Flexible control method for synthetic ammonia in wind-light-hydrogen ammonia source network load storage integrated park

Abstract

The invention relates to the technical field of new energy and renewable energy sources, and discloses a wind-light-hydrogen-ammonia source network load storage integrated park synthetic ammonia flexible control method, which comprises the following steps of: performing multistable load setting optimization by taking the maximum daily net income as a target through wind-light prediction, time-of-use electricity price, hydrogen storage capacity and ammonia market demand; obtaining a plurality of discrete steady-state loads and operation durations thereof; a rare earth doped catalyst and a synthesis tower are adopted for cooperative control, the low-load activity interval is widened, and operation parameters are optimized; executing ultra-short-term, short-term and medium-and-long-term source network load storage cooperative scheduling, and cooperating with security boundary adaptive control; and constructing an electricity-hydrogen-ammonia integrated digital twinborn model, realizing control parameters, and realizing whole-process closed-loop rolling optimization. According to the method, through multi-steady-state load setting optimization, the operation condition of the ammonia synthesis device is dispersed into a plurality of steady-state load grades, the operation duration is reasonably arranged, park source network load storage multi-element collaboration is achieved, and the overall ammonia production cost is reduced.

Description

Technical Field

[0001] This invention relates to the field of new energy and renewable energy technology, and in particular to a flexible control method for synthetic ammonia in an integrated wind-solar-hydrogen-ammonia-grid-load-storage park. Background Technology

[0002] Renewable energy sources are characterized by significant randomness, volatility, and intermittency, making their output difficult to predict and control precisely. This poses a significant challenge to traditional chemical production processes that rely on stable power grids. Ammonia synthesis, as a basic chemical raw material, is a typical high-energy-consuming continuous production process.

[0003] In existing technologies, to resolve the contradiction between renewable energy fluctuations and the stability of synthetic ammonia production, the following three main technical solutions have been developed: First, using constant steady-state operation combined with electrochemical energy storage or hydrogen storage, large-scale energy storage devices are used to smooth out wind and solar fluctuations, allowing the synthetic ammonia system to operate within the traditional economic load range; second, adopting a fully flexible synthetic ammonia process, enabling the unit load to be rapidly adjusted in real time to follow green electricity fluctuations within a wide range of continuous intervals; and third, developing a multi-steady-state flexible operation strategy, which achieves production within a wider load range by switching between several preset discrete steady-state operating conditions.

[0004] In the process of realizing the above technical solution, the inventors of this application discovered that the existing technology has at least the following technical problems: the existing load adjustment technology for synthetic ammonia plants lacks a systematic and coordinated consideration of multiple factors such as source, grid, load and storage, and cannot achieve the overall optimal matching of new energy consumption, energy storage utilization, electricity price cost and market demand. Summary of the Invention

[0005] To overcome the above shortcomings, this invention provides a flexible control method for synthetic ammonia in an integrated wind-solar-hydrogen-ammonia-grid-load-storage park. It aims to improve the existing synthetic ammonia plant load adjustment technology, which lacks systematic coordination of multiple factors such as source, grid, load, and storage, and cannot achieve the overall optimal matching of new energy consumption, energy storage utilization, electricity price cost, and market demand.

[0006] This invention provides the following technical solution: a flexible control method for synthetic ammonia in an integrated wind-solar-hydrogen-ammonia-grid-load-storage park, comprising the following steps:

[0007] S1. Based on the short-term power output forecast data of wind and solar power, the maximum hydrogen storage capacity of the hydrogen storage tank, the minimum hydrogen storage capacity of the hydrogen storage tank, the market price of synthetic ammonia and the time-of-use electricity price of the power grid, perform multi-steady-state load tuning optimization to obtain multiple discrete steady-state loads and their corresponding operating times for each day.

[0008] S2. Based on the discrete steady-state load, the catalyst activity and synthesis tower operating parameters are adjusted by a rare earth-doped catalyst and synthesis tower coordinated control method.

[0009] S3. Based on the discrete steady-state load and its operating time, perform source-grid-load-storage coordinated scheduling control to achieve ultra-short-term regulation, short-term regulation and medium-to-long-term regulation;

[0010] S4. During the execution of source-grid-load-storage coordinated scheduling control, the synthesis tower temperature, synthesis tower pressure, catalyst activity, hydrogen storage concentration and grid voltage are monitored in real time. When any monitored parameter approaches the safety boundary, safety boundary adaptive control is executed.

[0011] S5. By building an integrated digital twin model of electricity, hydrogen, and ammonia, rolling optimization and intelligent decision-making are carried out on the control parameters of multi-steady-state load tuning optimization, rare earth doped catalyst and synthesis tower collaborative control mode, source-grid-load-storage collaborative scheduling control, and safety boundary adaptive control, forming a closed-loop control.

[0012] Preferably, the specific process of performing multi-steady-state load tuning optimization in S1 is as follows:

[0013] With the goal of maximizing the daily net profit of the system, and constrained by the daily fluctuation of hydrogen storage not exceeding 20% ​​of the maximum hydrogen storage, the hydrogen-nitrogen partial pressure ratio being maintained at around 3:1, and ammonia production meeting daily demand, a mixed integer programming model is established to solve for multiple discrete steady-state loads and their corresponding operating times for each day. The discrete steady-state loads take values ​​ranging from 30%, 50%, 75%, 100%, to 110%.

[0014] Preferably, the specific process of using the rare earth-doped catalyst and the synthesis tower in S2 is as follows:

[0015] Ruthenium-based or iron-based catalysts doped with Sm or La were used to broaden the low-load activity range. At the same time, a quadratic polynomial surrogate model was used to optimize the bed size, flow channel structure and ammonia separation temperature of the synthesis tower, so that the ammonia separation temperature increases as the load decreases.

[0016] Preferably, the specific process of performing source-grid-load-storage coordinated scheduling control in S3 is as follows:

[0017] Within a 15-minute timescale, ultra-short-term regulation is achieved by adjusting the power of the electrolyzer and replenishing or releasing gas from the hydrogen storage tank.

[0018] Within an 8-hour timescale, operating conditions are switched according to the discrete steady-state load, with the load adjustment rate controlled at 0.5-1.0% / min to achieve short-term adjustment;

[0019] Within a daily or shift time scale, the power purchase and sale strategy and hydrogen storage scale are optimized by combining the grid time-of-use electricity price and ammonia demand forecast to achieve medium- and long-term regulation.

[0020] Preferably, the step of performing adaptive control of the safety boundary includes:

[0021] The lower limit of the operating pressure of the synthesis tower is set to 11 MPa and the temperature change rate is not more than 25℃ / h. When any monitoring parameter approaches the set boundary, emergency control is performed in the following order: rapid replenishment or release of hydrogen from the hydrogen storage tank, switching to the adjacent discrete steady-state load, and entering the zero-load hot standby mode.

[0022] The preferred process for building an integrated digital twin model of electricity, hydrogen, and ammonia in S5 is as follows:

[0023] By integrating mechanistic modeling and data-driven methods, a full-process simulation model is established, encompassing wind and solar power output prediction, electrolytic hydrogen production dynamics, hydrogen storage thermodynamics, and one-dimensional unsteady-state reaction heat transfer in ammonia synthesis. Furthermore, an advanced process control (APC) module and a real-time optimization (RTO) module are integrated to perform rolling optimization and predictive maintenance of multi-steady-state load tuning parameters, rare-earth-doped catalysts and synthesis tower operating parameters, and source-grid-load-storage scheduling instructions.

[0024] Preferably, the step of performing multi-steady-state load tuning optimization further includes setting 8 hours as the basic operating cycle of discrete steady-state loads, and adding a constraint to the mixed integer programming model to minimize the number of switching between adjacent discrete steady-state loads.

[0025] Preferably, the step of adjusting the power of the electrolytic cell within a 15-minute timescale includes: enabling the electrolytic cell to complete the power adjustment within 10 minutes of cold start or 30 seconds of hot start.

[0026] Preferably, the step of optimizing the synthesis tower using a quadratic polynomial surrogate model further includes: adding a catalytic bed temperature constraint to the surrogate model to maintain the catalytic bed temperature of the rare earth-doped catalyst at 410-460℃.

[0027] Preferably, the step of constructing the integrated digital twin model of electricity, hydrogen, and ammonia further includes:

[0028] Based on equipment degradation trends and wind and solar forecast error compensation, the multi-steady-state load setting parameters are dynamically corrected every 4-8 hours, and the corrected parameters are sent down to the execution layer to form a closed-loop adaptive control throughout the entire process.

[0029] The present invention has the following beneficial effects:

[0030] 1. This invention optimizes the operating conditions of the ammonia synthesis unit by discretizing it into several steady-state load levels and rationally arranging the operating time. It deeply integrates wind and solar power output forecasts, grid time-of-use electricity prices, hydrogen storage capacity and ammonia market demand to achieve multi-factor synergy of source, grid, load and storage in the industrial park, thereby reducing the overall cost of ammonia production.

[0031] 2. This invention uses rare earth-doped catalysts to broaden the low-load activity window, and with the synergistic optimization of the synthesis tower structure and operating parameters, the temperature of the catalytic bed is always kept in the optimal activity range, and the reaction pressure and temperature fluctuations are minimal, thus avoiding equipment fatigue damage caused by continuous large-scale load changes.

[0032] 3. This invention relies on an integrated digital twin model of electricity, hydrogen, and ammonia, combining mechanism modeling and data-driven approaches to continuously and dynamically optimize load setting, catalytic reaction conditions, and source-grid-load-storage scheduling strategies. Combined with a real-time safety boundary adaptive protection mechanism, the system can automatically adjust its operating status under various disturbances such as wind and solar fluctuations and slow degradation of equipment performance, maintaining long-term safe and stable operation and possessing predictive maintenance capabilities. Attached Figure Description

[0033] Figure 1 This is a flowchart of a flexible control method for synthetic ammonia in an integrated wind-solar-hydrogen-ammonia-grid-load-storage park proposed in this invention. Detailed Implementation

[0034] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0035] Reference Figure 1 This invention provides a flexible control method for synthetic ammonia in an integrated wind-solar-hydrogen-ammonia-grid-storage park, comprising the following steps:

[0036] S1. Based on the short-term power output forecast data of wind and solar power, the maximum hydrogen storage capacity of the hydrogen storage tank, the minimum hydrogen storage capacity of the hydrogen storage tank, the market price of synthetic ammonia and the time-of-use electricity price of the power grid, perform multi-steady-state load tuning optimization to obtain multiple discrete steady-state loads and their corresponding operating times for each day.

[0037] S2. Based on the discrete steady-state load, the catalyst activity and synthesis tower operating parameters are adjusted by a rare earth-doped catalyst and synthesis tower coordinated control method.

[0038] S3. Based on the discrete steady-state load and its operating time, perform source-grid-load-storage coordinated scheduling control to achieve ultra-short-term regulation, short-term regulation and medium-to-long-term regulation;

[0039] S4. During the execution of source-grid-load-storage coordinated scheduling control, the synthesis tower temperature, synthesis tower pressure, catalyst activity, hydrogen storage concentration and grid voltage are monitored in real time. When any monitored parameter approaches the safety boundary, safety boundary adaptive control is executed.

[0040] S5. By building an integrated digital twin model of electricity, hydrogen, and ammonia, rolling optimization and intelligent decision-making are carried out on the control parameters of multi-steady-state load tuning optimization, rare earth doped catalyst and synthesis tower collaborative control mode, source-grid-load-storage collaborative scheduling control, and safety boundary adaptive control, forming a closed-loop control.

[0041] Specifically, the discrete steady-state load generated by S1 is used to determine the target load level of the ammonia synthesis unit at different time periods. S2 and S3 adjust the reactor operating conditions and park-level energy allocation, respectively, and are executed based on the results of S1. S4 is used to ensure that the system operating parameters are kept within the preset safety range, and maintains operational stability by executing corresponding control actions when deviations occur. The digital twin model constructed by S5 models the wind and solar forecasts, electrolytic hydrogen production dynamics, hydrogen storage behavior, and ammonia synthesis reaction characteristics, and uses model predictive control methods to periodically correct the input parameters of each step to improve control accuracy and coordination between modules.

[0042] The solution for S1 is used as input for S2 through S4. S2 involves adjusting catalyst activity, bed operating conditions, and loading levels. Directly related. The scheduling strategy in S3 is based on The numerical values ​​determine the load switching times for different time periods. S4 uses the parameters monitored in real time by the model and performs boundary judgments based on the operating range determined by S1 to S3. S5 records the parameters generated in the above steps and inputs them into the constructed digital twin model for rolling prediction and parameter correction in subsequent control cycles.

[0043] This implementation method achieves coordinated control of wind and solar power generation, electrolytic hydrogen production, hydrogen storage, synthetic ammonia production, and grid interaction within the park through data transmission and parameter updates between S1 and S5.

[0044] Furthermore, the specific process of performing multi-steady-state load tuning optimization in S1 is as follows:

[0045] With the goal of maximizing the daily net profit of the system, and constrained by the daily fluctuation of hydrogen storage not exceeding 20% ​​of the maximum hydrogen storage, the hydrogen-nitrogen partial pressure ratio being maintained at around 3:1, and ammonia production meeting daily demand, a mixed integer programming model is established to solve for multiple discrete steady-state loads and their corresponding operating times for each day. The discrete steady-state loads take values ​​ranging from 30%, 50%, 75%, 100%, to 110%.

[0046] Specifically, short-term wind and solar power forecasts are used as inputs to the optimization model. Wind power output forecasts use a 5-minute forecast sequence, while solar power output forecasts use a 4- to 8-hour forecast sequence. The hydrogen storage tank provides maximum and minimum hydrogen storage capacities as constraints on the hydrogen storage range. The market price of synthetic ammonia and the time-of-use electricity price are used as economic parameters to determine operating costs and revenues under different load conditions.

[0047] In the multi-steady-state load setting optimization model, the daily operation process is divided into multiple time periods, and a load level is selected for each time period. The load level can be set to 30%, 50%, 75%, 100%, or 110%. This is achieved through integer variables. This indicates whether the k-th load level is selected during time period t, where ,satisfy: Where K represents the number of load levels.

[0048] ammonia production at each load level Hydrogen consumption rate The energy consumption parameters are all pre-calibrated fixed values ​​derived from the steady-state performance test of the device.

[0049] The objective function for maximizing the system's daily net return is expressed as:

[0050] ;

[0051] in, Indicates the number of time periods;

[0052] This indicates the price per unit of ammonia product;

[0053] Indicates the grid electricity price for time period t;

[0054] This represents the unit energy consumption at load level k;

[0055] This represents the operation and maintenance cost at load level k;

[0056] Indicates the length of a single time period.

[0057] The dynamic changes in hydrogen storage capacity are represented by the hydrogen storage balance equation:

[0058] ;

[0059] in, This indicates the amount of hydrogen in the hydrogen storage tank at time t;

[0060] This indicates the hydrogen production rate of the electrolysis unit at time t;

[0061] This indicates the hydrogen release rate from the hydrogen storage tank.

[0062] The daily fluctuation constraints for hydrogen storage are set as follows: ; to ensure that the hydrogen storage system maintains a reasonable dynamic range over a 24-hour period, among which This indicates the maximum hydrogen storage capacity of the hydrogen storage tank. This indicates the initial hydrogen storage capacity at the start of the cycle.

[0063] The hydrogen-nitrogen partial pressure ratio constraint is adopted in the following form: ;in, The partial pressure of hydrogen at load level k, The nitrogen partial pressure at load level kkk This is the allowable deviation.

[0064] The constraint that ammonia production must meet daily demand is expressed as: ;in, This represents the daily ammonia demand.

[0065] The constraint on the number of discrete steady-state load switching operations is achieved by introducing variables. This indicates the behavior of switching operating conditions, where: ; and by minimizing: To reduce unnecessary load switching.

[0066] In one embodiment, the basic operating cycle for discrete steady-state conditions is set to 8 hours, i.e., 8 consecutive hours. The same load level is used for each time period. The model uses mandatory constraints: ;in, This is to ensure that the minimum operating cycle meets the set value.

[0067] By solving the above mixed-integer programming model, combinations of several load levels (30%, 50%, 75%, 100%, and 110%) and their corresponding runtimes are obtained. The solution results serve as input parameters for subsequent S2 to S4 processes, determining the baseline load for catalyst control strategy, scheduling strategy, and safety boundary assessment.

[0068] Furthermore, the specific process of using the rare earth-doped catalyst and the synthesis tower in S2 is as follows:

[0069] Ruthenium-based or iron-based catalysts doped with Sm or La were used to broaden the low-load activity range. At the same time, a quadratic polynomial surrogate model was used to optimize the bed size, flow channel structure and ammonia separation temperature of the synthesis tower, so that the ammonia separation temperature increases as the load decreases.

[0070] Specifically, the synthesis tower employs a multi-layer catalyst bed structure. Ruthenium-based catalysts or iron-based catalysts doped with Sm or La are selected to improve the ammonia synthesis rate within the 30% to 75% loading range. Rare earth element doping of the catalyst adjusts the electronic structure around the metal center, enabling it to maintain stable reactivity under lower reaction temperatures and lower reactant partial pressures. The relationship between the activity and temperature of this type of catalyst can be expressed by an empirical function:

[0071] ;

[0072] in, This indicates the catalyst activity at temperature T;

[0073] Indicates the activity coefficient;

[0074] Indicates the apparent activation energy;

[0075] Represents the gas constant;

[0076] This indicates the operating temperature of the bed.

[0077] Rare earth doped catalysts Unlike undoped catalysts, this allows for the maintenance of suitable activity in a lower temperature range. Depending on the loading level... Set bed temperature satisfy: To ensure reaction stability, the temperature distribution is regulated by multiple temperature control points inside the synthesis tower.

[0078] The structural parameters of the synthesis tower were calibrated using a quadratic polynomial surrogate model. The surrogate model was based on experimental data fitting key geometric and operational parameters within the tower. The geometric parameters included bed diameter D, bed height H, and flow channel cross-sectional area S, while the operational parameters included gas inlet temperature. And the gas flow rate F. The surrogate model takes the form of:

[0079] ;

[0080] in,

[0081] Indicates the output variable (which can be bed outlet temperature, tower pressure drop, or unit bed reaction rate);

[0082] ~ Represents the fitting coefficient;

[0083] These represent the structural and operational parameters, respectively.

[0084] The above model is used to calculate the optimal combination of structure and operating parameters of the synthesis tower based on a given load level, so as to maintain stable temperature and pressure distribution in the reactor under different load conditions.

[0085] During load switching, the ammonia separation temperature is adjusted according to the load level. (Set ammonia separation temperature) for: ;in, This represents the baseline ammonia fractionation temperature under 100% load conditions; Indicates the load level; This represents the temperature regulation coefficient, determined based on experimental data. When the load decreases... Increase the pressure to ensure that the saturation temperature of the ammonia gas to be separated matches the pressure inside the tower. This adjustment method is used to maintain a stable ammonia separation efficiency under different loads.

[0086] The flow channel structure is adjusted by controlling the gas distribution device inside the synthesis tower. For different load levels, the gas channel cross-sectional area S and the total inlet flow rate F are configured according to the following relationship: ;in, This represents the apparent gas velocity required at load level k.

[0087] This parameter is determined based on numerical simulation results or device calibration experiments.

[0088] Based on the above catalyst activity model, structural proxy model, and ammonia fractionation temperature regulation formula, when implementing discrete load levels... At that time, the bed temperature setpoint can be determined in real time. Flow channel structure parameters Inlet temperature and apparent gas velocity This allows for the control of the reaction conditions in the synthesis tower.

[0089] The coordinated regulation of catalyst and synthesis tower operating parameters is performed based on the following logic: for each load level First calculate the output of the proxy model. If the output is within the preset operating range, the corresponding structural and operational parameters are used; if it deviates from the operating range, adjustments are made. , , This method brings the output back to an acceptable range.

[0090] By jointly adjusting the catalyst performance and the operating conditions inside the tower, the reaction rate, conversion rate, pressure drop and heat distribution corresponding to different load levels can meet the requirements for stable operation, providing a stable reaction condition basis for subsequent scheduling steps.

[0091] Furthermore, the specific process of performing source-grid-load-storage coordinated scheduling control in S3 is as follows:

[0092] Within a 15-minute timescale, ultra-short-term regulation is achieved by adjusting the power of the electrolyzer and replenishing or releasing gas from the hydrogen storage tank.

[0093] Within an 8-hour timescale, operating conditions are switched according to the discrete steady-state load, with the load adjustment rate controlled at 0.5-1.0% / min to achieve short-term adjustment;

[0094] Within a daily or shift time scale, the power purchase and sale strategy and hydrogen storage scale are optimized by combining the grid time-of-use electricity price and ammonia demand forecast to achieve medium- and long-term regulation.

[0095] Specifically, the source-grid-load-storage coordinated scheduling control is based on discrete steady-state load levels. It takes the operating period as input and divides the entire operating cycle into ultra-short-term, short-term and daily time periods to coordinate the material flow and energy flow of the electrolytic hydrogen production unit, hydrogen storage tank, grid interface and ammonia synthesis unit.

[0096] Execute ultra-short-term adjustments within a 15-minute timescale. Set the power setpoint for the electrolysis hydrogen production unit. ,satisfy:

[0097] ;

[0098] in, This is the baseline electrolysis power corresponding to the current load level;

[0099] This is the correction amount obtained based on the prediction deviation of wind and solar power.

[0100] The filling and discharging of hydrogen storage tanks is performed based on the difference between the hydrogen production rate and the hydrogen consumption rate:

[0101] ;

[0102] in, It indicates replenishing Qi. It indicates the release of gas.

[0103] The electrolyzer has both cold and hot start capabilities. The cold start time is set to 10 minutes, and the hot start time is set to 30 seconds. When performing ultra-short-term regulation, if the wind and solar power output suddenly increases and the electrolyzer is in a cold state, the scheduling module will initiate the cold start process, allowing it to reach the target power after 10 minutes of continuous cold operation; if it is in a hot state, the power adjustment will be completed within 30 seconds.

[0104] Short-term adjustments are implemented on an 8-hour timescale. These short-term adjustments are based on discrete steady-state load levels. Let L(t) be the target load. Set the total load L(t) of the ammonia synthesis unit and implement load switching according to a linear variation law, satisfying: ;in, This is the upper limit of the load regulation rate, ranging from 0.5% / min to 1.0% / min.

[0105] Short-term regulation is achieved through setpoint adjustments distributed across the synthesis tower, compressor units, and circulation system. Each unit receives load targets from the dispatch center and allocates them to execution parameters such as gas flow rate, circulation ratio, and compressor power.

[0106] During the 8-hour operating cycle, the load remains And at the end of the cycle, switch to the next load level according to the scheduling table. The load switching time is obtained from S1. Decide.

[0107] On a daily or shift-based timescale, medium- to long-term adjustments will be implemented. An electricity purchase and sale strategy combined with hydrogen storage capacity arrangements will be adopted to achieve overall energy balance. Daily purchase... Electricity sales And optimize based on the electricity price curve:

[0108] ;

[0109] in, This represents the electricity price during time period t.

[0110] Hydrogen storage capacity optimization is based on intraday hydrogen storage variation constraints:

[0111] ;

[0112] ;

[0113] To ensure the daily circulation stability of the hydrogen storage system.

[0114] Ammonia demand forecast input Used to determine minimum production requirements: This constraint is used to ensure that the medium- and long-term scheduling strategy can meet the established production requirements.

[0115] During the coordinated scheduling process, the energy interaction relationship between the electrolytic hydrogen production unit, hydrogen storage tank, and the power grid satisfies the following: ;in, It signifies the effort put into the scenery; This indicates the load of the industrial park excluding the electrolytic cells.

[0116] Short-term and medium-to-long-term scheduling are achieved by sharing the same hydrogen storage state variables. Achieving connectivity between time scales. Load switching actions performed during short-term scheduling are based on resource allocation strategies obtained from upper-level medium- and long-term scheduling, namely the target operating range of the hydrogen storage system and the power purchase and sale arrangements.

[0117] The coordination between electrolyzer power, hydrogen storage behavior, ammonia synthesis unit load, and power purchase and sale from the grid is determined by the control quantities calculated by the dispatch center based on the above constraints and then distributed to the controllers of each equipment. By continuously executing dispatch strategies for different time periods, the entire park maintains a stable material and energy balance under fluctuating wind and solar power conditions.

[0118] Furthermore, the step of performing adaptive control of the safety boundary includes:

[0119] The lower limit of the operating pressure of the synthesis tower is set to 11 MPa and the temperature change rate is not more than 25℃ / h. When any monitoring parameter approaches the set boundary, emergency control is performed in the following order: rapid replenishment or release of hydrogen from the hydrogen storage tank, switching to the adjacent discrete steady-state load, and entering the zero-load hot standby mode.

[0120] Specifically, the safety boundary adaptive control makes judgments based on real-time monitoring data of parameters such as temperature, pressure, catalyst activity, hydrogen concentration in the hydrogen storage tank, and grid voltage inside the synthesis tower. The system operates at a fixed time step. The above variables are sampled, and corresponding control actions are executed according to the decision logic.

[0121] The pressure boundary conditions for the synthesis tower are set as follows: ;in, This represents the operating pressure of the synthesis tower at time t; Set to 11 MPa.

[0122] The boundary conditions for the rate of temperature change of the synthesis tower are set as follows: ;in, This represents the bed temperature monitoring value at time t; This indicates the upper limit of the rate of temperature change.

[0123] Catalyst activity monitoring value Boundary judgment is performed based on the trend of activity changes. When the activity falls below a set threshold... If this occurs, the reaction conditions need to be adjusted to restore them to the acceptable range. These conditions are defined as follows: .

[0124] Hydrogen concentration monitoring values ​​in hydrogen storage tanks are expressed as a percentage of concentration. It indicates, and satisfies: ;in The standards are set based on the materials and operating conditions of the storage tank.

[0125] Grid voltage monitoring value Must meet: ;

[0126] By monitoring the changes in variables under the aforementioned boundary conditions, a basis for judgment can be provided for subsequent emergency control.

[0127] When any monitoring parameter meets one of the following conditions:

[0128] ;

[0129] This means it is judged to be close to the safety boundary. Among them, This is a preset allowable offset, used to trigger control actions in advance.

[0130] When the above conditions are met, three types of emergency control measures will be implemented in a fixed order.

[0131] The first type of regulation involves rapid replenishment or release of hydrogen from the storage tank. Based on the direction of pressure and temperature changes in the synthesis tower and the hydrogen consumption, the filling and discharging rates of the hydrogen storage tank are calculated:

[0132] ;

[0133] in, , For adjustment coefficients; This is a reference temperature.

[0134] This feature is used to adjust the hydrogen flow rate into the tower in a short period of time to balance pressure and temperature.

[0135] The second type of control involves switching to an adjacent discrete steady-state load. When the load level is... When switching targets to or The specific switching direction is determined based on the boundary deviation direction. The load adjustment command is set as follows:

[0136] ;

[0137] in, The value is taken as the load level interval, and the rate of change satisfies: ;

[0138] The third type of control involves entering a zero-load hot standby mode. In this mode, the ammonia synthesis unit stops feeding and maintains minimal heat input within the tower, keeping the bed temperature stable within a set range. Energy input for hot standby mode. satisfy:

[0139] ;

[0140] in, This refers to heat loss. Minimum heat compensation required to maintain the thermal stability of the equipment.

[0141] Duration of hot standby mode Set to no more than 168 hours.

[0142] The three types of control measures are implemented in sequence. If the first type of control measure can restore the monitored variable to the safe range, the second and third types of control measures are not implemented; if the first type of control measure is insufficient to restore the variable, the second type of control measure is implemented; if the second type of control measure is still insufficient to restore the variable, the third type of control measure is implemented.

[0143] Safety boundary adaptive control, through real-time monitoring, boundary determination, and sequential execution of control actions, enables the reactor to adjust its operating status in a timely manner when deviations occur in pressure, temperature, and catalyst state, thereby ensuring that the load execution and scheduling plans of subsequent steps can be carried out under predetermined safety conditions.

[0144] Furthermore, the specific process of building the integrated digital twin model of electricity, hydrogen, and ammonia in S5 is as follows:

[0145] By integrating mechanistic modeling and data-driven methods, a full-process simulation model is established, encompassing wind and solar power output prediction, electrolytic hydrogen production dynamics, hydrogen storage thermodynamics, and one-dimensional unsteady-state reaction heat transfer in ammonia synthesis. Furthermore, an advanced process control (APC) module and a real-time optimization (RTO) module are integrated to perform rolling optimization and predictive maintenance of multi-steady-state load tuning parameters, rare-earth-doped catalysts and synthesis tower operating parameters, and source-grid-load-storage scheduling instructions.

[0146] Specifically, the digital twin model includes a wind and solar sub-model, an electrolytic hydrogen production sub-model, a hydrogen storage tank model, and a synthetic ammonia reactor sub-model. The sub-models are connected through material flow, energy flow, and power flow parameters, and prediction and updates are performed according to a unified time step.

[0147] Wind and solar power output is described using a predictive model. Wind power prediction values. and photovoltaic power forecast Calculations based on historical data and short-term meteorological inputs satisfy the following:

[0148] ;

[0149] in, Indicates time Total renewable energy output.

[0150] The dynamics of hydrogen production by electrolysis are expressed by the relationship between electrolysis power and hydrogen production rate:

[0151]

[0152] in, Indicates electrolysis efficiency; Indicates the hydrogen production rate; This indicates the lower heating value of hydrogen.

[0153] The thermodynamic model of the hydrogen storage tank describes the changes in hydrogen storage capacity, tank pressure, and temperature over time. The hydrogen storage capacity is updated according to the following formula: ;in, For ammonia synthesis operation The rate of hydrogen consumption; The rate at which gas is replenished or released from the hydrogen storage tank.

[0154] The pressure in the storage tank can be expressed using the gas law as follows: ;in: For storage capacity, This refers to the hydrogen storage temperature.

[0155] The helium synthesis reactor employs a one-dimensional unsteady-state reaction heat transfer model. Let the axial coordinate be... The temperature is The mole fraction of helium is Its governing equations are: ;in, The density of the gas; Specific heat capacity at constant pressure; Thermal conductivity; The ammonia synthesis reaction releases heat; denoted as the reaction rate.

[0156] The reaction rate is expressed in Arrhenius form:

[0157] ;

[0158] in, Pre-exponential factor, For activation energy, This is a function related to the partial pressure of the reactants.

[0159] The coupling of bed outlet temperature and reaction rate is used to correspond with the surrogate model parameters in S2, realizing the online calculation closed loop of catalyst and column structure.

[0160] The aforementioned mechanistic model, combined with the data-driven dynamic model, forms the core of digital twin prediction. For components that are difficult to model precisely or exhibit performance aging, such as circulating compressors and heat exchangers, their performance changes are calibrated using a data-driven dynamic approach.

[0161]

[0162] in, Indicates estimated values ​​of equipment parameters; Indicates the measured value; Indicates the model output; To update the step size.

[0163] The digital twin model and control system are connected via APC and RTO modules. The APC module outputs and adjusts operational setpoints based on the predictive model, including helium fractionation temperature, inlet gas temperature, and circulation rate. RTO module operation optimization issues:

[0164]

[0165] in, To optimize decision variables; , As weight; For target load; The target hydrogen storage capacity.

[0166] The optimization cycle is set to 4 to 8 hours. At the end of each cycle, the multi-steady-state load setting parameters, synthesis tower operating parameters, and source-grid-load-storage scheduling instructions are updated, and the updated parameters are sent to the execution layer. The digital twin model also records equipment degradation trends. By comparing the predicted performance with the actual measurement deviation, if the deviation exceeds a set threshold, the equipment is added to the maintenance warning list to achieve predictive maintenance.

[0167] Through the above process, the digital twin model continuously receives real-time data and updates model parameters during the operation cycle, which are then used to adjust operating conditions in subsequent scheduling cycles to achieve the operation from S1 to S4.

[0168] Furthermore, the step of performing multi-steady-state load tuning optimization also includes setting 8 hours as the basic operating cycle of discrete steady-state loads, and adding a constraint to minimize the number of switching between adjacent discrete steady-state loads in the mixed integer programming model; the step of adjusting the power of the electrolytic cell within a 15-minute time scale includes: enabling the electrolytic cell to complete power adjustment within 10 minutes of cold start or 30 seconds of hot start.

[0169] Specifically, during multi-steady-state load tuning optimization, to ensure stable temperature, pressure, and catalyst activity ranges in the ammonia synthesis unit under steady-state conditions, the daily operating time is divided into several consecutive 8-hour periods, with each period using only one load level. The time step is set to... The number of time steps corresponding to an 8-hour period is: For each time period Let the load level be Then we have: ;in, It is a binary variable, representing the time period. Should I select a load level? ; The number of time periods corresponding to the basic operating cycle.

[0170] Based on the above constraints, the model will automatically generate a load distribution scheme that meets the minimum operating cycle requirement during the solution process. To reduce pressure and temperature disturbances caused by load switching, a constraint minimizing the number of switching operations is introduced into the mixed-integer programming model. Auxiliary variables are defined as follows: And set the target function for the number of switching operations as:

[0171] ;

[0172] in, Indicates from time period Switch to time period Changes in load levels; This represents the total number of running time periods.

[0173] The objective function and the main objective function are combined using weighted coefficients to form a multi-objective optimization problem, balancing economic efficiency and operational stability. When adjusting the electrolytic cell power within a 15-minute timescale, the electrolytic cell needs to select the corresponding startup mode based on its current state. The change in the electrolytic cell power setpoint is expressed by the following formula:

[0174] ;

[0175] in, For a moment The power of the electrolytic cell; This represents the permissible power adjustment amount. When the electrolytic cell is in a cold state, the maximum power change rate is set as follows: ;

[0176] in, To ensure the rated power is reached, the cold start process must be completed within 10 minutes and the specified power is achieved. When the electrolytic capacitor is hot, the maximum power change rate is set as follows: This adjustment capability allows for switching of the target power setpoint within 30 seconds. The electrolytic cell's operating state can be determined by temperature, internal pressure, and operating time. Let the set of characteristic parameters of the electrolytic cell be... When the state function satisfies: If the condition is met, it is determined to be in a hot state; otherwise, it is determined to be in a cold state. The hot-state determination value is determined based on the equipment calibration parameters.

[0177] The hydrogen replenishment and release strategies for the hydrogen storage tank are executed synchronously with the power adjustment of the electrolyzer. Changes in the hydrogen quantity in the storage tank satisfy the following: ;when When replenishing Qi, Gas release is performed at specific times. The upper limits for both gas replenishment and gas release rates are set as follows:

[0178] ;

[0179] in, This represents the maximum allowable charge / discharge rate for the hydrogen storage device.

[0180] The 15-minute adjustment cycle is interconnected with the 8-hour basic operating cycle. Ultra-short-term adjustment is used to offset short-term power deviations caused by wind and solar power fluctuations; short-term adjustment is used to execute the discrete steady-state load distribution derived from S1; both share the hydrogen storage capacity variable. And it is coordinated through the scheduling module.

[0181] During continuous operation, the dispatching system calculates power adjustment commands in real time based on wind and solar forecast errors, the dynamic response capability of the electrolyzers, and the remaining hydrogen storage capacity. and hydrogen storage tank regulation instructions This ensures that the system operating parameters meet the constraints of the optimization model.

[0182] Furthermore, the step of optimizing the synthesis tower using a quadratic polynomial surrogate model also includes: adding a catalytic bed temperature constraint to the surrogate model to keep the catalytic bed temperature of the rare earth doped catalyst at 410-460℃.

[0183] The steps for building the integrated digital twin model of electricity, hydrogen, and ammonia also include:

[0184] Based on equipment degradation trends and wind and solar forecast error compensation, the multi-steady-state load setting parameters are dynamically corrected every 4-8 hours, and the corrected parameters are sent down to the execution layer to form a closed-loop adaptive control throughout the entire process.

[0185] Specifically, when optimizing the synthesis tower structure and operating parameters using a surrogate model, it is necessary to ensure that the catalytic bed temperature is within the suitable operating range of the rare-earth-doped catalyst. Regarding the output variables of the surrogate model... Includes predicted bed temperature values Add the following constraints: ;in, The predicted values ​​are provided by the surrogate model:

[0186] ;

[0187] In the formula, The diameter of the bed layer; This refers to the height of the bed layer; The cross-sectional area of ​​the flow channel; Inlet temperature; This refers to the gas flow rate; These are the calibrated model coefficients; For elements in the set of input variables.

[0188] Based on the above constraints, when the surrogate model's output prediction shows a trend deviating from the temperature range, adjustments will be made. Operator variables are used to ensure the final calculation results meet temperature requirements. Furthermore, at load levels... At lower temperatures, the agent type may output a lower bed temperature. In this case, the ammonia separation temperature or inlet temperature will be automatically increased to compensate for the heat reduction caused by the decrease in load.

[0189] During the optimization process using the surrogate model, the final feasible solution must simultaneously satisfy the bed temperature constraint, the column pressure drop constraint, and the reaction rate constraint. The column pressure drop can be expressed as: ;in This is a function established based on the flow resistance formula; This represents the gas density.

[0190] The reaction rate constraint is expressed using a one-dimensional reaction kinetics expression:

[0191] ;

[0192] in, This is the rate term that depends on the voltage divider.

[0193] When the function r is lower than the set minimum rate value At that time, the reaction rate can be increased by adjusting the relevant input variables to keep it within an acceptable range.

[0194] When constructing the integrated digital twin model of electricity, hydrogen, and ammonia, it is necessary to periodically correct the multi-steady-state load setting parameters to reduce the impact of wind and solar forecast time lag, operational fluctuations, and equipment performance degradation on the operating results. The correction period is set to 4-8 hours. At the end of each correction period, the actual wind and solar power output is analyzed. Compared with the predicted value The deviation is calculated as follows: This deviation is used to update the power prediction model for the next cycle. The update formula is:

[0195] ;

[0196] in, To correct the gain coefficient.

[0197] Equipment degradation trends are obtained by monitoring the performance parameters of key equipment. For example, compressor efficiency. The update uses:

[0198] ;

[0199] in, This is a real-time power consumption measurement of the compressor; Predict power consumption for the model; To improve the update gain.

[0200] The input parameters of the multi-steady-state load tuning model are corrected based on the updated equipment performance parameters, including reaction rate, tower pressure drop, and equipment energy consumption curves.

[0201] The dynamic correction of the dopant number for multi-steady-state loading is achieved through the following means:

[0202] ;

[0203] in, and These are the revised load level and the original load level, respectively. This represents the operational deviation predicted by the digital twin model; This is the load correction factor.

[0204] After the correction is completed, the updated parameters are sent to the execution layer, including APC, DCS and scheduling module, through the control bus, so that the new optimal parameters are executed in subsequent operating cycles, realizing closed-loop adaptive control.

[0205] The periodic correction process of the digital twin model forms a cyclical relationship with S1 to S4: In each cycle, the digital twin model updates the model parameters based on the collected data and recalculates the load level, equipment operating conditions and scheduling instructions; in the next cycle, the execution layer runs according to the updated parameters, so that the system can automatically converge to a new operating point according to external resource fluctuations and equipment performance changes.

[0206] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A wind-solar-hydrogen-ammonia source, load and storage integrated park synthetic ammonia flexible control method, characterized in that, The method comprises the following steps: S1, according to the wind and light short-term output prediction data, the maximum hydrogen storage capacity of the hydrogen storage tank, the minimum hydrogen storage capacity of the hydrogen storage tank, the synthetic ammonia market price and the time-of-use electricity price, multi-steady-state load setting optimization is performed to obtain multiple discrete steady-state loads and corresponding running time lengths per day; S2, according to the discrete steady-state load, a rare earth doped catalyst and a synthetic tower are used in a collaborative control mode to adjust the catalyst activity and the synthetic tower operating parameters; S3, based on the discrete steady-state load and the running time length, source-grid-load-storage collaborative scheduling control is performed to achieve ultra-short-term regulation, short-term regulation and medium-long-term regulation; S4, during the execution of the source-grid-load-storage collaborative scheduling control, the synthetic tower temperature, the synthetic tower pressure, the catalyst activity, the hydrogen storage concentration and the grid voltage are monitored in real time, and when any monitored parameter approaches the safety boundary, safety boundary adaptive control is performed; S5, by building an electric-hydrogen-ammonia integrated digital twin model, the control parameters of the multi-steady-state load setting optimization, the rare earth doped catalyst and the synthetic tower collaborative control mode, the source-grid-load-storage collaborative scheduling control and the safety boundary adaptive control are rolled over and optimized and intelligently decided to form a closed-loop control.

2. The wind-solar-hydrogen-ammonia source, load and storage integrated park synthetic ammonia flexible control method according to claim 1, characterized in that, The specific process of performing multi-steady-state load setting optimization in S1 is as follows: A mixed integer programming model is established with the maximum system daily net income as the target, the daily hydrogen storage fluctuation not exceeding 20% of the maximum hydrogen storage capacity, the hydrogen-nitrogen partial pressure ratio being kept near 3:1, and the ammonia production meeting the daily demand as the constraints, to obtain multiple discrete steady-state loads and corresponding running time lengths per day, and the discrete steady-state load is in the range of 30%, 50%, 75%, 100% and 110%.

3. The wind-solar-hydrogen-ammonia source, load and storage integrated park synthetic ammonia flexible control method according to claim 1, characterized in that, The specific process of using the rare earth doped catalyst and the synthetic tower collaborative control mode in S2 is as follows: A ruthenium-based or iron-based catalyst doped with Sm or La is used to broaden the low-load activity interval, and a quadratic polynomial proxy model is used to optimize the synthetic tower bed size, flow channel structure and ammonia separation temperature, so that the ammonia separation temperature increases as the load decreases.

4. The wind-solar-hydrogen-ammonia source, load and storage integrated park synthetic ammonia flexible control method according to claim 1, characterized in that, The specific process of performing source-grid-load-storage collaborative scheduling control in S3 is as follows: In a 15-minute time scale, the electrolytic cell power and the hydrogen storage tank gas charging or releasing are adjusted to achieve ultra-short-term regulation; In an 8-hour time scale, the working condition is switched according to the discrete steady-state load, and the load regulation rate is controlled at 0.5-1.0% / min to achieve short-term regulation; In a daily or shift time scale, the electricity purchase and sale strategy and the hydrogen storage scale are optimized in combination with the time-of-use electricity price and the ammonia demand prediction to achieve medium-long-term regulation.

5. The wind-solar-hydrogen-ammonia source, load and storage integrated park synthetic ammonia flexible control method according to claim 1, characterized in that, The steps of performing safety boundary adaptive control include: The lower limit of the synthetic tower operating pressure is set to 11 MPa, and the temperature change rate is not more than 25℃ / h, when any monitored parameter approaches the set boundary, the emergency control is performed in the order of fast charging or releasing of the hydrogen storage tank, switching to the adjacent discrete steady-state load and entering the zero-load standby mode.

6. The wind-solar-hydrogen-ammonia source, load and storage integrated park synthetic ammonia flexible control method according to claim 1, characterized in that, The specific process of building an electric-hydrogen-ammonia integrated digital twin model in S5 is as follows: The fusion mechanism modeling and data-driven method establishes a full-process simulation model including wind and light output prediction, electrolytic hydrogen dynamic, hydrogen storage thermodynamics, and one-dimensional non-steady-state reaction heat of synthetic ammonia, and integrates advanced process control (APC) and real-time optimization (RTO) modules. The multi-steady-state load setting parameters, rare earth doped catalyst and synthetic tower operation parameters, source network load storage scheduling instructions are rolled and optimized and predictive maintenance is performed.

7. The wind-solar-hydrogen-ammonia source, load and storage integrated park synthetic ammonia flexible control method according to claim 1, characterized in that, The step of performing multi-steady-state load setting optimization further includes setting 8 hours as the basic operation cycle of discrete steady-state load, and adding a constraint of minimizing the number of switches between adjacent discrete steady-state loads in the mixed integer programming model.

8. The wind-solar-hydrogen-ammonia source, load and storage integrated park synthetic ammonia flexible control method according to claim 1, characterized in that, The step of adjusting the power of the electrolytic cell in a 15-minute time scale includes: making the electrolytic cell complete power adjustment within 10 minutes of cold start or 30 seconds of hot start.

9. The wind-solar-hydrogen-ammonia source, load and storage integrated park synthetic ammonia flexible control method according to claim 1, characterized in that, The step of optimizing the synthetic tower using a quadratic polynomial proxy model further includes adding a catalyst bed temperature constraint in the proxy model to keep the catalyst bed temperature of the rare earth doped catalyst at 410-460 ℃.

10. The wind-solar-hydrogen-ammonia source, load and storage integrated park synthetic ammonia flexible control method according to claim 1, characterized in that, The step of building an electric hydrogen ammonia integrated digital twin model further includes: Based on the equipment degradation trend and wind and light prediction error compensation, the multi-steady-state load setting parameters are dynamically corrected every 4-8 hours, and the corrected parameters are sent to the execution layer to form a full-process closed-loop adaptive control.

Images