ ba,charge,s,o ·Δt (28)

[0077] G ele (Δt)≥D ele (Δt) (29)

[0078] In the formula, G ele (Δt) represents the amount of electric energy supplied within the time interval Δt, D ele (Δt) represents the electric energy demand within the time interval Δt, θ fc(ele-generatioin),s,o represents the power generation efficiency of the fuel cell in state o in season s, θ ba,discharge,s,o and θ ba,charge,s,o They represent the discharge and charge efficiency of the battery in state o in season s, E ele (Δt) is the external power demand within the time interval Δt;

[0079] The supply and demand balance constraint of hydrogen energy is:

[0080] G H2 (Δt) = N ele ·θ ele,s,o· Δt (30)

[0081] D H2 (Δt) = E H2 (Δt)+N fc ·θ fc(H2-consumptiion),s,o ·Δt (31)

[0082] G H2 (Δt)≥D H2 (Δt) (32)

[0083] In the formula, G H2 (Δt) represents the amount of hydrogen energy supplied within the time interval Δt, D H2 (Δt) represents the hydrogen energy demand within the time interval Δt, E H2 (Δt) is the external hydrogen demand within the time interval Δt, θ fc(H2-consumption),s,o represents the hydrogen consumption efficiency of the fuel cell in state o in season s.

[0084] Compared with the prior art, the present invention has the following beneficial effects:

[0085] 1. The present invention comprehensively considers photovoltaic power generation and water electrolysis hydrogen production, including the carbon emission rights trading mechanism, and organically combines energy supply (photovoltaic power generation group), energy conversion (electrolyzer, fuel cell), energy storage (battery, hydrogen storage system) and energy trading (electricity market, hydrogen market, carbon emission rights trading market), and constructs a hydrogen-electricity storage system covering the entire industry chain. The electric energy generated by the photovoltaic power generation group first meets the internal electricity demand of the system (i.e., powering the electrolyzer), and the surplus electric energy is stored in the battery on the basis of meeting the external electricity demand (electricity market). The hydrogen generated by the electrolyzer first meets the external hydrogen demand (hydrogen market), and the surplus hydrogen is stored in the hydrogen storage system to achieve seasonal hydrogen storage. When the electric energy generated by the photovoltaic power generation group cannot meet the electricity demand of the electrolyzer and the external electricity demand, the external hydrogen demand is met by the hydrogen stored in the hydrogen storage system, and the external electricity demand is met by the fuel cell consuming the hydrogen stored in the hydrogen storage system to generate electric energy, which fully adapts to the seasonal fluctuations of photovoltaic power generation and the dynamic changes in external electricity and hydrogen demand.

[0086] 2. In order to more accurately characterize the seasonal fluctuations and output uncertainty of photovoltaic power generation, the photovoltaic power generation data for the whole year is divided by season, and the photovoltaic power generation data for each season is clustered into three states: high, medium, and low, so that the photovoltaic power generation state can be more accurately divided in the future, thereby providing a more reliable data basis for capacity configuration optimization and improving the accuracy of capacity configuration. In view of the uncertainty and nonlinear relationship of photovoltaic output, the traditional fuzzy C-means clustering algorithm requires uniform distribution of data when using Euclidean distance as a metric, which cannot accurately describe the complexity of photovoltaic power generation data. Therefore, the fuzzy C-means clustering algorithm based on Wasserstein distance is used to cluster photovoltaic power generation data, which can capture the complexity of photovoltaic power generation data distribution, especially uncertainty fluctuations and nonlinear relationships, and improve clustering accuracy.

[0087] Compared with the static description of state probability in the prior art, the present invention combines the state probability of photovoltaic power generation in different seasons and the state transition law between seasons to establish a state transfer matrix, and quantifies the dynamic distribution law of different photovoltaic power generation states throughout the year. This dynamic modeling method of photovoltaic power generation state is more in line with the actual situation, and can reflect the dynamic changes of the system in the operation throughout the year, providing strong support for the subsequent formulation of more scientific operation optimization strategies.

[0088] 3. The benefits of carbon emission rights trading are taken into account when constructing the objective function, so that the system can obtain additional benefits through carbon emission reduction during operation, further improve economic benefits, and take into account environmental benefits. While promoting the deep integration of photovoltaic power generation and hydrogen energy, a win-win situation of economic and environmental benefits is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0089] Figure 1 It is a system framework diagram of the present invention;

[0090] Figure 2 It is the overall flow chart of the present invention; DETAILED DESCRIPTION

[0091] Specific embodiments are given below in conjunction with the accompanying drawings. The specific embodiments are only used to introduce the technical solutions of the present invention in detail and are not intended to limit the protection scope of the present application.

[0092] The hydrogen electricity storage system involved in the present invention includes four modules: energy supply, energy conversion, energy storage and energy trading. The energy supply module is mainly composed of a photovoltaic generator set, which is used to convert solar energy into electrical energy; the energy conversion module includes an electrolyzer and a fuel cell. The electrolyzer converts the electrical energy generated by the photovoltaic generator set into hydrogen energy through a chemical reaction to meet the needs of the hydrogen sales market. At the same time, a certain scale of hydrogen energy storage capacity across seasons is planned to prevent the photovoltaic power station from being unable to operate for a long time at certain special moments. The fuel cell converts the stored hydrogen into electrical energy to supply external market demand when there is a power shortage; the energy storage module is mainly composed of a battery unit and a hydrogen storage system. The battery unit is used for short-term energy storage, and the stored energy comes from the photovoltaic power generation during the day. The battery only supplies the electrolyzer at night. This is done to avoid the electrolyzer from being frequently started and stopped to shorten its life, and to provide sufficient hydrogen production capacity. The battery is not associated with the energy trading module. The hydrogen storage system consists of several spherical high-pressure gaseous hydrogen storage tanks. The hydrogen energy of the hydrogen storage system is used to supply the hydrogen sales market, and the other part is used for seasonal hydrogen storage. Seasonal hydrogen storage can be used to supply fuel cell power generation to supply electricity demand during the system's dry period and at the same time be delivered to the hydrogen sales market to supply hydrogen energy demand; the energy trading module consists of two parts: the electricity sales market and the hydrogen sales market. Energy companies sign annual electricity sales contracts and hydrogen sales contracts with external markets. The system should continuously supply energy within a certain period of time to fulfill the contract. At the same time, as a new energy power generation and green hydrogen enterprise, the hydrogen-electricity storage coupling system can obtain carbon emission rights indicators allocated by the government, and the energy system theme operator can sell carbon emission rights quotas in the secondary carbon trading market according to annual production capacity to obtain additional income. The battery is mainly used in the following two situations: when the photovoltaic generator set operates well during the day, the battery stores excess electricity to supply the electrolyzer to produce hydrogen at night; at night, the battery provides electricity to the electrolyzer, but does not directly participate in the power supply of the hydrogen storage system. The battery preferentially supports the external needs of the system through hydrogen storage conversion. In the design, in order to avoid the impact of frequent start and stop of the electrolyzer unit on the equipment, the system limits the battery to complete the charge and discharge cycle within each scheduling cycle to ensure the stability of the equipment operation.

[0093] The present invention provides a method for configuring the capacity of an off-grid hydrogen-electric storage system taking into account seasonal hydrogen storage, comprising the following steps:

[0094] Step 1: Obtain the photovoltaic power generation data of historical years and divide them according to seasons; Use Latin Hypercube Sampling (LHS) to sample the photovoltaic power generation data of each season, generate new photovoltaic power generation data, achieve data enhancement, and solve the data shortage problem caused by the uncertainty of photovoltaic power generation; Use the fuzzy C-means clustering (FCM) algorithm based on Wasserstein distance to cluster the photovoltaic power generation data of each season, cluster photovoltaic power generation into three states: high, medium, and low, recorded as states 1, 2, and 3, and count the proportion of each state in each season as the initial probability of various states in each season;

[0095] The system operation mode is different under different photovoltaic power generation states. State 1 represents sufficient photovoltaic power generation. Photovoltaic power generation first meets the full-load operation of the electrolyzer for hydrogen production during the day, produces enough hydrogen, and has a surplus on the basis of meeting the external hydrogen demand. The surplus hydrogen is stored in the hydrogen storage system, and the remaining photovoltaic power generation is stored in the battery on the basis of meeting the external electricity demand, which is used for the electrolyzer to produce hydrogen at night; State 2 represents average photovoltaic power generation. Photovoltaic power generation first meets the electrolyzer to produce hydrogen at the minimum rated power during the day, and the produced hydrogen can just meet the external hydrogen demand, but there is no surplus hydrogen stored in the hydrogen storage system, and the remaining photovoltaic power generation can just meet the external electricity demand; State 3 represents low photovoltaic power generation. Photovoltaic power generation cannot meet the electricity demand of the electrolyzer and the external electricity demand. The external hydrogen demand is met by the hydrogen stored in the hydrogen storage system, and the external electricity demand is met by the fuel cell consuming the hydrogen stored in the hydrogen storage system to generate electricity.

[0096] Since the photovoltaic power generation state of the current day is only related to the photovoltaic power generation state of the previous day, that is, the probability distribution of the next state depends only on the current state and is not affected by the previous state, which conforms to the "no memory" of the Markov process. Therefore, the state transition probability matrix of photovoltaic power generation can be obtained based on the historical photovoltaic power generation data to describe the transition rules between different states. The uncertainty of photovoltaic output can be more accurately described by dynamically adjusting the weights of different states in different seasons using the state transition probability matrix and the initial probability distribution. The photovoltaic power generation state space is constructed based on the Markov model to obtain the state transition probability matrix of each season. Then the state transition probability matrix P of season s is (s) It is expressed as:

[0097]

[0098] In the formula, represents the probability of photovoltaic power generation transferring from state i to state o in season s, s = 1, 2, 3, 4 represent spring, summer, autumn, and winter respectively;

[0099] The state probability vector is used to describe the probability of photovoltaic power generation being in various states on each day of the season. The state probability vector of season s is It is expressed as:

[0100]

[0101] In the formula, represents the probability that the photovoltaic power generation is in state o on the tth day in season s and satisfies Representing high, medium and low states respectively;

[0102] Using the state transition probability matrix of season s, we can get the state probability vector of season s The update formula is:

[0103]

[0104] In the formula, represents the state probability vector of day t in season s;

[0105] The state probability vector according to season s Calculate the weights of photovoltaic power generation in various states on each day in season s; among them, the weight λ of photovoltaic power generation in state o on the tth day in season s is s,o (t) is calculated by the following formula:

[0106]

[0107] In the formula, λ s,o (0) is the initial weight of state o in season s, that is, the initial probability of state o in season s;

[0108] The weighted average of the weights of photovoltaic power generation in the same state on each day of the season is obtained to obtain the average weight of each state in the season; among which, the average weight of state o in season s is Calculated by the following formula:

[0109]

[0110] Where, T s Indicates the number of days included in season s.

[0111] By introducing the state probability Combining historical photovoltaic power generation information with the state probability vector, the weight of the photovoltaic power generation state is adaptively corrected, thereby more accurately characterizing the changes in the photovoltaic power generation state in different seasons, and dynamically reflecting the impact of the changes in the photovoltaic power generation state on the weight adjustment; compared with static weights, the dynamic adjustment method of weights is more flexible and robust, and can better adapt to complex seasonal influences and state changes, providing a basis for subsequent optimization goals that consider the seasonal characteristics of photovoltaic power generation.

[0112] Step 2: Construct the objective function and constraints for the capacity configuration of the hydrogen-electricity storage system; the objective function includes the planning and design cost function, the operation and maintenance cost function, and the net present value benefit function;

[0113] (1) The planning and design cost of the hydrogen-electricity-storage coupling system is the first optimization goal, which aims to configure the optimal capacity for each device and achieve the economic efficiency of the planning and design cost while meeting the demand for electricity and hydrogen. The planning and design cost function is:

[0114] minF1=C inv +C rep (6)

[0115] Where F1 is the planning and design cost of the system, C inv is the initial construction cost of the system, C rep It is the cost of updating the system after it is put into use;

[0116] The initial construction cost of the system C inv It is the sum of the initial construction costs of each equipment and is calculated as follows:

[0117] G inv =∑ m N m β m ;m∈{pv,ele,fc,h,ba} (7)

[0118] Where N m is the planned capacity of device m, β m is the construction cost per unit capacity of equipment m; m is pv for photovoltaic generator, m is ele for electrolyzer, m is fc for fuel cell, m is h for hydrogen storage system, and m is ba for battery;

[0119] Update cost after the system is put into use C rep It is the cost incurred due to equipment degradation and damage during the system's use, which requires timely scrapping and replacement of equipment. It is the sum of the renewal costs of each device and is calculated by the following formula:

[0120] C rep =∑ m N m η m ;m∈{pv,ele,fc,h,ba} (8)

[0121] Where η m It is the update cost of the unit capacity of equipment m, which is related to the number of times the equipment needs to be updated and the unit price of investment cost. The calculation formula is as follows:

[0122]

[0123] Where, L plan Indicates the design service life of the system, which is set to 30 years in this embodiment; L m It indicates the expected service life of the equipment m. After the expected service life is exceeded, it needs to be scrapped and replaced with a new equipment.

[0124] (2) The operation and maintenance cost of the hydrogen-electricity-storage coupling system is the second optimization goal, which aims to design the output dispatch of each device and achieve operational economy under the premise of meeting the balance between electricity and hydrogen supply and demand. The operation cost function is:

[0125] minF2=∑ m C op,m ; m∈{pv,ele,fc,h,ba} (10)

[0126] In the formula, C op,m is the operation and maintenance cost of device m;

[0127] The operation and maintenance cost of the equipment mainly comes from the daily operation and maintenance cost, such as equipment maintenance and repair costs, component cleaning costs, and unit start-up and shutdown costs. The operation and maintenance cost C of equipment m is op,m Calculated by the following formula:

[0128]

[0129] O m,s,o =N m ·θ m,s,o ; m∈{pv,ele,fc,h,ba} (12)

[0130] Where, T y Indicates the number of days in a year, O m,s,o represents the daily production capacity of equipment m in state o in season s, φ m represents the operation and maintenance cost per unit output of equipment m, θ m,s,o is the efficiency of equipment m in state o in season s.

[0131] (3) The net present value (NPV) of the hydrogen-electricity storage coupling system is the third optimization objective, which aims to measure the economic efficiency of the entire system operation. The net present value function is:

[0132]

[0133] In the formula, R l is the total system revenue in year l, C op =∑ m C op,m is the system operation and maintenance cost; r is the discount rate, and the empirical value is 0.08;

[0134] The total system revenue includes hydrogen sales revenue, electricity sales revenue and carbon emission rights trading revenue. The total system revenue in year l is R l Calculated by the following formula:

[0135] R l =R H2,l +R elec,l +R carbon,l (14)

[0136] In the formula, R H2,l , R elec,l and R carbon,l They are the hydrogen sales revenue, electricity sales revenue and carbon emission rights trading revenue in the first year respectively;

[0137] The income from hydrogen sales in the first year is R H2,l Calculated by the following formula:

[0138]

[0139] In the formula, π H2,l is the hydrogen sales coefficient in the first year, reflecting the proportion of hydrogen sales, and its value range is (0,1); H2,s,o is the hydrogen production of the electrolyzer in state o in season s, V H2,l is the hydrogen selling price in year l;

[0140] The electricity sales revenue in the first year is R elec,l Calculated by the following formula:

[0141]

[0142] In the formula, π elec,l is the electricity sales coefficient in the first year, reflecting the proportion of electricity sales, and its value range is (0,1); pv,s,o is the photovoltaic power generation in state o in season s, V elec,l represents the electricity selling price in year l;

[0143] The carbon emission trading income R in the first year carbon,l Calculated by the following formula:

[0144]

[0145] In the formula, π carbon,l It represents the electricity-carbon conversion coefficient, reflecting the carbon reduction corresponding to the unit photovoltaic power generation. carbon,l Represents the carbon emission rights trading price in the lth year.

[0146] Constraints include photovoltaic generators, electrolyzers, fuel cells, hydrogen storage systems, battery constraints, and supply and demand balance constraints for electricity and hydrogen;

[0147] PV generator constraints include:

[0148]

[0149] 0≤O pv,s,o ≤N pv (19)

[0150] O pv,s,o ≤OPV actual (20)

[0151] Where N pv is the planned capacity of the photovoltaic generator set, and Respectively represent the upper and lower boundaries of the planned capacity of the photovoltaic power generation unit, OPV actual It indicates the available photovoltaic power generation obtained by the system based on historical data combined with real-time monitoring;

[0152] Electrolyser constraints include:

[0153]

[0154] Where N ele is the planned capacity of the electrolyzer, and Respectively represent the upper and lower boundaries of the planned capacity of the electrolyzer, θ ele,s,o is the efficiency of the electrolyzer in state o in season s, and They represent the upper and lower bounds of the efficiency of the electrolyzer in state o in season s, respectively;

[0155] Fuel cell constraints include:

[0156]

[0157] Where N fc is the planned capacity of the fuel cell, and Respectively represent the upper and lower boundaries of the fuel cell planning capacity, θ fc,s,o is the efficiency of the fuel cell in state o in season s, and They represent the upper and lower bounds of the efficiency of the fuel cell in state o in season s, respectively;

[0158] The constraints of the hydrogen storage system are:

[0159]

[0160] Where N h is the planned capacity of the hydrogen storage system, and They represent the upper and lower boundaries of the planned capacity of the hydrogen storage system respectively;

[0161] The battery constraints are:

[0162]

[0163] Where N ba is the planned capacity of the battery, and Respectively represent the upper and lower boundaries of the planned battery capacity;

[0164] The power supply and demand balance constraint means that within any time interval, the power supply of the system is greater than or equal to the power demand, ensuring that the system can meet the power demand at any time and avoid insufficient power supply. The power supply and demand balance constraint is:

[0165] G ele (Δt) = N pv ·θ pv,s,o ·Δt+N fc ·θ fc(ele-generation),s,o ·Δt+N ba ·θ ba,discharge,s,o ·Δt (27)

[0166] D ele (Δt) = E ele (Δt)+N ele ·θ ele,s,o ·Δt+N ba ·θ ba,charge,s,o ·Δt (28)

[0167] G ele (Δt)≥D ele (Δt) (29)

[0168] In the formula, G ele (Δt) represents the amount of electric energy supplied within the time interval Δt, including the amount of photovoltaic power generation, the amount of power generated by the fuel cell during hydrogen-to-electricity conversion, and the amount of discharge of the battery within the time interval Δt; D ele (Δt) represents the power demand within the time interval Δt, including the external power demand E within the time interval Δt ele (Δt), the amount of electricity consumed by the electrolyzer to produce hydrogen and the amount of electricity stored in the battery; θ fc(ele-generation),s,o represents the power generation efficiency of the fuel cell in state o in season s, θ ba,discharge,s,o and θ ba,charge,s,o They represent the discharge and charge efficiencies of the battery in state o in season s, respectively;

[0169] The supply and demand balance constraint of hydrogen energy means that within any time interval, the system hydrogen energy supply is greater than or equal to the hydrogen energy demand, ensuring that the system can meet the hydrogen demand at any time and avoid the situation of insufficient hydrogen energy supply. The supply and demand balance constraint of hydrogen energy is:

[0170] G H2 (Δt) = N ele ·θ ele,s,o ·Δt (30)

[0171] D H2 (Δt) = E H2 (Δt)+N fc ·θ fc(H2-consumption),s,o ·Δt (31)

[0172] G H2 (Δt)≥D H2 (Δt) (32)

[0173] In the formula, G H2 (Δt) represents the amount of hydrogen energy supplied within the time interval Δt, that is, the amount of hydrogen produced by the electrolyzer within the time interval Δt; D H2 (Δt) represents the hydrogen energy demand within the time interval Δt, including the external hydrogen demand E within the time interval Δt H2 (Δt) and the amount of hydrogen consumed by the fuel cell when converting hydrogen into electricity; θ fc(H2-consumption),s,o represents the hydrogen consumption efficiency of the fuel cell in state o in season s.

[0174] Step 3: Taking into account the constraints, solve the objective function of the hydrogen storage system capacity configuration to obtain the Pareto front solution set, obtain the optimal solution from the Pareto front solution set, and then obtain the optimal planning capacity of each device;

[0175] The multiple objective functions, constraints and decision variables (planned capacity of each device) of the hydrogen storage system lead to extremely high dimensionality of the optimization problem. High-dimensional optimization problems may contain multiple local optimal solutions. Traditional methods are difficult to effectively find all optimal solutions. Therefore, Gaussian Process Regression (GPR) is selected as a proxy model to solve the optimization objective function. On the one hand, the optimal planned capacity of each device is obtained under the multiple objectives of system planning and design cost, operation and maintenance cost and net present value benefit. On the other hand, the planning and design cost, operation and maintenance cost and net present value benefit of the system under the optimal planned capacity are obtained. The Pareto front solution set consists of multiple solutions without priority order. Each solution is better than other solutions in some objectives, but cannot achieve the optimal in other objectives. Therefore, it is necessary to comprehensively consider multiple factors to select the optimal solution, and then obtain the optimal planned capacity of each device.

[0176] Any matters not described in the present invention are applicable to the prior art.

Claims

1. A method for configuring the capacity of an off-grid hydrogen storage system considering seasonal hydrogen storage, characterized in that: The method comprises the following steps: Step 1: Obtain the photovoltaic power generation data of historical years, divide it by season and perform data enhancement; cluster the photovoltaic power generation data of each season, divide photovoltaic power generation into three states: high, medium and low, and count the proportion of each state in each season as the initial probability of each state in each season; construct the photovoltaic power generation state space based on the Markov model, and obtain the state transition probability matrix of each season, where the state transition probability matrix P of season s is (s) It is expressed as: In the formula, represents the probability of photovoltaic power generation transferring from state i to state o in season s, s = 1, 2, 3, 4 represent spring, summer, autumn, and winter respectively; The state probability vector is used to describe the probability of photovoltaic power generation being in various states on each day of the season. The state probability vector of season s is It is expressed as: In the formula, represents the probability that the photovoltaic power generation is in state o on the tth day in season s, Representing high, medium and low states respectively; According to the state transition probability matrix of season s, the state probability vector of season s is obtained The update formula is: In the formula, represents the state probability vector of day t in season s; According to the state probability vector of season s, the weights of photovoltaic power generation in various states on each day in season s are calculated, and the weight λ of photovoltaic power generation in state o on the tth day in season s is s,o (t) is calculated by the following formula: In the formula, λ s,o (0) is the initial probability of state o in season s; The weighted average of the weights of photovoltaic power generation in the same state on each day of the season is obtained to obtain the average weight of each state in the season. The average weight of state o in season s is Calculated by the following formula: Where, T s Indicates the number of days included in season s; Step 2: Construct the objective function of hydrogen storage system capacity configuration, including planning and design cost function, operation and maintenance cost function and net present value benefit function; The planning and design cost function is: minF1=C inv +C rep (6) Where F1 is the planning and design cost of the system, C inv is the initial construction cost of the system, C rep It is the cost of updating the system after it is put into use; The operating cost function is: minF2=C op =∑ m C op,m ;m∈{pv,ele,fc,h,ba} (10) In the formula, C op,m is the operation and maintenance cost of equipment m, where pv,ele,fc,h,ba represent the photovoltaic generator set, electrolyzer, fuel cell, hydrogen storage system and battery respectively; The operation and maintenance cost C of device m op,m Calculated by the following formula: YOU ARE m,s,o N m ·θ m,s,o m∈{pv,ele,fc,h,ba} (12) Where, L plan represents the design service life of the system, T y Indicates the number of days in a year, O m,s,o represents the daily production capacity of equipment m in state o in season s, φ m Represents the operation and maintenance cost of equipment m per unit output, N m is the planned capacity of device m, θ m,s,o is the efficiency of equipment m in state o in season s; The net present value benefit function is: In the formula, R l is the total system revenue in year l, r is the discount rate, C op It is the system operation and maintenance cost; Total system revenue R in year l l Calculated by the following formula: R l =R H2,l +R elec,l +R carbon,l (14) In the formula, R H2,l , R elec,l and R carbon,l They are the hydrogen sales revenue, electricity sales revenue and carbon emission rights trading revenue in the first year respectively; The income from hydrogen sales in the first year is R H2,l Calculated by the following formula: In the formula, π H2,l is the hydrogen sales coefficient in the first year, and its value range is (0,1); H2,s,o is the hydrogen production of the electrolyzer in state o in season s, V H2,l is the hydrogen selling price in year l; The electricity sales revenue in the first year is R elec,l Calculated by the following formula: In the formula, π elec,l is the electricity sales coefficient in the first year, and its value range is (0,1); pv,s,o is the photovoltaic power generation in state o in season s, V elec,l represents the electricity selling price in year l; The carbon emission trading income R in the first year carbon,l Calculated by the following formula: In the formula, π carbon,l Represents the electricity-to-carbon conversion coefficient, V carbon,l represents the carbon emission rights trading price in year l; Step 3: Taking into account the constraints, solve the objective function of the capacity configuration of the hydrogen storage system to obtain the Pareto frontier solution set, obtain the optimal solution from the Pareto frontier solution set, and then obtain the optimal planned capacity of each equipment.

2. The off-grid hydrogen storage system capacity configuration method considering seasonal hydrogen storage according to claim 1, characterized in that: The initial construction cost of the system is C inv Calculated by the following formula: C inv (∑) m N m β m m∈{pv,ele,fc,h,ba} (7) In the formula, β m is the construction cost per unit capacity of equipment m; Update cost after the system is put into use C rep Calculated by the following formula: C rep (∑) m N m η m m∈{pv,ele,fc,h,ba} (8) Where η m is the update cost per unit capacity of device m, L m Represents the expected service life of device m.

3. The off-grid hydrogen storage system capacity configuration method considering seasonal hydrogen storage according to claim 1 or 2, characterized in that: The constraints include photovoltaic power generation units, electrolyzers, fuel cells, hydrogen storage systems, battery constraints, and supply and demand balance constraints for electricity and hydrogen energy; PV generator constraints include: 0≤O pv,s,o ≤N pv (19) THE pv,s,o ≤OPV actual (20) Where N pv is the planned capacity of the photovoltaic generator set, and Respectively represent the upper and lower boundaries of the planned capacity of the photovoltaic power generation unit, OPV actual It indicates the available photovoltaic power generation; Electrolyser constraints include: Where N ele is the planned capacity of the electrolyzer, and Respectively represent the upper and lower boundaries of the planned capacity of the electrolyzer, θ ele,s,o is the efficiency of the electrolyzer in state o in season s, and They represent the upper and lower bounds of the efficiency of the electrolyzer in state o in season s, respectively; Fuel cell constraints include: Where N fc is the planned capacity of the fuel cell, and Respectively represent the upper and lower boundaries of the planned capacity of the electrolyzer, θ fc,s,o is the efficiency of the fuel cell in state o in season s, and They represent the upper and lower bounds of the efficiency of the fuel cell in state o in season s, respectively; The constraints of the hydrogen storage system are: Where N h is the planned capacity of the hydrogen storage system, and They represent the upper and lower boundaries of the planned capacity of the hydrogen storage system respectively; The battery constraints are: Where N ba is the planned capacity of the battery, and Respectively represent the upper and lower boundaries of the planned battery capacity; The supply and demand balance constraint of electric energy is: G ele (Δt)=N pv ·i pv,s,o ·Δt+N fc ·i fc(ele-generation),s,o ·Δt+N ba ·i ba,discharge,s,o ·Δt (27) D ele (Δt)=E ele (Δt)+N ele ·i ele,s,o ·Δt+N ba ·i ba,charge,s,o ·Δt (28) G ele (Δt)≥D ele (Δt) (29) In the formula, G ele (Δt) represents the amount of electric energy supplied within the time interval Δt, D ele (Δt) represents the electric energy demand within the time interval Δt, θ fc(ele-generation),s,o represents the power generation efficiency of the fuel cell in state o in season s, θ ba,discharge,s,o and θ ba,charge,s,o They represent the discharge and charge efficiency of the battery in state o in season s, E ele (Δt) is the external power demand within the time interval Δt; The supply and demand balance constraint of hydrogen energy is: G H2 (Δt)=N ele ·i ele,s,o ·Δt (30) D H2 (Δt)=E H2 (Δt)+N fc ·i fc(H2-consumption),s,o ·Δt (31) G H2 (Δt)≥D H2 (Δt) (32) In the formula, G H2 (Δt) represents the amount of hydrogen energy supplied within the time interval Δt, D H2 (Δt) represents the hydrogen energy demand within the time interval Δt, E H2 (Δt) is the external hydrogen demand within the time interval Δt, θ fc(H2-consumption),s,o represents the hydrogen consumption efficiency of the fuel cell in state o in season s.

Images