Hydrogen energy park planning method and system based on heuristic algorithm

Abstract

The application discloses a hydrogen energy park planning method and system based on a heuristic algorithm, relates to the technical field of intelligent energy management, and comprises the following steps: a mathematical model of a park comprehensive energy system is constructed, wherein the system is composed of a renewable energy power generation unit, an energy conversion device, a carbon capture device, a gas power device and a plurality of types of energy storage units; based on the mathematical model, a cost objective function is established with the lowest annual comprehensive cost of the park comprehensive energy system as an objective, constraint conditions are set, an energy management strategy is formulated; and according to the energy management strategy, the improved Tianji Horse Race Optimization (THRO) algorithm is used for solving, and a hydrogen energy park is planned according to the solving result. The application realizes internalization of environmental externality and maximization of economic benefits.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of smart energy management, in particular to a hydrogen energy park planning method and system based on a heuristic algorithm. BACKGROUND

[0002] At present, hydrogen energy is mainly used as a chemical raw material in the industrial field for the production of chemical products such as ammonia and methanol. The related technology is relatively mature and can achieve stable large-scale supply. In the transportation field, fuel cell vehicle hydrogenation technology is gradually popularized, and the construction and operation of hydrogenation stations have formed certain norms to meet the hydrogenation needs of vehicles. In terms of hydrogen production technology, alkaline electrolytic cells, proton exchange membrane electrolytic cells and other hydrogen production equipment have been put into practical application, and can use new energy power to produce hydrogen, and the hydrogen production efficiency is continuously improved. At the same time, the hydrogen storage technology also has certain development, and the high-pressure gaseous hydrogen storage, low-temperature liquid hydrogen storage and solid hydrogen storage technologies are applied in different scenes, providing support for the transportation and use of hydrogen energy.

[0003] In terms of hydrogen energy grid-connected power generation, the related technology is also continuously developing. In the fuel cell grid-connected power generation technology, proton exchange membrane fuel cells and solid oxide fuel cells have certain grid-connected capability and can convert hydrogen energy into electric energy and input into the power grid. Hydrogen gas turbine grid-connected power generation technology can use the mixed combustion of hydrogen energy and natural gas to generate electricity, realizing the application of hydrogen energy in traditional power generation equipment. In order to ensure the stability of grid connection, the existing power grid dispatching system will regulate and control hydrogen energy power generation to some extent, monitor the output power of hydrogen energy power generation, adjust the output of other power sources, and maintain the frequency and voltage stability of the power grid. In addition, the metering and billing technology of hydrogen energy power generation is also gradually improved, which can realize accurate metering and reasonable billing of hydrogen energy power generation.

[0004] When solving the capacity configuration problem of comprehensive energy systems, traditional heuristic algorithms have a wide range of applications. Genetic algorithm optimizes the configuration of device capacity in the system by simulating the genetic, mutation and selection operations in the biological evolution process, and can select the better configuration scheme from multiple candidate schemes. Particle swarm optimization algorithm simulates the foraging behavior of bird swarm, and finds the optimal capacity configuration solution through information sharing and cooperation between particles, which is efficient in handling small and medium-sized capacity configuration problems. Simulated annealing algorithm simulates the physical annealing process by controlling the reduction of temperature, so that the algorithm has the opportunity to jump out of the local optimal solution, which improves the optimization effect of capacity configuration to a certain extent. These algorithms provide technical support for the reasonable configuration of device capacity in the planning and design of comprehensive energy systems, helping to optimize the system cost, efficiency and other targets.

[0005] There are obvious deficiencies in the existing technology of hydrogen energy utilization in integrated energy systems. In the industrial field, hydrogen energy is only used as a chemical raw material, and it is not effectively combined with power generation, heat supply and other links, resulting in that the energy value of hydrogen energy is not fully utilized and a large amount of energy potential is wasted. In the transportation field, the hydrogen refueling technology of fuel cell vehicles lacks coordination and interaction with other energy networks such as power grids and heat networks, and cannot realize the cascade utilization and optimal allocation of energy. In the process of hydrogen production, a large amount of waste heat is directly discharged in the existing technology, without being combined with the heating and industrial heat demand of the park, resulting in a huge waste of energy. In the field of hydrogen storage technology, high-pressure gaseous hydrogen storage has the problems of high energy consumption and safety to be improved; low-temperature liquid hydrogen storage has high cost and large evaporation loss; and solid-state hydrogen storage technology faces challenges such as low hydrogen storage capacity and high desorption temperature, which limits the large-scale application of hydrogen energy.

[0006] The existing technology of hydrogen energy grid-connected power generation also has many shortcomings. When fuel cells are connected to the grid, the output power is easily affected by factors such as hydrogen supply and operating temperature, resulting in large fluctuations and impacting the stability of the grid, and the existing control technology is difficult to quickly and accurately respond to such fluctuations. In the hydrogen gas turbine grid-connected power generation, it is difficult to control the mixing ratio of hydrogen energy and natural gas, and an improper ratio will affect the combustion efficiency and power generation stability, and may increase pollutant emissions. The grid dispatching system lacks flexibility in regulating hydrogen energy generation, and it is difficult to achieve efficient coordination between hydrogen energy generation and new energy generation such as wind power and photovoltaic power. When new energy output fluctuates, hydrogen energy generation cannot adjust the power output in time, affecting the safe and stable operation of the grid. In addition, the cost of hydrogen energy generation is relatively high, and it lacks obvious economic advantages compared with traditional thermal power and hydropower, limiting its large-scale grid-connected application.

[0007] Traditional heuristic algorithms for solving the capacity configuration problem of integrated energy systems have technical shortcomings. Genetic algorithm is prone to slow convergence speed and local optimal solution when dealing with multi-objective and high-dimensional capacity configuration problems, resulting in unsatisfactory optimization results. Particle swarm optimization algorithm has a significant decline in optimization performance when facing complex constraints and large-scale variables, and it is difficult to find the global optimal solution. The parameter setting of the algorithm has a great influence on the optimization result, and improper parameter selection will lead to poor optimization effect. The optimization effect of simulated annealing algorithm is greatly affected by the initial temperature and cooling rate, and it often takes a long time to calculate when dealing with complex capacity configuration problems, and the efficiency is low. At the same time, these traditional heuristic algorithms lack consideration of dynamic factors in the system, and cannot well adapt to dynamic scenarios such as wind and light output fluctuations and hydrogen demand changes, resulting in poor robustness of the configuration scheme and difficulty in meeting the actual operation requirements of integrated energy systems. SUMMARY

[0008] In order to solve the above problems, the purpose of the present application is to provide a hydrogen energy park planning technology based on heuristic algorithm, which aims to break the limitation of single application in the prior art through hydrogen energy multi-element utilization structure, and realize the multi-scene cooperation of hydrogen energy in the energy system.

[0009] In order to achieve the above technical purpose, the present application provides a hydrogen energy park planning method based on heuristic algorithm, comprising the following steps:

[0010] A mathematical model of the park comprehensive energy system is constructed, wherein the system is composed of renewable energy power generation units, energy conversion devices, carbon capture, gas power devices and multiple types of energy storage units;

[0011] Based on the mathematical model, a cost objective function is established with the lowest annual comprehensive cost of the park comprehensive energy system as the target, and constraint conditions are set to develop an energy management strategy;

[0012] According to the energy management strategy, the improved THRO is solved, and the hydrogen energy park is planned according to the solving result, wherein when the improved THRO is obtained, the Latin hypercube sampling is used to replace the random initialization to establish an initial population uniform space distribution model; the cosine annealing strategy is used to reconstruct the linear decay weight to establish a nonlinear time-varying parameter model; the elite reservation mechanism is introduced in the horse racing strategy to construct an elite guided position update operator; a hybrid learning mechanism with time-varying probability control is designed to introduce an optimal solution disturbance operator; and a convergence criterion based on historical solution stability is established.

[0013] Preferably, when the mathematical model of the park comprehensive energy system is constructed, carbon capture modeling is performed by fixing energy consumption operation energy consumption;

[0014] Based on the carbon capture modeling, the mathematical model of the park comprehensive energy system is constructed through electric-gas equipment modeling, hydrogen-doped gas turbine modeling, hydrogen-doped gas boiler modeling, carbon trading model, hydrogen fuel cell modeling, energy storage model and hydrogen power automobile modeling.

[0015] Preferably, when the cost objective function is constructed, the cost objective function is constructed according to the annual comprehensive total cost of the system, the annual investment cost, the annual maintenance and depreciation cost, the carbon trading, the carbon sequestration cost, the natural gas purchase cost, and the abandoned wind / light cost.

[0016] Preferably, when the constraint conditions are set, the installed capacity constraint, the wind / solar power supply constraint, the energy storage device constraint, the power balance constraint, the carbon sequestration capacity limit, the carbon capture constraint, the gas turbine output and climbing constraint, the gas boiler constraint, the fuel cell constraint and the P2H constraint are set as the constraint conditions.

[0017] Preferably, when the energy management strategy is developed, the energy management strategy comprises:

[0018] Mode E1: The wind-solar output has met the electricity load demand, the battery has residual capacity, and the maximum charging power of the battery is greater than the residual power , then the residual power is all stored in the battery.

[0019] Mode E2: The wind-solar output has met the electricity load demand, part of the residual electricity is stored in the battery, and the other part is converted into hydrogen gas by the electrolytic cell.

[0020] Mode E3: The wind-solar output has met the electricity load demand, part of the residual electricity is stored in the battery, the electrolytic cell is operated at the maximum power, water is electrolyzed to convert into hydrogen gas, and the other part is absorbed by carbon capture to absorb carbon dioxide.

[0021] Mode E4: The wind-solar output has met the electricity load demand, the residual power is jointly consumed by electricity-to-gas and carbon capture.

[0022] Mode E5: The wind-solar output cannot meet the electricity load demand, and the system lacks the power is jointly supplied by the battery and the fuel cell, and the gas turbine.

[0023] Mode T1: The fuel cell and gas turbine output have met the heat load demand, the heat storage tank has residual capacity, and the maximum charging power of the heat storage tank is greater than the residual power , then the residual power is all stored in the heat storage tank.

[0024] Mode T2: The fuel cell and gas turbine output have met the heat load demand, the heat storage tank has residual capacity, and the maximum charging power of the heat storage tank is less than the residual power , the heat storage tank is charged at the maximum charging power, and the residual energy is abandoned.

[0025] Mode T3: The fuel cell and gas turbine output have met the heat load demand, the heat storage tank is at the upper limit of the energy state, and the residual energy is abandoned.

[0026] Mode T4: The fuel cell and gas turbine output cannot meet the electricity load demand, and the system lacks the power is jointly supplied by the battery and the fuel cell, and the gas turbine.

[0027] Mode H1: The electrolytic cell output has met the hydrogen load and hydrogen blending demand of gas, the hydrogen storage tank has residual capacity, and the maximum charging power of the hydrogen storage tank is greater than the residual power , then the residual power is all stored in the hydrogen storage tank.

[0028] Mode H2: The electrolyzer has met the hydrogen load and gas hydrogen blending demand, and the hydrogen storage tank has residual capacity, part of the residual hydrogen is stored in the hydrogen storage tank, and the other part is used to start the fuel cell to supplement the power and heat load demand in mode E5 and T4.

[0029] Mode H3: The electrolyzer has met the hydrogen load and gas hydrogen blending demand, and the hydrogen storage tank has residual capacity, part of the residual hydrogen is stored in the hydrogen storage tank, part of the residual hydrogen is used to run the fuel cell at maximum power to supplement the power and heat load demand in mode E5 and T4, and the other part is sent to the methane reactor.

[0030] Mode H4: The electrolyzer has met the hydrogen load and gas hydrogen blending demand, and the hydrogen storage tank has no residual capacity, the residual hydrogen, and the residual power are consumed by the fuel cell and the methane reactor jointly.

[0031] Mode H5: The electrolyzer cannot meet the hydrogen load and gas hydrogen blending demand, and the system lacks power supplied by the hydrogen storage tank.

[0032] Preferably, in the improvement of the Tim Horses Race Optimization algorithm THRO, the elite reservation mechanism is represented as: the elite individuals are reserved in each generation population and introduced into the position update formula to guide the population to move to the high-quality solution area, avoid the loss of high-quality solutions and accelerate the convergence.

[0033] Preferably, after introducing the elite reservation mechanism, the position update is improved through elite guidance.

[0034] Preferably, in the improvement of the Tim Horses Race Optimization algorithm THRO, local disturbance search is performed near the global optimal solution.

[0035] Preferably, in the improvement of the Tim Horses Race Optimization algorithm THRO, an early stop mechanism is set, when it is detected that the standard deviation of the optimal fitness of 50 consecutive generations is less than 10 -6 , the iteration is terminated in advance.

[0036] The application discloses a hydrogen energy park planning system based on a heuristic algorithm, which is used for realizing the hydrogen energy park planning method based on the heuristic algorithm.

[0037] A model construction module is configured to construct a mathematical model of the park comprehensive energy system, wherein the system is composed of a renewable energy power generation unit, an energy conversion device, a carbon capture device, a gas power device, and a plurality of types of energy storage units.

[0038] A strategy making module is configured to establish a cost objective function and set a constraint condition based on the mathematical model, so as to make an energy management strategy with the lowest annual comprehensive cost of the park comprehensive energy system as the target.

[0039] The park planning module is used for planning the hydrogen energy park according to the solving result of the improved Tianji Horse Race Optimization Algorithm THRO according to the energy management strategy, wherein, when the improved Tianji Horse Race Optimization Algorithm THRO is acquired, Latin hypercube sampling is used to replace random initialization to establish an initial population uniform space distribution model; a nonlinear time-varying parameter model is established by using a cosine annealing strategy to reconfigure linear decay weights; an elite guiding position update operator is constructed by introducing an elite reservation mechanism in the horse race strategy; a hybrid learning mechanism with time-varying probability control is designed, and an optimal solution disturbance operator is introduced; and a convergence criterion based on historical solution stability is established.

[0040] The following technical effects are disclosed in the application:

[0041] (1) Hydrogen energy multi-element utilization realizes the cascade development and full value chain application of hydrogen energy. The waste heat generated by hydrogen fuel cells during power generation is not wasted, but is converted into heat energy for the park, allowing the same part of hydrogen to play multiple roles in the energy conversion chain. Compared with single power generation or heating mode, the comprehensive utilization rate of energy is greatly improved, and the hydrogen power generation process only produces water, without carbon dioxide, sulfide and other pollutant emissions; when used in hydrogen refueling stations, hydrogen fuel cell vehicles also achieve zero emissions, reducing the environmental burden of the park from the source. At the same time, compared with traditional fossil energy utilization methods, the dependence on non-renewable resources such as coal and natural gas is effectively reduced, helping the park to build a clean and low-carbon energy consumption structure. The hydrogen multi-element utilization system also builds a flexible and reliable energy network. When the park's electricity demand fluctuates, the hydrogen fuel cell power generation system can quickly adjust the output power, complementing the output of new energy and ensuring stable power supply. The heat supply is combined with gas boilers, gas turbines and hydrogen fuel cells through heat storage equipment, and can be flexibly adjusted according to seasonal and time requirements. The hydrogen refueling station relies on the hydrogen production and storage facilities within the park to avoid the risk of supply interruption caused by external transportation and ensure continuous hydrogen supply. In terms of economic value, the long-term operation advantage is obvious. With the maturity of hydrogen production technology and the emergence of scale effect, the cost of hydrogen production gradually decreases. At the same time, through the cascade utilization of energy, not only the expenditure of purchased electricity and heat is reduced, but also the carbon dioxide emissions are reduced, thereby reducing the environmental cost.

[0042] (2) The energy management strategy is formulated, so that the system ensures stable supply of electric load through multi-energy complementation, reduces the risk of interruption of single energy supply, improves the reliability and anti-interference ability of the system, and can quickly respond to power regulation, avoids equipment wear caused by excessive energy storage, simplifies the regulation process, and ensures safe operation of the system. In terms of energy utilization, the energy utilization approach is expanded, and the system's ability to accept wind and light fluctuations is improved, which not only avoids waste of wind and light energy and reduces energy waste, but also improves energy utilization. In addition, through the peak shaving effect of the energy storage device, the supply and demand fluctuations of energy are balanced. In addition, the system enhances the energy synergy capability, meets the hydrogen reserve while relieving the supply pressure of electric and heat loads, ensures the continuity of heat supply, and meets the stable heat demand of the park production and life, which reduces the dependence on external energy and reduces the operating cost, reduces carbon emissions, conforms to the low-carbon environmental protection concept, improves the environmental protection of energy conversion, realizes the synergy of energy utilization and ecological protection, and enhances the environmental benefits of the system.

[0043] (3) The algorithm replaces random initialization with Latin hypercube sampling to establish an initial population uniform spatial distribution model; a nonlinear time-varying parameter model is established by reconstructing the linear decay weight through the cosine annealing strategy; an elite reservation mechanism is introduced in the horse race strategy to construct an elite-guided position update operator; a hybrid learning mechanism with time-varying probability control is designed, an optimal solution disturbance operator is introduced, and a convergence criterion based on the stability of historical solutions is established. The improved algorithm realizes Pareto improvement in three dimensions of exploration ability, development efficiency and solution quality, and provides a more efficient solution framework for complex optimization problems. BRIEF DESCRIPTION OF DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0045] Figure 1 is a schematic diagram of the park comprehensive energy system structure described in the present application;

[0046] Figure 2 is a schematic diagram of the method flow described in the present application. DETAILED DESCRIPTION

[0047] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative work belong to the scope of protection of the present application.

[0048] As shown in Figures 1-2 , the present application provides a hydrogen energy park planning technology based on heuristic algorithm, specifically including the following contents:

[0049] 1. Constructing a mathematical model of a park comprehensive energy system

[0050] The hydrogen energy multi-element utilization park comprehensive energy system architecture provided by the present application is shown in Figure 1 . The system is mainly composed of renewable energy power generation units (wind turbine, photovoltaic array), energy conversion devices (electricity to gas, hydrogen fuel cell), carbon capture, gas power devices (gas turbine, gas boiler) and multi-type energy storage units (electricity, heat, hydrogen), and adopts an off-grid operation mode of electricity to avoid the risk of power grid impact. In terms of load demand: the electric load is supplied by renewable energy, hydrogen fuel cell, gas turbine and battery; the heat load is supplied by the waste heat recovery and heat storage system of the gas device; and the hydrogen load is satisfied by the cooperation of the electricity-to-hydrogen device and the hydrogen storage tank. The specific structure is shown in Figure 1 .

[0051] (1) Carbon capture modeling:

[0052] The total energy consumption of carbon capture includes fixed energy consumption and operating energy consumption , wherein the operating energy consumption is proportional to the amount of CO2 treated by the regenerator, and the specific expression is as follows:

[0053] (1)

[0054] In the formula, e c is the energy consumption for treating unit mass of CO2, ; represents the mass of CO2 treated by the regenerator of carbon capture at time t; since the energy consumption of carbon capture is very large, the present application adopts CCS with flue gas diversion, and actively discharges CO2 to achieve the purpose of controlling the energy consumption of carbon capture, and the specific expression is as follows:

[0055] (2)

[0056] In the formula: and respectively total emissions of the unit at time t and the amount of CO2 absorbed by CCS at time t; is the amount of CO2 emitted into the atmosphere by flue gas at time t.

[0057] (2) Modeling of electricity-to-gas equipment:

[0058] Considering that the overall conversion efficiency of P2G is low, and in order to fully utilize the low-carbon benefits of hydrogen, the present application subdivides P2G into hydrogen production (P2H) and methane treatment two processes to meet the hydrogen-doped needs of gas boilers and gas turbines. The specific mathematical model is as follows:

[0059] (3)

[0060] In the formula: and are the electric power consumed by electrolysis and the hydrogen production power at time t; and are the hydrogen consumption power and gas production power of the methane reactor at time t; and are the operating efficiencies of P2H and the methane reactor, respectively.

[0061] Since the volume of CO2 required for methanation is equal to the volume of methane generated, the mass of CO2 required for methanation is determined by the following expression:

[0062] (4)

[0063] (5)

[0064] In the formula: and are the amounts of CO2 methanation and sequestration of CCS regeneration at time t, respectively; is the calorific value of natural gas.

[0065] (3) Hydrogen-doped gas turbine:

[0066] The mathematical model of the gas turbine is as follows:

[0067] (6)

[0068] In the formula: and are the electric and thermal efficiencies of the gas turbine; , respectively represent the output electric power and thermal power of the gas turbine at time t; and respectively represent the power corresponding to the natural gas and hydrogen consumed by the gas turbine at time t; is the hydrogen blending ratio (volume ratio) at time t.

[0069] (4) Hydrogen-blended gas boiler:

[0070] The relevant standard requires that when the hydrogen-blended mixed gas is used in the gas boiler, the molar mass proportion of hydrogen is kept within the range of 2% to 20%, and the mathematical model of the gas boiler is as follows:

[0071] (7)

[0072] In the formula: is the thermal conversion efficiency of the gas boiler; is the boiler heat production power at time t; and respectively represent the power corresponding to the hydrogen and natural gas consumed by the gas boiler at time t; and are the relative molecular masses of hydrogen and natural gas; is the hydrogen blending ratio (molar mass) of the gas boiler at time t.

[0073] (5) Carbon trading model:

[0074] The actual carbon emission amount refers to the net emission amount of CO2 generated by the system minus the amount of CO2 treated by the P2G, and the system CO2 net amount includes and two parts, respectively, the CO2 directly discharged into the atmosphere through flue gas diversion and the CO2 indirectly discharged into the atmosphere due to the carbon capture efficiency not reaching 100%, and the calculation formula is as follows:

[0075] (8)

[0076] The total amount of CO2 generated by the system is calculated as follows:

[0077] (9)

[0078] In the formula: is the carbon emission amount corresponding to the natural gas consumed per unit power.

[0079] The initial amount of carbon emission rights is distributed free of charge, and its free share is determined based on the baseline method. The carbon emission rights distribution object of the present application covers gas turbines and coal-fired units.

[0080] (10)

[0081] In the formula: is the free carbon emission right allocated to the system at time t; is the carbon emission right allocation for unit power supply of the gas turbine unit; is the carbon emission right allocation for unit heat supply of the gas turbine unit.

[0082] After the free carbon emission quota and the actual carbon emission of the system are calculated by formulas (8)-(10), the carbon emission right actually participating in the carbon trading market transaction is calculated .

[0083] (11)

[0084] To strengthen the carbon emission constraint, the present application borrows the step price mechanism, and calculates the carbon trading cost according to the carbon emission interval on the basis of unified carbon trading. The emission part exceeding the carbon quota corresponds to a higher price, and the remaining amount is sold when it is lower than the quota, and a compensation coefficient is introduced to enhance the emission reduction incentive. The step carbon trading cost model is defined as follows:

[0085] (12)

[0086] In the formula: is the carbon trading base price; L is the length of the carbon emission interval; is the carbon trading price growth rate; is the compensation coefficient; is the carbon trading cost of the system at time t, which is positive for purchase and negative for sale.

[0087] (6) Hydrogen fuel cell:

[0088] (13)

[0089] In the formula: represents the electric power output by the fuel cell at time t; represents the heat power output by the fuel cell at time t; is the conversion efficiency, which is taken as 0.65 here; represents the hydrogen mass input by the fuel cell at time t; represents the hydrogen power input by the fuel cell at time t.

[0090] (7) Energy storage model:

[0091] Taking a hydrogen storage tank as an example, the mathematical model thereof is established as follows:

[0092] (14)

[0093] In the formula: is the hydrogen storage amount of the hydrogen storage tank at t+1 time; is the hydrogen dissipation rate of the hydrogen storage tank; is the hydrogen storage amount of the hydrogen storage tank at t time; is the hydrogen charging power of the hydrogen storage tank at t time; is the hydrogen discharging power of the hydrogen storage tank at t time; is the hydrogen charging efficiency of the hydrogen storage tank, and the present application takes 0.95;

[0094] energy storage state is the ratio of the hydrogen storage amount at t time to the capacity, that is,

[0095] (15)

[0096] In the formula: is the maximum capacity of the hydrogen storage tank.

[0097] (8) hydrogen power automobile:

[0098] The daily travel distribution of the automobile conforms to the logarithmic normal distribution law.

[0099] (16)

[0100] In the formula: standard deviation The value is 0.88; the expectation The value is 3.20.

[0101] The vehicle end-of-day return time t satisfies the normal distribution, and the probability density function is

[0102] (17)

[0103] The probability distribution characteristics of the daily travel mileage and the end-of-day return time are comprehensively considered, and the recent traffic statistical data is combined, and the Monte Carlo simulation method is used to predict a plurality of key parameters of the new energy automobile in the park, including the battery capacity, the charging starting time, the travel distance and the charging power in each period. Based on the above prediction results, the residual hydrogen amount of the hydrogen energy automobile in different states is further calculated, so that the charging load demand of the hydrogen energy automobile in each different time period is evaluated.

[0104] 2, cost objective function:

[0105] The present application takes the lowest annual comprehensive cost of the park comprehensive energy system as the target, and the target function established is:

[0106] (18)

[0107] In the formula: C all , C inv , C repThe system annual comprehensive total cost, annual investment cost, and annual maintenance and depreciation cost, respectively. , , , The carbon trading, carbon sequestration cost, natural gas purchase cost, and wind / photovoltaic curtailment cost, respectively.

[0108] (1) The system annual investment cost is:

[0109] (19)

[0110] In the formula, r is the discount rate, Y is the service life, and C is the unit capacity investment cost. The total installed capacity and unit capacity investment cost of the hydrogen storage tank, battery, heat storage tank, fuel cell, and P2G, respectively.

[0111] (2) The annual maintenance and depreciation cost is:

[0112] (20)

[0113] In the formula, C is the operation and maintenance cost coefficient, and S is the residual value coefficient. (3) The carbon trading and carbon sequestration cost is:

[0114] (21)

[0115] In the formula, C is the cost of sequestration per unit mass of CO2.

[0116] (4) The annual natural gas purchase cost is:

[0117]

[0118] (22) In the formula, Q is the purchase amount of natural gas at time t, and P is the natural gas price.

[0119] (5) The wind and photovoltaic curtailment penalty cost is:

[0120]

[0121] (23) In the formula, Pw and Pp are the wind and photovoltaic curtailment prices at time t, respectively.

[0122] In the formula, Pw and Pp are the predicted wind and photovoltaic output values, respectively.

[0123] 3. Construction of constraint conditions:

[0124] (1) Installed capacity constraint:​​​​​​

[0125] The installed capacity of the hydrogen storage tank, battery, heat storage tank, fuel cell, and P2G device should not exceed the maximum value to avoid wasting too much capacity and save costs, i.e.,

[0126] (24)

[0127] wherein, Pmax, wind, Pmax, PV, Pmax, tank, Pmax, fuel cell, and Pmax, P2G represent the maximum installed capacity of wind power, photovoltaic power, electrolyzer, hydrogen storage tank, fuel cell, and P2G device, respectively.

[0128] (2) Wind and light power supply constraints:

[0129] Wind power constraints:

[0130] (25)

[0131] Light power constraints:

[0132] (26)

[0133] wherein, , Pwind, t and PV, t represent the amount of abandoned wind and light at time t, respectively. Pwind, t and PV, t represent the predicted output of wind power and photovoltaic power at time t, respectively. Pwind, t and PV, t represent the wind power and photovoltaic power supply at time t, respectively.

[0134] (3) Energy storage device constraints:

[0135] To avoid simultaneous charging and discharging of the energy storage device, a unique charging and discharging state flag should be set. For example, the following constraints are set for the hydrogen storage tank, and the heat storage tank and battery are equivalent, as follows:

[0136] (27)

[0137] To extend the service life of the energy storage device, an SOC constraint should be added, i.e., the state of charge of the hydrogen storage tank at time t should be within the range of , and to avoid simultaneous charging and discharging of the hydrogen storage tank, a charging and discharging state flag should be set, which takes a value of 0 or 1; at the same time, the state of charge of the hydrogen storage tank at time 0 should be consistent with the state of charge at the last time.

[0138] (4) Power balance constraints:

[0139] To ensure the supply and demand balance between the system and the load side, the following constraints are imposed on the four energy flows in the system, i.e.,

[0140] (28)

[0141] where: represents the electric power consumed, outputted by the battery and the electric load demand of the system at time t; represents the thermal power absorbed, outputted by the thermal tank and the thermal load demand of the system at time t; represents the volume of natural gas purchased by the system at time t; represents the volume of natural gas outputted by the methane reactor at time t; represents the volume of natural gas consumed by the gas turbine, gas boiler at time t; represents the hydrogen power absorbed, released by the hydrogen tank and the hydrogen load demand of the system at time t;

[0142] (5) Carbon sequestration capacity constraint

[0143] (29)

[0144] where: respectively represent the sequestration amount of CO2 at time t and the maximum sequestration amount of CO2 of the system;

[0145] (6) Carbon capture constraint

[0146] Operating energy consumption of CCS Cannot exceed the maximum operating power :

[0147] (30)

[0148] (7) Gas turbine output and ramping constraint

[0149] (31)

[0150] where: is the total output of the gas turbine respectively represent the electric power and the thermal power outputted by the gas turbine at time t; is the upper and lower limit of the electric output of the gas turbine; is the upper and lower limit of the thermal output of the gas turbine; is the upper and lower limit of the ramping output of the gas turbine;

[0151] (8) Gas boiler constraint

[0152] (32)

[0153] where, is the upper and lower limit of the thermal output of the gas boiler; is the upper and lower limit of the ramping output of the gas boiler;

[0154] (9) Fuel cell constraint

[0155] (33)

[0156] wherein, Pmax is the upper limit of fuel cell output;

[0157] (10) P2H constraint

[0158] (34)

[0159] P2G operating energy consumption to be within the range ; P2Hmax and P2Hmin represent the upper and lower limits of P2H ramp-up output, respectively;

[0160] 4. Formulate energy management strategy:

[0161] Mode E1: wind-solar output has met the demand for electrical load, and there is remaining capacity in the battery, and the maximum charging power of the battery is greater than the remaining power , then the remaining power is all stored in the battery.

[0162] Mode E2: wind-solar output has met the demand for electrical load, and part of the remaining power is stored in the battery, and the other part is converted into hydrogen gas using an electrolyzer.

[0163] Mode E3: wind-solar output has met the demand for electrical load, and part of the remaining power is stored in the battery, and the electrolyzer is operated at maximum power to convert water into hydrogen gas, and the other part is absorbed by carbon capture.

[0164] Mode E4: wind-solar output has met the demand for electrical load, and the remaining power is jointly consumed by electricity-to-gas and carbon capture.

[0165] Mode E5: wind-solar output cannot meet the demand for electrical load, and the system lacks power is jointly supplied by the battery and fuel cell, and the gas turbine.

[0166] Mode T1: fuel cell, gas turbine output has met the demand for heat load, and there is remaining capacity in the heat storage tank, and the maximum charging power of the heat storage tank is greater than the remaining power , then the remaining power is all stored in the heat storage tank.

[0167] Mode T2: fuel cell, gas turbine output has met the demand for heat load, and there is remaining capacity in the heat storage tank, and the maximum charging power of the heat storage tank is less than the remaining power , the heat storage tank is charged at maximum power, and the remaining energy is treated as curtailment of wind and solar power.

[0168] Mode T3: Fuel cell, gas turbine output has met the heat load demand, the thermal storage tank is in the upper boundary of the energy state, the remaining energy is abandoned.

[0169] Mode T4: Fuel cell, gas turbine output cannot meet the electrical load demand, the system lacks power Supplied by the battery and fuel cell, gas turbine joint.

[0170] Mode H1: Electrolytic tank output has met the hydrogen load and gas hydrogen blending demand, the hydrogen storage tank has remaining capacity, the maximum charging power of the hydrogen storage tank is greater than the remaining power , then the remaining power All stored in the hydrogen storage tank.

[0171] Mode H2: Electrolytic tank has met the hydrogen load and gas hydrogen blending demand, the hydrogen storage tank has remaining capacity, part of the remaining hydrogen is stored in the hydrogen storage tank, and the other part is used to start the fuel cell to supplement the electrical and thermal load demand in modes E5 and T4.

[0172] Mode H3: Electrolytic tank has met the hydrogen load and gas hydrogen blending demand, the hydrogen storage tank has remaining capacity, part of the remaining hydrogen is stored in the hydrogen storage tank, part of it is operated at the maximum power of the fuel cell to supplement the electrical and thermal load demand in modes E5 and T4, and the other part is sent to the methane reactor.

[0173] Mode H4: Electrolytic tank has met the hydrogen load and gas hydrogen blending demand, the hydrogen storage tank has no remaining capacity, the remaining hydrogen, the remaining power Absorbed by the fuel cell and methane reactor jointly.

[0174] Mode H5: Electrolytic tank cannot meet the hydrogen load and gas hydrogen blending demand, the system lacks power Supplied by the hydrogen storage tank.

[0175] 5. Algorithm improvement

[0176] The Tianji Horse Racing Optimization (THRO) algorithm is improved in many ways to improve its global search ability, convergence speed and solution quality. The following is an academic description of the improvements:

[0177] (1) Population initialization optimization: By introducing Latin hypercube sampling (LHS) instead of random initialization, ensure the uniform distribution of the initial population in the solution space, enhance the diversity of the initial solution and the global exploration potential.

[0178] (35)

[0179] In the formula: LHS represents the Latin hypercube design matrix; n is the population size; Dim is the population dimension; Up, Low respectively represent the upper and lower limits of the decision variable;

[0180] (2) Nonlinear adaptive weight: cosine annealing strategy is adopted to adjust the weight factor, and the exploration and exploitation capabilities of the dynamic balance algorithm are dynamically balanced. The weight decays nonlinearly with the number of iterations, which is specifically shown as:

[0181] (36)

[0182] In the formula, P represents the adaptive weight; t1 represents the current iteration number; T represents the maximum iteration number.

[0183] (3) Elite preservation mechanism: elite individuals are preserved in each generation of population and introduced into the position update formula to guide the population to move towards the high-quality solution area, avoid the loss of high-quality solutions and accelerate convergence.

[0184] (37)

[0185] (38)

[0186] In the formula, respectively represent the elite solutions of the chariot and the king populations; is the fitness function; is the individual of the chariot and the king populations

[0187] (4) Position update improvement: introduction of elite guidance

[0188] (39)

[0189] In the formula, Xnew represents the updated new position; e is the elite vector solution; β represents the Levy disturbance coefficient.

[0190] (5) Dynamic learning strategy: design of time-varying probability control to switch exploration and exploitation behavior. The probability of Levy flight exploration decreases linearly with the number of iterations, while the probability of learning from elite individuals increases correspondingly. This strategy makes the algorithm focus on global exploration in the early stage and local exploitation in the later stage.

[0191] (40)

[0192] In the formula, represents the linear decay of Levy exploration probability from 0.7 to 0.1.

[0193] (6) Convergence acceleration mechanism: in the later stage of the algorithm, local disturbance search is performed around the global optimal solution to enhance the local exploitation capability.

[0194] (41)

[0195] In the formula, represents the candidate solution vector; represents the current optimal solution; represents the perturbation amplitude coefficient; represents the Gaussian noise vector.

[0196] (7) Early stopping mechanism: when the standard deviation of the optimal fitness of 50 consecutive generations is detected to be less than 10 -6 , terminate the iteration in advance to reduce unnecessary computational overhead.

[0197] (42)

[0198] wherein, represents the standard deviation function; represents the optimal fitness sequence.

[0199] Through the above improvements, the algorithm significantly improves the global search ability, convergence speed and solution quality on the basis of maintaining the original competitive framework, and especially shows stronger robustness on multi-peak complex optimization problems.

[0200] 6、Model solution:

[0201] According to the characteristics of the seasons, the year is divided into seasons, and a representative day is selected for each of the transition season, summer and winter. March to May and September to November are defined as the transition period, June to August is summer, and December to February is winter. A representative day is selected for each season. The relevant device parameter inputs include the upper and lower limits of the power of the device, the upper and lower limits of the climbing, the operating efficiency, and the predicted wind and light output values, natural gas prices and other parameter information, which are solved by the improved chariot race optimization algorithm. This configuration method can improve the consumption of renewable energy in the industrial park and the reliability of the power grid. Reduce carbon dioxide emissions, improve energy efficiency, reduce dependence on fossil fuels, and reduce fuel procurement expenses, thereby achieving environmental and economic benefits.

[0202] 7、Algorithm evaluation:

[0203] Through the sensitivity analysis of the original chariot race optimization algorithm and the chariot race optimization algorithm improved by the method under the same test function, it can be known that under the improvement mode of the application, the algorithm not only improves the convergence speed and enhances the global search, but also improves the solution quality, enhances the stability, and optimizes the local search ability.

[0204] 8、Model evaluation:

[0205] By using the improved Tianji horse racing optimization algorithm to solve the model, the supply of electricity, heat and hydrogen, as well as the carbon dioxide emission situation can be analyzed. Under the guidance of the energy management strategy formulated above, the park comprehensive energy structure considering hydrogen energy multi-use shows excellent performance. The energy supply is stable and efficient to meet the demand of the park for electricity, heat and hydrogen. The energy structure successfully realizes the efficient complementation between energies. This dynamic energy allocation mechanism keeps the entire energy system in an optimal operating state at all times. And it can realize efficient source-level CO2 capture and storage, thereby significantly reducing carbon emissions and promoting effective recycling of carbon in the park. This combination of technical optimization and economic optimization not only internalizes environmental externalities, but also maximizes economic benefits, thereby providing a sustainable and economically viable strategy for the development of industrial parks.

[0206] In the industrial field, hydrogen energy is not only used as a chemical raw material, but also converted into electricity and heat through fuel cells, hydrogen gas turbines and other equipment, deeply integrated with electricity and heat demand in industrial production, fully realizing the energy value of hydrogen energy and avoiding the waste of energy potential caused by using it only as a raw material in the prior art. In the transportation field, hydrogen energy multi-use realizes the linkage of hydrogen refueling stations and park energy networks. Hydrogen refueling stations not only provide fuel for fuel cell vehicles, but also supply electricity to the park during peak electricity demand through fuel cells, forming a closed-loop system of "hydrogen energy-transportation-electricity", which makes up for the lack of coordination between hydrogen energy and other energy networks in the transportation field in the prior art. Moreover, through the deep coupling of hydrogen energy and multi-energy systems, a "hydrogen energy-electricity-heat-gas" collaborative network is constructed. Hydrogen energy can be converted into electricity and heat through fuel cells to meet the internal electricity and heat load demand, while excess electricity can be stored by electrolysis to form an energy complementary cycle. For example, hydrogen is stored during the peak output of photovoltaic power, and hydrogen is used to generate electricity at night to make up for the lack of photovoltaic power, while the heat generated during electricity generation is used for heating in the park, fully utilizing hydrogen energy as an energy hub to adjust and improve the efficiency of the cascade utilization of the comprehensive energy system. Hydrogen is added to the gas turbine and gas boiler to reduce carbon emissions at the source. At the same time, a methane reactor is introduced to promote the chemical reaction between hydrogen and CO2 collected by the carbon capture system (CCS) to generate methane and other fuels, realizing the recycling and efficient use of carbon-hydrogen resources, and further improving the environmental value and resource utilization rate of hydrogen energy in the energy system.

[0207] Compared to grid-connected integrated energy systems in industrial parks, off-grid integrated energy systems eliminate dependence on the power grid, effectively avoiding a series of problems associated with grid-connected hydrogen power generation. In terms of stability, off-grid systems, through internal energy buffering and regulation mechanisms, can precisely respond to power fluctuations in hydrogen power generation. For example, when the output power of fuel cells fluctuates due to hydrogen supply fluctuations, the energy storage devices within the system (such as lithium batteries and supercapacitors) can quickly respond to mitigate the impact on grid stability under grid-connected conditions, solving the problem of difficulty in quickly and accurately responding to fluctuations in existing grid-connected technologies. In terms of economics, off-grid systems do not incur additional costs such as grid connection fees and standby capacity fees, and also avoid the deep linkage between hydrogen production costs and peak / valley electricity prices during grid connection. In off-grid systems, hydrogen production and consumption are completed in a closed loop within the industrial park, allowing for flexible adjustments to hydrogen production and consumption strategies based on energy supply and demand. This avoids the situation where peak electricity prices during grid connection cause a sharp increase in hydrogen production costs, while also eliminating the investment in transformers, switchgear, and other equipment required for grid connection, thus improving the project's economic feasibility. In terms of low carbon emissions, the off-grid system relies on renewable energy sources such as wind and solar power within the industrial park to produce hydrogen, completely avoiding the problem of increased "grey hydrogen" caused by purchasing thermal power when connected to the grid. The energy flow within the system is fully controllable, ensuring the low carbon attributes of hydrogen production and solving the defect of indirect carbon emissions from the power grid diluting the low carbon emissions of hydrogen in existing grid-connected technologies.

[0208] The improved Tian Ji Horse Racing Optimization Algorithm, drawing inspiration from the strategy of "Tian Ji's Horse Racing," demonstrates significant advantages in solving the capacity configuration problem of off-grid integrated energy systems in industrial parks. Compared to traditional heuristic algorithms, this algorithm enhances its ability to solve multi-objective, high-dimensional problems through hierarchical optimization and dynamic game theory mechanisms. When dealing with the complex variables arising from the diversified utilization of hydrogen energy, it avoids the problem of traditional algorithms getting trapped in local optima, finding the globally optimal configuration scheme through a "combination of advantages" strategy. In handling dynamic scenarios, the improved Tian Ji Horse Racing Optimization Algorithm possesses stronger real-time response capabilities, adjusting the capacity configuration strategy in real time based on dynamic factors such as fluctuations in wind and solar power output and changes in hydrogen energy demand within the park, thus addressing the shortcomings of traditional algorithms in considering dynamic factors and exhibiting poor robustness. Furthermore, through prioritizing and combining multi-dimensional objectives, this algorithm can find the optimal balance between cost, efficiency, and low-carbon goals.

[0209] The computer program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other processing device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other processing device to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions of the flowchart block(s) or step(s) of the flowchart block(s). Figure 1 one or more of the flowchart or flowchart blocks Figure 1 one or more of the flowchart or flowchart blocks

[0210] In the description of the present application, it is to be understood that the terms "first", "second", "third" and the like, merely identify features belonging to distinct categories, and do not imply or imply a relative importance or a specific number thereof. Thus, a feature identified as "first", "second", or "third" can explicitly or implicitly include one or more of the features. In the description of the present application, the meaning of "a plurality" is two or more, unless otherwise expressly and specifically limited.

[0211] Obviously, various modifications and changes can be made to the present application by those skilled in the art without departing from the spirit and scope of the present application. Thus, it is intended that the present application also include such modifications and changes insofar as they come within the scope of the claims of the present application and their equivalents.

Claims

1. A hydrogen energy park planning method based on heuristic algorithm, characterized in that, The method comprises the following steps: constructing a mathematical model of the park comprehensive energy system, wherein the system is composed of a renewable energy power generation unit, an energy conversion device, carbon capture, a gas power device, and multiple types of energy storage units; based on the mathematical model, establishing a cost objective function with the lowest annual comprehensive cost of the park comprehensive energy system as the target, setting constraints, and formulating an energy management strategy; according to the energy management strategy, solving based on the improved THRO, and planning the hydrogen energy park according to the solving result, wherein when the improved THRO is obtained, a uniform space distribution model of the initial population is established by replacing random initialization with Latin hypercube sampling, specifically as follows: By introducing Latin hypercube sampling to replace random initialization, the uniform distribution of the initial population in the solution space is ensured: ， wherein LHS represents a Latin hypercube design matrix, n is the population size, Dim is the population dimension, Up and Low represent the upper and lower limits of the decision variable, respectively; reconstructing the linear decay weight by introducing the cosine annealing strategy to establish a nonlinear time-varying parameter model, specifically as follows: The cosine annealing strategy is used to adjust the weight factor to dynamically balance the exploration and development capabilities of the algorithm; the weight decreases nonlinearly with the number of iterations, specifically as follows: ， wherein P represents the adaptive weight, t1 represents the current number of iterations, and T represents the maximum number of iterations; An elite reservation mechanism is introduced in the horse racing strategy to construct an elite-guided position update operator, specifically as follows: elite individuals are reserved in each generation of the population and introduced into the position update formula to guide the population to move towards the high-quality solution region: ， ， In the formula, Respectively, as Tianji, the king of the population elite solution; Fitness function; Tianji, the king of the population individual; A hybrid learning mechanism with time-varying probability control is designed, an optimal solution disturbance operator is introduced, and a convergence criterion based on historical solution stability is established: the optimal solution disturbance operator is introduced, specifically as follows: a time-varying probability control is designed to switch the exploration and development behavior, the probability of Levy flight exploration decreases linearly with the number of iterations, and the probability of learning from elite individuals increases correspondingly, which makes the algorithm focus on global exploration in the early stage and local development in the later stage: ， wherein is represented as Levy exploration probability decaying linearly from 0.7 to 0.

1.

2. The hydrogen energy park planning method based on a heuristic algorithm according to claim 1, wherein: when constructing the mathematical model of the park comprehensive energy system, carbon capture modeling is performed by fixing energy consumption operation energy consumption; based on the carbon capture modeling, the mathematical model of the park comprehensive energy system is constructed by modeling the electric-to-gas equipment, the hydrogen-doped gas turbine, the hydrogen-doped gas boiler, the carbon trading model, the hydrogen fuel cell, the energy storage model, and the hydrogen-powered vehicle.

3. The hydrogen energy park planning method based on a heuristic algorithm according to claim 2, wherein: when constructing the cost objective function, the cost objective function is constructed based on the annual comprehensive total cost of the system, the annual investment cost, the annual maintenance and depreciation cost, the carbon trading, the carbon sequestration cost, the natural gas purchase cost, and the abandoned wind / solar cost.

4. The hydrogen energy park planning method based on a heuristic algorithm according to claim 3, wherein: In setting the constraint condition, the installed capacity constraint, the wind and light power supply constraint, the energy storage device constraint, the power balance constraint, the carbon sequestration capacity limit, the carbon capture constraint, the gas turbine output and climbing constraint, the gas boiler constraint, the fuel cell constraint and the P2H constraint are taken as the constraint condition. 5.The hydrogen energy park planning method based on heuristic algorithm according to claim 4, characterized in that: In formulating the energy management strategy, the energy management strategy is formulated, including: Mode E1 : wind and solar power have satisfied the electrical load demand, there is remaining capacity in the battery, the maximum charging power of the battery is greater than the remaining power then the remaining power is stored in the battery; Mode E2: The wind and light output has met the electrical load demand, and part of the remaining electrical energy is stored in the battery, and the other part is converted into hydrogen by the electrolytic cell; Mode E3: The wind and light output has met the electrical load demand, and part of the remaining electrical energy is stored in the battery, the electrolytic cell runs at the maximum power, the electrolytic water is converted into hydrogen, and the other part of the remaining electrical energy is absorbed by carbon capture to absorb carbon dioxide; Mode E4: The wind and solar power output has met the demand of the electrical load, and the remaining power By the combination of power-to-gas and carbon capture Mode E5: wind and solar power cannot meet the demand of electrical load, the system lacks power Combined supply by battery and fuel cell, gas turbine Mode T1 : Fuel cell, gas turbine output has satisfied the heat load demand, there is residual capacity in the thermal storage tank, the maximum thermal charging power of the thermal storage tank is greater than the residual power then the residual power is stored in the thermal storage tank; Mode T2: fuel cell, gas turbine output has met the heat load demand, there is residual capacity in the heat storage tank, and the maximum heat charging power of the heat storage tank is less than the residual power , the heat storage tank is charged at the maximum charging power, and the remaining energy is treated as abandoned wind and light; Mode T3: The fuel cell, the gas turbine output has met the heat load demand, the heat storage tank is in the upper limit of the energy state, and the remaining energy is treated as abandoned wind and light; Mode T4: Fuel cell, gas turbine power cannot meet the electrical load demand, system lacks power Supplied jointly by the battery and the fuel cell, gas turbine; Mode H1 : The electrolyzer output has met the hydrogen load and gas hydrogen blending demand, and the hydrogen storage tank has residual capacity, and the maximum charging power of the hydrogen storage tank is greater than the residual power Then the residual power All stored in the hydrogen storage tank; Mode H2: The electrolytic cell has met the hydrogen load and gas hydrogen blending demand, the hydrogen storage tank has remaining capacity, part of the remaining hydrogen is stored in the hydrogen storage tank, and the other part is used to start the fuel cell to supplement the electrical and heat load demand in mode E5 and T4; Mode H3: The electrolytic cell has met the hydrogen load and gas hydrogen blending demand, the hydrogen storage tank has remaining capacity, part of the remaining hydrogen is stored in the hydrogen storage tank, part of the remaining hydrogen runs at the maximum power of the fuel cell to supplement the electrical and heat load demand in mode E5 and T4, and the other part is sent to the methane reactor; Mode H4: The electrolytic tank has met the hydrogen load and gas hydrogen blending demand, there is no remaining capacity in the hydrogen storage tank, there is remaining hydrogen, and there is remaining power By fuel cell and methane reactor combined consumption; Mode H5: The electrolyzer cannot meet the hydrogen load and the hydrogen blending demand of the gas, and the system lacks power Supplied by the hydrogen storage tank. 6.The hydrogen energy park planning method based on heuristic algorithm according to claim 5, characterized in that: In improving the THRO, the elite reservation mechanism is represented as: the elite individuals in each generation population are reserved and introduced into the position update formula to guide the population to move to the high-quality solution area, avoid the loss of high-quality solutions and accelerate the convergence. 7.The hydrogen energy park planning method based on heuristic algorithm according to claim 6, characterized in that: After introducing the elite reservation mechanism, the position update is improved through elite guidance. 8.The hydrogen energy park planning method based on heuristic algorithm according to claim 7, characterized in that: In improving the THRO, local perturbation search is performed near the global optimal solution. 9.The hydrogen energy park planning method based on heuristic algorithm according to claim 8, characterized in that: In the improvement of THRO, early stopping mechanism is set. When the standard deviation of the optimal fitness of 50 consecutive generations is less than 10, the iteration is terminated in advance. -6 ​ 10. A hydrogen energy park planning system based on heuristic algorithm, for implementing a hydrogen energy park planning method based on heuristic algorithm as claimed in claim 1, characterized in that, It includes: A model construction module is configured to construct a mathematical model of the park comprehensive energy system, wherein the system is composed of renewable energy power generation units, energy conversion devices, carbon capture, gas power devices and multiple types of energy storage units; A strategy formulation module is configured to establish a cost objective function based on the mathematical model, set a constraint condition and formulate an energy management strategy with the lowest annual comprehensive cost of the park comprehensive energy system as the target. The park planning module is used for planning the hydrogen energy park according to the solving result of the improved Tianji Horse Race Optimization (THRO) algorithm according to the energy management strategy, wherein, when the improved THRO algorithm is obtained, Latin hypercube sampling is used to replace random initialization to establish an initial population uniform space distribution model; a cosine annealing strategy is used to reconfigure a linear decay weight to establish a nonlinear time-varying parameter model; an elite reservation mechanism is introduced in the horse race strategy to construct an elite-guided position update operator; a hybrid learning mechanism with time-varying probability control is designed to introduce an optimal solution disturbance operator; and a convergence criterion based on historical solution stability is established.

Images