Method and system for collaborative control and optimal configuration of ammonia-hydrogen power supply access to power distribution network

Abstract

The application provides a kind of ammonia hydrogen power access power distribution network collaborative control and optimization configuration method and system, belongs to power distribution network optimization operation and integrated energy scheduling field.The application constructs ammonia hydrogen power system mathematical model covering electric-hydrogen-ammonia-hydrogen-electric full energy conversion process, forms the dual energy storage complementary architecture combining hydrogen storage short cycle and ammonia storage long cycle, introduces hydrogen production uncertainty operation constraint;Build a multi-time scale layered collaborative control architecture matching the response characteristics of each link of ammonia hydrogen system, realize multi-region parallel scheduling combined with the distributed iterative solution framework based on boundary consistency constraint;Through the output configuration scheme of wind-solar-fire-ammonia coupling double-layer optimization scheduling model, the renewable energy consumption level can be effectively improved, and the economy and reliability of power distribution network operation are guaranteed.

Description

Technical Field

[0001] This invention belongs to the field of power distribution network optimization operation and integrated energy dispatching technology, specifically involving a method and system for coordinated control and optimized configuration of ammonia-hydrogen power supply access to the power distribution network. Background Technology

[0002] Electrolysis hydrogen production technology is widely regarded as an efficient means to mitigate the volatility of wind and solar renewable energy and achieve more flexible resource optimization across a wider range. Ammonia, as a high-density, easily liquefied hydrogen storage carrier, has significantly lower storage and transportation costs than high-pressure gaseous hydrogen storage. The energy storage cycle process produces no secondary pollution and can achieve near-zero carbonization in the "electricity-hydrogen-ammonia-electricity" energy storage cycle, providing long-term energy storage technology support for distributed energy systems.

[0003] The ammonia-hydrogen power system, as a novel energy coupling form, is a crucial support for building a clean, low-carbon, safe, and efficient modern energy system, and is of great significance for improving the regulation capacity and multi-energy complementarity of power distribution networks. Currently, under the "dual-carbon" strategy, accelerating the engineering and large-scale development of cutting-edge energy technologies such as hydrogen and ammonia energy is an important path to promote energy structure transformation, ensure energy security, and lead the construction of new power systems.

[0004] However, due to the complex multi-stage linkage characteristics of the ammonia-hydrogen energy coupling system, existing models mostly cover single stages and lack full-process coupling constraints. Furthermore, the response time scales of various stages within the system—chemical energy storage, co-firing power generation, and grid dispatch—differ significantly, and there is a lack of suitable hierarchical collaborative control mechanisms. Therefore, the widespread application of this type of system in actual distribution networks still faces many technical obstacles. Especially in the absence of systematic planning, scenario-based adaptation solutions, and collaborative control mechanisms, the ammonia-hydrogen power system struggles to achieve efficient collaboration and complementarity with various distributed resources in the distribution network, affecting its role as a regulatory resource and hindering the development of intelligent and multi-energy integrated distribution networks.

[0005] With the continuous increase in energy demand in urban and rural areas, especially in rural, mountainous, and island regions, the "ammonia-hydrogen" power system, characterized by its cleanliness, flexibility, and self-sufficiency, has become an important technological choice for promoting on-site energy conversion and improving power supply reliability in these areas due to the high economic and technical difficulties in implementing centralized power supply. Without prior system research and resource allocation strategy design in typical scenarios, it will be difficult to achieve precise matching of resources and demand, affecting the effective integration and operational efficiency of new power systems. This could even lead to resource redundancy and low coordination efficiency, thus hindering the green and efficient development of regional energy systems.

[0006] Existing technologies struggle to simultaneously meet the multiple requirements of system economy, power supply reliability, and operational flexibility. There is an urgent need to develop a systematic collaborative control method and optimized configuration scheme for scenarios involving ammonia-hydrogen power supply integration into the distribution network. Through scientific planning and forward-looking technological research, not only can the rationality of the distribution network structure be improved, but the safe and stable operation of the power supply system can also be guaranteed. This is of great significance for guiding the implementation of new energy technologies, accelerating the construction of a multi-energy integrated energy system, and improving urban and rural energy security and regional competitiveness. Summary of the Invention

[0007] To address the shortcomings and deficiencies of existing technologies, this invention provides a method and system for the coordinated control and optimized configuration of ammonia-hydrogen power sources integrated into the power distribution network. The method first constructs a mathematical model of the ammonia-hydrogen power system covering the entire energy conversion process from electricity to hydrogen to ammonia to hydrogen to electricity. In this model, short-cycle rapid response energy storage is achieved through hydrogen storage devices, while long-cycle large-scale energy storage is achieved through ammonia storage devices, forming a complementary dual-energy storage architecture, and incorporating the uncertainty constraint of hydrogen production. Second, a multi-timescale hierarchical coordinated control architecture matching the response characteristics of each component of the ammonia-hydrogen system is constructed. The response speeds, from slowest to fastest, correspond to energy storage scheduling, co-firing regulation, and disturbance mitigation levels, respectively, achieving coordinated scheduling of the ammonia-hydrogen power source and distributed resources in the power distribution network. Third, a distributed iterative solution framework based on boundary consistency constraints is adopted, combined with power flow constraints in the power distribution network, to complete parallel solutions in multiple regions. Finally, a two-layer optimized scheduling model coupled with wind, solar, thermal, and ammonia power sources is constructed, and the coordinated optimized configuration scheme of the ammonia-hydrogen power source and distributed resources is output through iterative output of the upper and lower layers. This invention effectively solves the technical problems of difficult modeling, difficult control coordination, and unreasonable configuration when ammonia-hydrogen power supply is connected to the power distribution network by organically combining full-process modeling, hierarchical collaborative control, and dual-layer optimized configuration.

[0008] The specific technical solution adopted by this invention to solve its technical problem is as follows:

[0009] To address the technical challenges of integrating ammonia-hydrogen power sources into distribution networks, such as inaccurate modeling, mismatch between control strategies and system characteristics, poor privacy in multi-regional collaborative scheduling, and lack of scenario adaptability in resource allocation, this invention provides a collaborative control and optimized configuration method and system for integrating ammonia-hydrogen power sources into distribution networks. Through a combination of refined modeling throughout the entire process, multi-timescale collaborative control, distributed privacy-preserving solutions, and scenario-based dual-layer optimization, this invention achieves efficient collaboration between ammonia-hydrogen power sources and distributed resources in the distribution network, comprehensively improving the economy, reliability, and low-carbon operation of the distribution network.

[0010] In a first aspect, the present invention provides a method for coordinated control and optimized configuration of ammonia-hydrogen power supply connected to a power distribution network:

[0011] First, a mathematical model and operational constraints for an ammonia-hydrogen power system covering the entire energy conversion process from electricity to hydrogen to ammonia to hydrogen to electricity are constructed. This model uses hydrogen storage devices for short-cycle, rapid-response energy storage and ammonia storage devices for long-cycle, large-scale energy storage, forming a complementary dual-energy storage architecture, and includes operational constraints related to uncertainties in hydrogen production. Next, a multi-timescale hierarchical collaborative control architecture matching the response characteristics of each stage of the ammonia-hydrogen system is constructed, with energy storage scheduling, co-firing regulation, and disturbance mitigation levels corresponding to response speeds from slow to fast, achieving collaborative scheduling of the ammonia-hydrogen power source and distributed resources in the distribution network. Then, a distributed iterative solution framework based on boundary consistency constraints is adopted, combined with power flow constraints in the distribution network, to complete parallel solutions in multiple regions. Finally, a two-layer optimized scheduling model coupling wind, solar, thermal, and ammonia is constructed, outputting a collaborative optimized configuration scheme for the ammonia-hydrogen power source and distributed resources through iterative output of the upper and lower layers. Finally, the collaborative optimized configuration scheme is output.

[0012] The dual-energy storage complementary architecture fully leverages the advantages of both types of energy storage media: hydrogen storage is suitable for short-term power regulation needs, while ammonia storage is suitable for long-term energy transfer needs. The two complement each other, balancing response speed and energy storage scale. The hierarchical collaborative control architecture divides the control levels according to the response time constants of different components of the ammonia-hydrogen system. The upper-level scheduling output serves as the constraint boundary for the lower-level control, forming a top-down multi-scale collaborative system. This avoids the problem that single-time-scale control cannot adapt to the response differences of multiple components. Boundary consistency constraints are used to ensure that the electrical variables of adjacent regions remain consistent at the common connection boundary, enabling the distributed solution results to have global consistency. At the same time, each region only needs to interact with boundary variables without uploading the original operating data, balancing privacy protection and solution efficiency.

[0013] Furthermore, to address the modeling accuracy requirements of the co-firing power generation stage and the need to characterize the uncertainties in the hydrogen production process, the hydrogen power generation stage and the ammonia power generation stage of the ammonia-hydrogen power system both employ a co-firing ratio coefficient to decouple the energy contributions of the base fuel and the alternative fuel. This accurately distinguishes the energy proportion and material consumption of different fuels, ensuring the accuracy of energy balance calculations and fundamentally avoiding the fundamental problem of attributing all total power generation to alternative fuels in traditional modeling. The hydrogen production uncertainty operation constraint is implemented using a three-layer structure, which includes baseline conversion efficiency calculation, catalyst dynamic decay correction, and characterization of random fluctuations in hydrogen production. This comprehensively covers the main deterministic and uncertain influencing factors in the ammonia decomposition hydrogen production process, enabling a more accurate reflection of efficiency changes and random fluctuation characteristics in the ammonia decomposition hydrogen production process and improving the anti-interference capability of the scheduling scheme.

[0014] Furthermore, to adapt to the time scale characteristics of different control levels, in the multi-time-scale hierarchical collaborative control architecture, the energy storage scheduling level adopts a model predictive control method to continuously optimize the power output plan, adapting to the long-term scheduling needs of the slow-response energy storage link, and relying on medium- and long-term prediction results to smooth system power fluctuations and achieve global economic optimization; the co-firing regulation level adopts a reinforcement learning method to output adaptive decisions to cope with short- and medium-term power fluctuations and operating condition changes, and learns the optimal decision logic under complex uncertainties through the interaction and iteration between the agent and the environment; the disturbance smoothing level adopts a robust control method to suppress power fluctuations, achieving rapid disturbance smoothing at the second level, and designs anti-interference control laws for known disturbances such as hydrogen production delay and fuel cell dynamic response lag, ensuring voltage and frequency stability without the need for an accurate disturbance model.

[0015] Furthermore, to balance the multiple optimization objectives in the reinforcement learning process, the reward function of the reinforcement learning method includes an operating cost term, a distribution network operating pressure penalty term, and an equipment adjustment smoothness penalty term. While optimizing the economic efficiency of system operation, it suppresses branch overload and voltage fluctuations based on branch power flow level and node voltage deviation, and avoids frequent and large-scale adjustments of equipment output based on the power changes of ammonia decomposition, gas-fired power generation with hydrogen, and coal-fired power generation with ammonia, thereby reducing equipment operating losses and extending equipment service life.

[0016] Furthermore, to improve the convergence performance and scalability of the multi-region distributed solution, the distributed iterative solution framework adopts the alternating direction multiplier method, constructing an augmented Lagrangian function containing a dual term and a quadratic penalty term. This decomposes the original high-dimensional global optimization problem into independent low-dimensional subproblems in each sub-region, and uses the original residual and dual residual as convergence criteria to complete the iterative solution. The original residual is used to measure the consistency between local variables and shared boundary variables in each region, while the dual residual is used to measure the optimal convergence of the iterative process. The combination of the two ensures the stability of the solution and the optimality of the result, and has the characteristics of fast convergence speed and strong stability, making it suitable for the distributed scheduling requirements of large-scale distribution networks.

[0017] Furthermore, to address the issues of disconnect between planning and operation in traditional solutions, and the difficulty of achieving both global optimization and solution efficiency in single-dimensional optimization, the upper layer of the proposed two-layer optimization scheduling model aims to maximize the system's coordinated operation indicators. These indicators comprehensively reflect multiple dimensions such as renewable energy consumption levels, power supply reliability, and low carbon emissions. The upper layer optimizes the net output power of wind and solar power and the total output of coal-fired units as the power boundary for the lower layer optimization. The lower layer aims to minimize the all-time operating cost. Within the given power boundary, it coordinates the material consumption and equipment operating costs of the ammonia-hydrogen system. Through iterative interaction between the upper and lower layers, the globally optimal configuration and scheduling scheme is obtained.

[0018] Furthermore, to adapt to the continuous operation characteristics of chemical production units, the ammonia-hydrogen power supply system mathematical model adopts a quasi-steady-state regulation model for the ammonia synthesis stage. The first-order inertial transient process is used to characterize the dynamic characteristics of the operating condition switching, reflecting the transition time constant and response lag of the ammonia synthesis unit when the operating conditions change. This avoids the problem of unexecutable instructions caused by the ideal step model. In addition, the adjustable range of operating conditions and the ramp rate constraint are set to ensure that the scheduling instructions meet the safe operation requirements of chemical production, making the model more in line with the actual engineering operation law.

[0019] Furthermore, to adapt to the differentiated needs of different power distribution network scenarios, differentiated configuration strategies are adopted for three types of power distribution network scenarios: rural, mountainous, urban, and island. In rural and mountainous scenarios, the focus is on improving the renewable energy absorption capacity, configuring ammonia synthesis and ammonia decomposition devices to achieve medium- and long-term energy transfer, and converting surplus wind and solar power into ammonia for storage and release on demand. In urban scenarios, the focus is on optimizing the interaction between ammonia-hydrogen equipment and the power grid, participating in peak shaving and valley filling in combination with the daily periodic characteristics of urban load, and improving the load regulation level. In island scenarios, the focus is on matching electrolytic hydrogen production and fuel cell capacity to ensure off-grid power supply, improving the reliability of independent power supply systems, and gradually reducing dependence on high-pollution diesel generator sets.

[0020] Secondly, this invention provides a collaborative control and optimized configuration system for ammonia-hydrogen power supply connected to a power distribution network:

[0021] The system comprises a model building module, a collaborative control module, an optimization configuration module, and an output module. The model building module constructs a mathematical model and operational constraints for an ammonia-hydrogen power system covering the entire energy conversion process from electricity to hydrogen to ammonia to hydrogen to electricity, forming a dual-energy storage complementary architecture combining short-cycle hydrogen storage and long-cycle ammonia storage, and generating operational constraints for uncertain hydrogen production. The collaborative control module constructs a multi-timescale hierarchical collaborative control architecture that matches the response characteristics of each stage of the ammonia-hydrogen system, and combines a distributed iterative solution framework to achieve collaborative scheduling of ammonia-hydrogen power sources and distributed resources in the power grid. The optimization configuration module constructs a two-layer optimized scheduling model coupling wind, solar, thermal, and ammonia, generating a collaborative optimized configuration scheme through iterative generation at the upper and lower layers. The output module outputs the collaborative optimized configuration scheme.

[0022] Thirdly, the present invention provides a computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, it implements the method for coordinated control and optimized configuration of ammonia-hydrogen power supply connected to the power distribution network as described in any of the above technical solutions.

[0023] Compared with the prior art, the present invention and its preferred embodiments have at least the following beneficial effects:

[0024] First, by using a dual energy storage complementary architecture of "hydrogen storage-ammonia storage" and refined full-process modeling, especially for the energy decoupling and hydrogen production uncertainty characterization in the co-firing process, the physical modeling of the ammonia-hydrogen coupling system is more in line with the actual engineering operation characteristics, which significantly improves the accuracy and robustness of subsequent optimization and control results.

[0025] Secondly, the proposed multi-timescale hierarchical collaborative control architecture, which matches the response characteristics of the ammonia-hydrogen system, effectively solves the problem of timescale mismatch between the large inertia and slow response characteristics of the ammonia-hydrogen system and the rapid fluctuations of renewable energy and load. It achieves efficient collaboration under all operating conditions and significantly improves the dynamic response performance and anti-interference capability of the system.

[0026] Furthermore, the distributed solution framework based on boundary consistency constraints takes into account both the data privacy protection requirements of multi-regional distribution networks and the solution efficiency of large-scale collaborative optimization problems, making it more suitable for the current decentralized and market-oriented development of distribution networks.

[0027] Finally, through a two-layer optimized scheduling model that couples wind, solar, thermal, and ammonia energy and a scenario-based adaptation strategy, multiple optimization goals of high-proportion renewable energy consumption, power supply reliability, and operational economy are achieved. This model can effectively adapt to the energy development needs of different scenarios such as cities, rural areas, and islands, and has broad engineering application prospects. Attached Figure Description

[0028] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments:

[0029] Figure 1 This is a flowchart illustrating the method for coordinated control and optimized configuration of ammonia-hydrogen power supply connected to the power distribution network according to an embodiment of the present invention.

[0030] Figure 2 This is a structural block diagram of the ammonia-hydrogen power supply access to the power distribution network collaborative control and optimization configuration system according to an embodiment of the present invention;

[0031] Figure 3 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention;

[0032] Figure 4 This is a schematic diagram of the ammonia-hydrogen power supply connected to the power distribution network according to an embodiment of the present invention;

[0033] Figure 5 This is a schematic diagram of the structure of a two-layer optimization model for a wind-solar-thermal-ammonia coupled integrated energy system according to an embodiment of the present invention. Detailed Implementation

[0034] To make the features and advantages of the present invention more apparent and understandable, specific embodiments are described below in detail:

[0035] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of this application. Unless otherwise specified, all technical and scientific terms used in this specification have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0036] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this application. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0037] This invention proposes a collaborative control and optimization configuration method and system for ammonia-hydrogen power supply integration into the distribution network. Addressing the challenges and operational risks of distribution networks with high proportions of renewable energy integration, it constructs a complete technical system around the entire energy conversion and storage process of "electricity-hydrogen-ammonia-hydrogen-electricity," encompassing refined modeling, multi-scenario adaptation, hierarchical collaborative control, and multi-objective optimization configuration. Simultaneously, it incorporates power quality collaborative regulation capabilities based on federated learning, achieving a comprehensive improvement in the safety, reliability, and economy of the distribution network.

[0038] At the system modeling level, static and dynamic mathematical models covering the entire process of hydrogen production, hydrogen storage, hydrogen power generation, ammonia synthesis, ammonia storage, ammonia decomposition, and ammonia power generation are established, clarifying the power-material conversion relationship, upper and lower operating limits, and ramp-up constraints for each stage. For gas-fired power generation with hydrogen blending and coal-fired power generation with ammonia blending, a blending ratio coefficient is used to decouple the energy contributions of the base fuel and alternative fuels, establishing the corresponding relationships between the consumption rates of natural gas / coal and hydrogen / ammonia and the total power generation, which conforms to energy conservation and engineering practice. Dynamic material balance equations are used for hydrogen and ammonia storage, with effective storage ranges of 10%~100% and 0%~100%, respectively. A quasi-steady-state adjustment model is added to the ammonia synthesis stage, using a first-order inertial transient process to characterize the dynamic characteristics of operating condition switching, while setting adjustable operating ranges and ramp-up rate constraints to adapt to the operational requirements of continuous chemical production.

[0039] To address the uncertainties in system operation, two models were established: a renewable energy fluctuation model and a hydrogen production fluctuation model. The renewable energy model, based on the thermal balance mechanism of photovoltaic modules, characterizes the dynamic characteristics of battery operating temperature, reflecting the impact of light intensity and ambient temperature on output. The hydrogen production model constructs a three-layer structure: baseline conversion efficiency, catalyst dynamic decay, and stochastic fluctuation process. The baseline efficiency, combined with the Arrhenius equation, reflects the influence of reaction temperature and catalyst activity on conversion efficiency. The relative catalyst activity decays exponentially over time, and its range is limited by an interval projection operator. The stochastic fluctuation of the hydrogen production rate is characterized by a first-order autoregressive process, including normally distributed independent perturbations, time correlation coefficients, and upper and lower limits of fluctuation, comprehensively reflecting the dynamic uncertainties of ammonia decomposition for hydrogen production.

[0040] At the collaborative control level, a hierarchical, multi-timescale collaborative control architecture is constructed, integrating three methods: model predictive control, reinforcement learning control, and robust control. Model predictive control is used for regional long-term scheduling, relying on predictive models of load, renewable energy, and hydrogen production to continuously optimize the output plans of each resource, smoothing medium- and long-term fluctuations and achieving global economic optimization. Reinforcement learning is used for minute-level regional coordination, defining a state space that includes distribution network node voltage, branch power flow, load level, energy storage status, and ammonia-hydrogen equipment operating conditions; an action space that includes ammonia decomposition power, hydrogen- and ammonia-blended power generation, energy storage charging and discharging, demand response adjustment, and reactive power support; and a reward function that considers operating costs, grid operating pressure, and equipment adjustment smoothness. Through interactive iterative learning between the agent and the environment, adaptive decision-making under complex uncertainties is achieved. Robust control is used for equipment-level second-level regulation, designing anti-interference control laws to quickly smooth short-term power fluctuations and ensure voltage and frequency stability, addressing disturbances such as hydrogen production delays and fuel cell dynamic response lags.

[0041] To address the distributed solution requirements of multi-regional distribution networks, a distributed solution framework based on the standard alternating direction multiplier method is adopted. The power flow constraints of the distribution network are handled by linearizing the DistFlow equations, transforming the nonlinear power flow problem into a linear form that can be solved in a distributed manner. A standard augmented Lagrangian function containing dual and quadratic penalty terms is constructed, and constraint indicator functions are introduced to ensure that variables satisfy the feasible region boundary. Iterative iteration is performed using a tree-parallel architecture of root node-branch node-leaf node, with nodes interacting only with their adjacent nodes. The original residual and dual residual are used as convergence criteria, improving the solution efficiency and convergence stability of large-scale systems while protecting data privacy.

[0042] At the level of scenario adaptation and optimized configuration, the core characteristics of ammonia-hydrogen power sources—miniaturization, modularity, and strong site adaptability—are identified. Differentiated distributed resource combination schemes and collaborative control strategies are designed for three typical power distribution network scenarios: rural / mountainous areas, cities, and islands. In rural / mountainous areas, the focus is on enhancing renewable energy absorption capacity by configuring ammonia synthesis and decomposition devices to achieve medium- and long-term energy transfer. In urban areas, the interaction between ammonia-hydrogen equipment and the grid is optimized based on load characteristics to improve load regulation capabilities. In island scenarios, system independence and off-grid power supply reliability are strengthened by matching electrolytic hydrogen production and fuel cell capacity to ensure continuous power supply. Based on this, a two-layer optimized scheduling model coupling wind, solar, thermal, and ammonia is constructed. The upper layer aims to maximize system coordinated operation indicators, outputting net wind and solar power output and total coal-fired unit output as the power boundary for the lower layer optimization. The lower layer aims to minimize overall operating costs, coordinating the material consumption and operating costs of the electro-ammonia conversion unit to achieve global optimized configuration of multiple energy systems.

[0043] At the power quality support level, a four-in-one control system is constructed, integrating adaptive federated learning training, voltage limit detection, ammonia-hydrogen power supply regulation capacity calculation, and distributed resource collaborative matching. The power quality assessment model is updated through a federated learning model that combines multi-regional local training and global parameter aggregation. Each region only uploads model parameters, without disclosing raw operational data. Real-time monitoring of voltage amplitude, phase, fluctuation frequency and harmonics, and three-phase imbalance at each node of the distribution network quickly identifies abnormal states such as voltage limits exceeding limits, sudden rises and falls, and distortion. The active power regulation margin, reactive power support capacity, regulation response speed, and continuous regulation duration of the ammonia-hydrogen power supply are dynamically calculated to quantify its voltage smoothing capability. Finally, coordinated control commands are generated by linking distributed photovoltaic, wind power, and energy storage resources within the distribution network. These commands are executed rapidly through a hierarchical collaborative control architecture to smooth voltage anomalies and ensure the power quality of the distribution network meets standards and operates safely and stably.

[0044] This invention, through refined modeling and multi-dimensional collaborative optimization throughout the entire process, fully leverages the complementary advantages of short-cycle rapid response hydrogen storage and long-cycle large-scale ammonia energy storage. It can significantly improve the level of renewable energy consumption, reduce system operating costs and losses caused by frequent equipment adjustments, enhance the reliability and operational stability of the power distribution network, and adapt to the power distribution network transformation needs of different regions and scenarios.

[0045] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of 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, 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.

[0046] like Figure 1 As shown, the process of implementing the collaborative control and configuration optimization method for connecting ammonia-hydrogen power supply to the power distribution network in this embodiment specifically includes the following steps:

[0047] Step S101: Establish a mathematical model of the "ammonia-hydrogen" power system, including a complete dynamic mathematical model of the ammonia-hydrogen power system. The "ammonia-hydrogen" power model includes full-process modeling of "electricity-hydrogen-ammonia-hydrogen-electricity", namely hydrogen production, hydrogen storage, hydrogen power generation, ammonia synthesis, ammonia storage, ammonia decomposition, and ammonia power generation, and analyze the operating characteristics of the "ammonia-hydrogen" power system.

[0048] In this step, the operational constraints of the hydrogen production stage in the dynamic mathematical model of the ammonia-hydrogen power supply system include the hydrogen production power expressed in the following formula. and hydrogen production rate Conversion relationship (comprehensive energy consumption for hydrogen production) The value is 56.00 (kW·h) / kg, and there are upper and lower limits for hydrogen production power.

[0049]

[0050] In the formula: and These represent the minimum and maximum load rates of the electrolytic cell, respectively. Let represent the installed capacity of the electrolytic cell; the subscript y, t represents the time t of the y-th planning year; given the planning period index set Y={1, 2, ..., Y}, with a step size of 1 year; for each planning period, the running time index set is T={1, 2, ..., N} (N=8760), with a time step size of... It takes 1 hour.

[0051] Operational constraints for hydrogen storage include the amount of hydrogen stored at different times, expressed by the following formula. and hydrogen production rate (hydrogen production from electrolyzer) Ammonia decomposition to produce hydrogen ) and the rate of hydrogen use (hydrogen used in ammonia synthesis) Hydrogen for gas turbine units Hydrogen for fuel cells The relationship between the hydrogen storage capacity and the upper and lower limits of hydrogen storage are considered. Due to the use of gas tanks for hydrogen storage, the hydrogen cannot be completely released due to the minimum pressure of the front and rear compressors. Therefore, the hydrogen storage range is set to 10%~100%.

[0052]

[0053] In the formula: and These represent the minimum and maximum load rates of the hydrogen storage tank, respectively. This refers to the installed capacity of the hydrogen storage tank.

[0054] The operational constraints of the hydrogen power generation process are achieved through hydrogen blending in gas turbines, as shown in the following formula, which represents the power output of the gas turbine unit using hydrogen blending. and the rate of hydrogen use The relationship between the two, and the hydrogen consumption coefficient corresponding to the equivalent contribution power per unit of hydrogen. The value is 0.04959 t / (MW·h); The proportion of power generated by hydrogen to the total power generated; This represents the power generation ratio corresponding to natural gas. The natural gas consumption rate coefficient corresponding to the electricity contribution per unit of natural gas.

[0055]

[0056] The operational constraints of the ammonia synthesis process include the ammonia production power expressed in equation (a). With ammonia production rate The conversion relationship, electricity consumption per ton of ammonia The value is taken as 645 (kW·h) / t; Equation (b) represents the adjustable range constraint of the ammonia synthesis operating conditions, which is usually 30%~110%; Equation (c) represents the ramp constraint, with a maximum ramp rate of 20% of the rated load per hour; Equation (d) represents the calculation relationship between the rated operating conditions and the capacity of the ammonia synthesis equipment, as well as the material balance relationship between the hydrogen consumption rate and the ammonia production rate, based on the mass conservation relationship of the ammonia synthesis reaction. The hydrogen-ammonia mass conversion coefficient represents the mass of ammonia corresponding to a unit mass of hydrogen, and is taken as 5.663×10-3t / kg (i.e. 5.663 kgammonia / kghydrogen). It is applicable to both the ammonia production calculation in ammonia synthesis and the ammonia consumption calculation in ammonia decomposition. Equation (e) represents the quasi-steady-state regulation model for ammonia synthesis.

[0057] (a)

[0058] (b)

[0059]

[0060] (d)

[0061] (e)

[0062]

[0063]

[0064] In the formula: and These represent the minimum and maximum loading rates for ammonia synthesis, respectively. The rated capacity of synthetic ammonia is usually expressed as the output of synthetic ammonia under 8000h rated operating conditions. This indicates the rated operating conditions for ammonia synthesis, expressed here as the hydrogen consumption rate under rated conditions; and These represent the maximum forward and reverse ramp rates for ammonia synthesis, respectively. This represents the switching process between different quasi-steady states in ammonia synthesis, where... The quasi-steady-state adjustment function is represented by a first-order transient process; This indicates the quasi-steady-state operating condition for ammonia synthesis; This is the set of indices for quasi-steady-state operating conditions. This refers to the period of quasi-steady-state switching in ammonia synthesis. This is the time constant for the ammonia synthesis unit to transition from one quasi-steady-state condition to another.

[0065] The constraints in ammonia storage include the amount of ammonia stored at different times, as expressed in Formula 1 below. and the rate of ammonia production from ammonia synthesis ammonia decomposition ammonia consumption rate Ammonia rate for coal-fired power units (It consists of two parts: ammonia doping power and ammonia doping start-up, represented as:) ,in For unit start-up capacity, The relationship is as follows: (the ammonia consumption coefficient for starting capacity is given by formula 2, which can be selected based on the actual start-up and shutdown characteristics of the unit). Formula 2 below represents the upper and lower limits of ammonia storage capacity. Since the ammonia storage tank stores ammonia in liquid form, the ammonia storage range is set to 0%~100%.

[0066]

[0067]

[0068] In the formula: and These represent the minimum and maximum load rates of the ammonia storage tank, respectively. This refers to the installed capacity of the ammonia storage tank.

[0069] Considering that the cost of storing ammonia is far lower than that of storing hydrogen, liquid ammonia can be used as a medium for long-term hydrogen storage, supplying hydrogen through the ammonia decomposition process. Its operational constraints include the ammonia decomposition power expressed in equation (a). and hydrogen production rate The conversion relationship, energy consumption coefficient The value is 755 (kW·h) / t; Equation (b) represents the upper and lower limits of the ammonia decomposition operating condition, typically ranging from 0% to 100%; Equation (c) represents the ammonia consumption rate during the ammonia decomposition process. With hydrogen production rate The material balance relationship.

[0070]

[0071]

[0072]

[0073] In the formula: and These represent the minimum and maximum loading rates for ammonia decomposition, respectively. This represents the rated capacity for ammonia decomposition, expressed here as the hydrogen production rate under rated operating conditions.

[0074] The power generation process employs ammonia-hydrogen co-combustion, as shown in the following formula, which represents the ammonia consumption rate in a coal-fired unit using ammonia-infused power generation. With power generation The relationship between them. This represents the proportion of ammonia-related power generation to total power generation. This represents the proportion of power generation corresponding to coal; The ammonia consumption coefficient corresponding to the electricity contributed per unit of ammonia; The coal consumption coefficient corresponding to the electricity contribution per unit of coal.

[0075]

[0076] In this embodiment, the key parameters of the ammonia-hydrogen power supply system are determined based on publicly available technical data, equipment manufacturer technical parameters, engineering experience values, on-site commissioning data, or historical operating data. Specifically, the comprehensive energy consumption for hydrogen production, the electricity consumption per ton of ammonia in synthetic ammonia, the operating condition adjustment range for synthetic ammonia, the ramp-up rate for synthetic ammonia, the energy consumption for ammonia decomposition, the hydrogen consumption coefficient for gas-fired power generation with hydrogen blending, the ammonia consumption coefficient for coal-fired power generation with ammonia blending, and the ammonia consumption coefficient for ammonia blending startup can be determined based on publicly available parameters or engineering experience values ​​of similar ammonia-hydrogen energy storage systems, thermal power generation systems with hydrogen-ammonia blending, and related equipment. The hydrogen-to-ammonia material conversion coefficient is derived from the mass conservation relationship of the synthetic ammonia reaction formula. The hydrogen storage range is determined based on the minimum safe pressure of the hydrogen storage tank, the compressor inlet pressure, and the safety margin. The ammonia storage range is determined based on the effective capacity of the liquid ammonia storage tank, the safe liquid level, and operational safety constraints.

[0077] The above parameters are the values ​​used for modeling and simulation verification in this embodiment. In actual engineering applications, they can be replaced, corrected or calibrated according to the specific equipment manufacturer's nameplate parameters, on-site measured data or historical operating data.

[0078] Step S102: Based on the characteristics of the "ammonia-hydrogen" power system, study typical distribution network scenarios suitable for "ammonia-hydrogen" power access, and propose differentiated distribution network distributed resource optimization combination schemes and collaborative control strategies.

[0079] In this step, the study combines the characteristics of the "ammonia-hydrogen" power system to investigate typical power distribution network scenarios applicable to "ammonia-hydrogen" power supply access, summarizes the core characteristics of "ammonia-hydrogen" power supply, and clarifies its basic advantages for power distribution network access, such as the three core features of ammonia-hydrogen power supply: miniaturization (containerized / distributed deployment, adaptable to distributed loads), modularity (capacity can be expanded on demand, reducing initial investment), and site adaptability (low dependence on terrain, climate, and infrastructure, can be deployed in remote / unpowered areas). At the same time, it clarifies its technical shortcomings, such as the dynamic response speed of ammonia cracking to produce hydrogen, the efficiency degradation characteristics of fuel cells, and the safety constraints of ammonia and hydrogen storage.

[0080] This study outlines the complete technology chain from "ammonia preparation → ammonia storage (liquid ammonia / solid ammonia) → ammonia cracking to hydrogen production → fuel cell power generation → grid-connected / off-grid power supply," clarifying key parameters such as efficiency, cost, and response time for each stage, and providing basic data for subsequent modeling.

[0081] This study analyzes the adaptability of typical systems and conducts a comprehensive survey of three core scenarios. Based on these scenarios, it constructs distributed resource optimization schemes adapted to regional differences: Rural / Mountainous Areas: Focus on enhancing the absorption capacity of renewable energy sources such as photovoltaic and wind power, and configure ammonia synthesis and decomposition devices to achieve medium- and long-term energy transfer; Island Areas: Emphasize system independence and stability, and strengthen the matching of electrolytic hydrogen production and fuel cell capacity to ensure reliable power supply under off-grid operation; Urban Areas: Combine electric vehicle charging and discharging behavior to optimize the interaction between ammonia-hydrogen equipment and the existing power grid, and improve load regulation capacity and environmental friendliness. A multi-objective optimization model is constructed considering operating costs, energy efficiency, carbon emissions, and response speed, and uncertainties such as load forecasting deviations and hydrogen production fluctuations are introduced to enhance the robustness of resource allocation.

[0082] (1) Resource endowment in rural mountainous areas, installed capacity, output characteristics and annual utilization hours of renewable energy such as distributed photovoltaic, small hydropower and small wind power; energy demand: temporal distribution and seasonal fluctuation characteristics of agricultural production load (irrigation, processing) and residential load; current status of power grid: distribution network structure, weak links and existing energy storage / generator configuration.

[0083] (2) Resource endowment of urban load center scenarios, distribution and capacity of distributed photovoltaic and commercial building energy storage; daily / weekly periodicity and peak-valley difference characteristics of industrial and commercial load and residential load in energy demand, and high requirements for power supply reliability and power quality; current status of power grid: high-density distribution network structure, load fluctuation characteristics, demand response mechanism, and backup power configuration.

[0084] (3) The island scenario has abundant resources, high volatility and intermittency of sea wind and photovoltaic power, and the characteristics of an isolated power grid without large power grid coverage; the load structure of island residents' life, tourism and fishery production in energy demand has the ultimate requirements for the reliability of independent power supply; the current power grid is an isolated microgrid mainly composed of diesel generators, with insufficient energy storage configuration, high power supply cost and high carbon emissions.

[0085] Assessment Dimensions Construction: An adaptability evaluation system is established from four dimensions: technical feasibility, economic feasibility, environmental feasibility, and operational feasibility. Adaptability Conclusion Output: Combining the characteristics of energy production and consumption in various regions, the access priority and optimal deployment mode (such as off-grid / grid-connected, centralized / distributed) of ammonia-hydrogen power sources are clarified for different scenarios, providing scenario boundaries for subsequent modeling and optimization.

[0086] Design of a Hierarchical Cooperative Control Strategy for the Ammonia-Hydrogen System: This phase focuses on the cooperative control mechanism between the ammonia-hydrogen system and other distributed energy sources. Considering the delay in ammonia-to-hydrogen production and the response characteristics of fuel cells, a hierarchical (regional / equipment level) and graded (temporal / spatial) cooperative control architecture is constructed. Advanced control methods such as model predictive control and reinforcement learning are integrated to achieve intelligent response and dynamic scheduling of the system to load changes, thereby improving the economy, flexibility, and stability of the power grid operation.

[0087] Mathematical modeling and computational analysis of typical scenarios are first performed by modeling the uncertainty, characterizing the intraday / seasonal fluctuations of wind and solar power output, quantifying the probability distribution of power output uncertainty, and modeling the fluctuations of renewable energy. Combining ammonia cracking reaction kinetics, feedstock supply stability, and equipment efficiency decay characteristics, a stochastic fluctuation model of ammonia production is established to analyze the impact of factors such as temperature, pressure, and catalyst activity on hydrogen production efficiency and hydrogen production fluctuations.

[0088] The fluctuation model for renewable energy is as follows:

[0089]

[0090] In the formula: This refers to the actual operating temperature. Ambient temperature; The nominal operating temperature; The ambient reference temperature corresponding to the NOCT test; This represents the actual light intensity at the current moment. The light intensity under NOCT test conditions; This represents the maximum power generation efficiency of the component under standard conditions. Temperature coefficient of component efficiency; is the thermal time constant.

[0091] Hydrogen production fluctuations are modeled as follows:

[0092]

[0093] In the formula: The baseline conversion efficiency for the time period; The conversion efficiency is for reference operating conditions; The relative activity of the catalyst over a given time period; It is the apparent activation energy; The reaction temperature; This is the reference reaction temperature.

[0094] This is the catalyst deactivation rate coefficient; The length of the scheduling period; This is the interval projection operator. The time correlation coefficient; The intensity of the fluctuation; For independent random perturbations; and These are the upper and lower limits of the fluctuation. This represents the relative fluctuation over a given period. The baseline hydrogen production rate for the ammonia decomposition unit during the specified time period; The mass flow rate of ammonia entering the ammonia decomposition unit during a given time period; The theoretical mass conversion coefficient for the conversion of ammonia to hydrogen; To account for the actual conversion efficiency considering temperature and catalyst activity.

[0095] As a further preferred implementation, the aforementioned hydrogen production uncertainty model is used to generate multiple sets of typical fluctuation scenarios, which are then used as constraint inputs to the model predictive control, two-layer optimization scheduling, and robust control links. This ensures that the scheduling scheme fully considers the impact of random fluctuations in hydrogen production, thereby improving the anti-disturbance capability and robustness in actual operation.

[0096] The multi-method fusion control strategy is based on system operating characteristics, integrating advanced methods such as model predictive control, reinforcement learning control, and robust control to form an adaptive collaborative control strategy:

[0097] Model Predictive Control (MPC): Suitable for long-term regional scheduling, it relies on predictive models of load, renewable energy, and hydrogen production to continuously optimize the output plans of each resource, smooth out medium- and long-term fluctuations, and achieve optimal global economic efficiency. Reinforcement Learning Control: Suitable for minute-level regional coordination, it learns optimal decisions under complex uncertainties through interactive iteration between the agent and the environment, coordinating the output allocation of ammonia-hydrogen power sources, energy storage, and distributed energy, and improving the system's adaptability to disturbances. Robust Control: Suitable for equipment-level, second-level control, it designs anti-interference control laws (such as H∞ control, μ synthesis, etc.) to quickly smooth short-term power fluctuations, ensure voltage / frequency stability, and enhance the system's anti-interference capability, addressing disturbances such as hydrogen production delays and fuel cell dynamic response lags.

[0098] (1) The hierarchical collaborative control solution algorithm for the "ammonia-hydrogen" system divides the multi-region distribution network into sub-problems based on the independence and collaboration of the "ammonia-hydrogen" system in each region. After the variable update and coordination are completed locally, the central coordinator iteratively summarizes the results, thereby improving the scalability and convergence efficiency of the large-scale system.

[0099] Core objective function: The objective is to minimize the operating cost over the entire time period (t=1 to T). The overall objective function is:

[0100]

[0101] In the formula, It is the overall objective function; It is the grid interaction cost, which is the net value of the cost of electricity purchased by producers and consumers from the grid and the revenue from selling electricity; It refers to the operating cost of energy storage, namely the maintenance cost of charging and discharging energy storage. It is the demand response cost, which is the cost of compensating for load deviations from baseline forecasts. and yes The on-grid electricity price and the off-grid electricity price for different time periods; and area Power purchased from the grid and power sold to the grid; It is the operation and maintenance and depreciation cost coefficient per unit of electricity of the energy storage system; and These are the energy storage charging power and discharging power; It is the penalty coefficient for demand response; It is the actual load after participating in demand response; This is the baseline load.

[0102] (2) Power flow linearization and optimization problem transformation: The power flow is approximated by linearization. In order to solve the nonlinear constraints of the power flow in the distribution network, the linearized DistFlow equation is adopted to transform the power flow constraints into a linear form that can be solved in a distributed manner, as shown in the following equation:

[0103]

[0104] In the formula, and It is a node Always Total active and reactive power injected during the time period; and It is a node Power injection from distributed power sources; and It is a node Flow to branch The active and reactive power of the branch circuit.

[0105] The constrained optimization problem is transformed into an augmented Lagrangian function. By introducing a quadratic penalty term and an indicator function, a distributed decomposition of constraints is achieved:

[0106]

[0107] In the formula For the region At the border The local boundary variables maintained above include tie line active power, reactive power, and boundary node voltage; For each adjacent region to the boundary Shared variables that are jointly maintained; These are the dual variables corresponding to the consistency constraints; The ADMM penalty parameter is positive.

[0108] in To constrain the indicator function, ensure that the variables satisfy the feasible region boundary conditions. As shown in the following formula:

[0109]

[0110] (3) The distributed parallel solution mechanism is a tree-based distributed architecture: a tree topology of root node-branch node-leaf node is adopted, each node corresponds to a producer / consumer / sub-region, the root node coordinates the whole, the branch node transmits the iteration signal, the leaf node performs local optimization, and the nodes only need to interact with the adjacent nodes to achieve privacy protection and parallel computing.

[0111] The proximal operator is defined as follows:

[0112]

[0113] In the formula, For set indicator functions; Equivalent to the Euclidean projection operator. Used to smooth local updates while preserving information about the objective function, ensuring iterative stability.

[0114] The European projection is as follows:

[0115]

[0116] In the feasible set Finding variables in The closest point in Euclidean distance ; This is the standard projection symbol.

[0117] Distributed parallel solution mechanism based on augmented Lagrange function Perform iterative solutions. In the next iteration, each regional node makes decisions based on its local regional decision variables. Shared variables between adjacent regions and dual variables This completes the local optimization of shared variable updates and dual variable updates.

[0118] To determine whether the distributed solution has converged, based on the boundary consistency constraint of the standard augmented Lagrangian function, the original residual and dual residual of the l-th iteration are defined as follows:

[0119] Definition of the first The original residual of the next iteration is:

[0120]

[0121] The standard dual residual is:

[0122]

[0123] in Representing regional decision variables Shared variables with neighboring regions Consistency deviation between them This indicates the degree of change of the shared variable between two consecutive iterations. Distributed solution is considered to have converged when the following condition is met:

[0124]

[0125]

[0126] in, and These are the original residual threshold and the dual residual threshold, respectively.

[0127] At the same time, to avoid excessive iterations, a maximum number of iterations is set. When the above convergence condition is met, or the number of iterations reaches... The iteration terminates when the time is right, and the decision variables for each region are output. This is the result of coordinated control of ammonia-hydrogen power sources, energy storage, and distributed resources.

[0128] (4) In this embodiment, reinforcement learning control is used for minute-level coordinated adjustment after the ammonia-hydrogen power supply is connected to the distribution network. To ensure that the reinforcement learning control corresponds to the aforementioned ammonia-hydrogen power supply model and distribution network power flow model, the state space of the reinforcement learning agent is defined as follows:

[0129]

[0130] in , and This refers to the voltage at distribution network nodes and the power flow status of branches. and The baseline load and the actual load after demand response. and For energy storage charging and discharging power, and For hydrogen storage and ammonia storage, , , , and These represent the hydrogen production, ammonia decomposition, hydrogen blending in gas-fired power plants, ammonia blending in coal-fired power plants, and hydrogen consumption states in ammonia synthesis.

[0131] Define the action space of a reinforcement learning agent as follows:

[0132]

[0133] in , and These are control commands for ammonia decomposition power, gas-fired power generation with hydrogen blending, and coal-fired power generation with ammonia blending, respectively. and For energy storage charging and discharging control commands, For the load after demand response adjustment, This refers to the reactive power support of the node.

[0134] The reward function of the reinforcement learning agent remains consistent with the original operating cost objective, and penalties are introduced for the operating pressure of the distribution network and the smoothness of the ammonia-hydrogen equipment regulation.

[0135]

[0136]

[0137]

[0138] in Used to characterize the operational pressure on the distribution network caused by changes in branch power flow and node voltage. Used to characterize hydrogen-ammonia fuel consumption and ammonia decomposition, and the adjustment range of gas-fired power plants and coal-fired power plants with hydrogen and ammonia blending. and The weighting coefficients are used. Through this reward function, the reinforcement learning agent can reduce grid interaction costs, energy storage operation costs, and demand response costs while suppressing branch power flow pressure, voltage fluctuations, fuel consumption, and frequent equipment adjustments.

[0139] As a preferred implementation of this embodiment, the Proximal Policy Optimization (PPO) algorithm is used to implement the aforementioned reinforcement learning collaborative control. This algorithm has the advantages of strong training stability and high sample utilization in power system dispatching scenarios with continuous action space, and can adapt to the collaborative optimization requirements of multi-dimensional states and continuous actions in this embodiment. The core hyperparameters of the PPO algorithm are configured as follows: the learning rate of both the policy network and the value network is set to 3×10^-4, the discount factor γ is set to 0.99 to balance immediate rewards and long-term operating benefits, the generalized advantage estimation parameter λ is set to 0.95, the policy pruning coefficient ε is set to 0.2, the batch size is set to 64, the number of network update rounds per iteration is set to 10, and the maximum number of training steps is set to 1×10^6. Both the policy network and the value network adopt a three-layer fully connected neural network architecture, with the number of hidden layer neurons being 256, 128, and 64 respectively. The ReLU function is used for the hidden layer activation function, and the Tanh function is used for the action output layer to map the control quantity to the rated adjustment range of the corresponding device. The training process employs an experience replay mechanism, with the replay buffer capacity set to 1×10^5. In each iteration, a batch of data is randomly sampled from the buffer to update the network parameters, thereby breaking the temporal correlation of samples and improving the convergence stability of the algorithm.

[0140] Step S103: By load forecasting and power output analysis, the load power demand and power characteristics are clarified respectively. Considering the power demand and multiple power sources including "ammonia-hydrogen" power, and taking into account various economic and reliability indicators, a multi-objective optimization configuration method for "ammonia-hydrogen" power system and distributed resources is proposed.

[0141] In this step, a multi-objective optimization method for the "ammonia-hydrogen" power system and distributed resource optimization is derived. This involves constructing a two-layer optimization scheduling model for a wind-solar-thermal-ammonia coupled integrated energy system. Based on wind power / solar / electric load power prediction, the model aims to maximize system coordinated operation indicators (while considering wind and solar power absorption, reliability, and low carbon emissions). Constrained by the output of wind / solar / coal-fired units and the balance of electrical power, the net output power of wind and solar power and the total output power of coal-fired units are output as the power boundary for the lower-level optimization.

[0142] The operating cost of thermal power is as follows:

[0143]

[0144] In the formula, It is an indicator of the stability of the output of coal-fired power units throughout the day; This is the total number of time periods; It is the first Actual power generation of coal-fired units during the specified time period; It is the average power generation of coal-fired power units throughout the day; It is the average daily utilization rate of coal-fired power units; This is the maximum technical output of the coal-fired power unit; This is the total daily operating cost of a coal-fired power unit; It is the procurement cost per unit mass of coal; It is the first The quality of coal consumed by coal-fired power units during a given period; yes The operation and maintenance costs of coal-fired power units in the first period; This is the total daily operating cost of the electro-ammonia conversion unit; It is the water consumption cost of the electro-ammonia conversion process; The operation and maintenance costs of electro-ammonia conversion equipment; It is the cost per unit mass of water; It is the first The mass of water consumed in the electro-ammonia conversion reaction during a given period.

[0145] The constraints are as follows:

[0146]

[0147]

[0148] In the formula, yes Net grid-connected wind power in the first time period; It is the first Forecasted wind power output for the specified time period; It is the first Net grid-connected photovoltaic power during the specified time period; It is the first Forecasted photovoltaic power output for the specified time period; and These are the minimum and maximum technical outputs of coal-fired power units; It is the first The wind power absorbed during a given period; This is the maximum permitted proportion of wind power consumption; It is the first The photovoltaic power absorbed during the specified time period; This is the maximum permitted proportion of photovoltaic power generation.

[0149] Step S104: Construct a mathematical model of the "ammonia-hydrogen" power source, a mathematical model of the distributed power source, and a mathematical model of the load. Verify whether the proposed method can meet the power supply requirements under various typical scenarios through power balance analysis. Optimize the proposed method using mathematical models to obtain the optimal method for the coordinated optimization of the "ammonia-hydrogen" power system and the distributed resources of the distribution network.

[0150] In this step, the optimal method for coordinated optimization of the "ammonia-hydrogen" power system and the distributed resources of the distribution network is described, wherein the hydrogen-ammonia-hydrogen system is as follows:

[0151]

[0152] for The mass of hydrogen produced by ammonia decomposition during the electro-ammonia conversion process. Energy conversion efficiency for ammonia-to-hydrogen production. The mass coefficient of hydrogen gas that can be obtained from the decomposition of a unit mass of ammonia. for The mass of ammonia used for hydrogen production during the specified time period. The comprehensive specific power consumption of the entire electro-ammonia conversion process encompasses the total power consumption of the water electrolysis hydrogen production and ammonia synthesis sections. It is used to characterize the power consumption level of the entire chain for producing a unit mass of ammonia, and is different from the power consumption coefficient per ton of ammonia, which only reflects the energy consumption of a single ammonia synthesis section. . The duration of the i-th scheduling period is expressed in hours (h). for The mass of nitrogen gas consumed during the ammonia synthesis process over a period of time. for Wind power used for PSA nitrogen production during certain periods for Photovoltaic power used for PSA nitrogen production during certain periods The actual specific power consumption required for a synthetic ammonia unit to produce a unit mass of nitrogen.

[0153] Step S105: Based on actual conditions, test the proposed methods and strategies, and verify through mathematical models whether the proposed methods can meet the power supply requirements under various typical scenarios. If they do not meet the requirements, re-optimize. Ensure comprehensive optimization of the distribution network in terms of safety, reliability, and economy.

[0154] As a further preferred implementation, this method also includes a power quality collaborative regulation step: training a power quality assessment model based on a federated learning model, monitoring the voltage and power quality indicators of distribution network nodes in real time, dynamically calculating the active and reactive power regulation capabilities of the ammonia-hydrogen power supply, linking distributed resources to generate collaborative regulation commands, suppressing voltage overruns and power quality anomalies, and ensuring the stable operation of the distribution network.

[0155] In summary, this application presents a method and system for coordinated control and optimized configuration of ammonia-hydrogen power sources in distribution networks. It establishes a complete static and dynamic mathematical model of the ammonia-hydrogen power system. The ammonia-hydrogen power model includes full-process modeling of "electricity-hydrogen-ammonia-hydrogen-electricity," encompassing hydrogen production, hydrogen storage, hydrogen power generation, ammonia synthesis, ammonia storage, ammonia decomposition, and ammonia power generation. The operating characteristics of the ammonia-hydrogen power system are analyzed. Based on the characteristics of the ammonia-hydrogen power system, typical distribution network scenarios suitable for ammonia-hydrogen power source integration are proposed, and differentiated distributed resource optimization combination schemes and coordinated control strategies for distribution networks are studied. By clarifying load power demand through load forecasting and power source characteristics based on power output analysis, and considering power demand and various power sources including ammonia-hydrogen, a multi-objective optimized configuration method for the ammonia-hydrogen power system and distributed resources is proposed, comprehensively considering various economic and reliability indicators. Mathematical models of ammonia-hydrogen power sources, distributed power sources, and loads are constructed. Power balance analysis is used to verify whether the proposed method can meet the power supply requirements under various typical scenarios. The mathematical models are then used to optimize the proposed method, ultimately obtaining the optimal method for the coordinated optimization of ammonia-hydrogen power systems and distributed resources in the distribution network.

[0156] Please see Figure 2 The diagram shows a structural block diagram of a collaborative control and optimization configuration system for ammonia-hydrogen power supply connected to a power distribution network, as described in this application.

[0157] like Figure 2 As shown, the ammonia-hydrogen power supply access to the distribution network collaborative control and optimization configuration system 200 includes an adaptive federated learning training module 210, a voltage over-limit sensing module 220, an ammonia-hydrogen power supply regulation capacity calculation module 230, and a distributed resource collaborative matching module 240.

[0158] The adaptive federated learning training module 210 is configured to collect distributed data on multi-regional power distribution network operation, all operational parameters of the ammonia-hydrogen power supply, and distributed power output data. It trains the evaluation model locally in each region. The evaluation model includes an input layer, a feature extraction layer, a fusion layer, and an output layer. The input layer receives regional power distribution network operation data and ammonia-hydrogen power supply operation data; the feature extraction layer extracts features of node voltage changes, branch power flow changes, and ammonia-hydrogen power supply output changes; the fusion layer merges the power distribution network operation status and the ammonia-hydrogen power supply operation status; and the output layer outputs the voltage exceedance risk level, active power regulation margin, reactive power support capacity, regulation response speed, and continuous regulation duration. Only model parameters are uploaded to global nodes for federated iterative aggregation. Without disclosing the original operational data, the adaptive updating and optimization of the power quality regulation capability evaluation model are achieved, accurately adapting to the dynamic operating characteristics of ammonia-hydrogen power supply connected to the power distribution network. The training process includes: global model initialization, local model training, model parameter uploading, global model aggregation, global model distribution, and iterative updates; training is terminated when the model validation error is less than a set threshold or the training rounds reach the maximum number of communication rounds.

[0159] As a preferred implementation of this embodiment, the FedAvg algorithm is used as the global model aggregation algorithm. This algorithm has low computational complexity and low communication overhead, and can adapt to the communication constraints and data privacy protection requirements of distributed training in multi-regional power distribution networks. During global aggregation, the local model parameters are weighted and averaged according to the proportion of local training sample size in each region to the total sample size to obtain the updated global model parameters. To address the task characteristics of this embodiment, where the input is multivariate time-series operational data and the output is multi-dimensional power quality assessment indicators, the local assessment model employs a Long Short-Term Memory (LSTM) neural network structure. This structure effectively captures the time-series dependency features of distribution network operation data and ammonia-hydrogen power supply operation data. Specifically, the network structure is as follows: the number of neurons in the input layer matches the dimension of the state features, corresponding to input features such as node voltage, branch power flow, hydrogen storage, ammonia storage, and equipment output; the hidden layer contains two LSTM units, each with 128 neurons, used to extract deep time-series features; subsequently, two fully connected layers are connected, with 64 and 32 neurons respectively, using the ReLU activation function; the number of neurons in the output layer matches the dimension of the assessment indicators, corresponding to output results such as voltage exceedance risk level, active power regulation margin, and reactive power support capacity. During local model training, the batch size is set to 32, the learning rate is set to 1×10^-3, the optimizer uses the Adam algorithm, the number of local training rounds within a single communication round is set to 5, the global maximum number of communication rounds is set to 50, and the model validation error threshold is set to 0.01.

[0160] The voltage over-limit sensing module 220 is configured to monitor in real time the voltage amplitude, phase, fluctuation frequency and harmonics, three-phase imbalance and other power quality indicators of each grid-connected node in the distribution network. It focuses on collecting electrical operation data from the ammonia-hydrogen power supply grid connection point, quickly identifying abnormal states such as voltage over-limit, voltage spikes and drops, and voltage distortion, accurately locating weak nodes in the distribution network, determining the over-limit level and duration, and outputting standardized voltage anomaly characteristic data. The ammonia-hydrogen power supply regulation capacity calculation module 230 is configured to dynamically calculate the active power regulation margin, reactive power support capacity, regulation response speed and continuous regulation duration of the ammonia-hydrogen power supply based on the operating constraints, load rate and output characteristics of each component of the ammonia-hydrogen power supply. It quantitatively evaluates the ammonia-hydrogen power supply's ability to mitigate voltage over-limits in the distribution network and its potential to support power quality, outputting the real-time adjustable output range and regulation constraints. The distributed resource collaborative matching module 240 is configured to use the voltage over-limit sensing results and the ammonia-hydrogen power supply regulation capacity calculation data to link distributed resources such as distributed photovoltaic, wind power, and energy storage in the distribution network, coordinate and optimize the output allocation, regulation timing and response strategies of various resources, generate collaborative control instructions adapted to voltage over-limit management, and execute them quickly through a hierarchical collaborative control architecture to smooth out voltage anomalies in the distribution network and ensure that the power quality of the distribution network meets the standards and that it operates safely and stably after the ammonia-hydrogen power supply is connected.

[0161] It should be understood that Figure 2 The modules and references described in the document Figure 1 The steps described in the text correspond to those in the method described above. Therefore, the operations, features, and corresponding technical effects described above also apply to the method described in the text. Figure 2 The various modules in the document will not be described in detail here.

[0162] In other embodiments, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor performs the ammonia-hydrogen power supply access to the distribution network coordinated control and optimization configuration method in any of the above method embodiments.

[0163] In one embodiment, the computer-readable storage medium of the present invention stores computer-executable instructions, which are configured as follows:

[0164] A mathematical model of the ammonia-hydrogen power system was established, and a complete dynamic mathematical model of the ammonia-hydrogen power system was built. The ammonia-hydrogen power model includes the modeling of the entire process of "electricity-hydrogen-ammonia-hydrogen-electricity", and the operating characteristics of the ammonia-hydrogen power system were analyzed.

[0165] Based on the characteristics of the ammonia-hydrogen power system, this study investigates typical distribution network scenarios suitable for ammonia-hydrogen power access, and explores differentiated distributed resource optimization schemes and collaborative control strategies for distribution networks.

[0166] By clarifying the load power demand through load forecasting and the power output characteristics through power source analysis, and considering the power demand and various power sources including "ammonia-hydrogen" power sources, and comprehensively considering various economic and reliability indicators, a method for optimizing the allocation of "ammonia-hydrogen" power systems and distributed resources is derived through multi-objective optimization.

[0167] Mathematical models of ammonia-hydrogen power sources, distributed power sources, and loads are constructed. Power balance analysis is used to verify whether the proposed method can meet the power supply requirements under various typical scenarios. The mathematical models are then used to optimize the proposed method, ultimately obtaining the optimal method for the coordinated optimization of ammonia-hydrogen power systems and distributed resources in the distribution network.

[0168] Based on actual conditions, the proposed methods and strategies are tested, and mathematical models are used to verify whether the proposed methods can meet the power supply requirements under various typical scenarios. If they do not meet the requirements, they are re-optimized. This ensures comprehensive optimization of the distribution network in terms of safety, reliability, and economy.

[0169] Computer-readable storage media may include a stored program area and a stored data area, wherein the stored program area may store an operating system and an application program required for at least one function; the stored data area may store data created based on the use of the ammonia-hydrogen power supply access to the distribution network coordinated control and optimization configuration system. Furthermore, the computer-readable storage medium may include high-speed random access memory, and may also include memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some embodiments, the computer-readable storage medium may optionally include memory remotely configured relative to a processor, which can be connected to the ammonia-hydrogen power supply access to the distribution network coordinated control and optimization configuration system via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0170] Figure 3 This is a schematic diagram of the structure of the electronic device provided in the embodiment of the present invention, such as... Figure 3 As shown, the device includes a processor 310 and a memory 320. The electronic device may also include an input device 330 and an output device 340. The processor 310, memory 320, input device 330, and output device 340 can be connected via a bus or other means. Figure 3 Taking a bus connection as an example, the memory 320 is the computer-readable storage medium described above. The processor 310 executes various server functions and data processing by running non-volatile software programs, instructions, and modules stored in the memory 320, thereby realizing the ammonia-hydrogen power supply access to the power distribution network collaborative control and optimization configuration method described in the above embodiment. The input device 330 can receive input digital or character information. The output device 340 may include a display screen or other display device.

[0171] The aforementioned electronic device can execute the method provided in the embodiments of the present invention, and has the corresponding functional modules and beneficial effects for executing the method. Technical details not described in detail in this embodiment can be found in the method provided in the embodiments of the present invention.

[0172] In one implementation, the aforementioned electronic device is applied in an ammonia-hydrogen power supply access distribution network coordinated control and optimization configuration system, used as a client, and includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to:

[0173] A mathematical model of the ammonia-hydrogen power system was established, and a complete dynamic mathematical model of the ammonia-hydrogen power system was built. The ammonia-hydrogen power model includes the modeling of the entire process of "electricity-hydrogen-ammonia-hydrogen-electricity", and the operating characteristics of the ammonia-hydrogen power system were analyzed.

[0174] Figure 4 This is a diagram of the ammonia-hydrogen power supply connection to the power distribution network provided in an embodiment of the present invention, as shown below. Figure 4 As shown, based on the characteristics of the "ammonia-hydrogen" power system, this paper studies typical distribution network scenarios suitable for "ammonia-hydrogen" power access, and investigates differentiated distributed resource optimization combination schemes and collaborative control strategies for distribution networks.

[0175] By clarifying the load power demand through load forecasting and the power output characteristics through power source analysis, and considering the power demand and various power sources including "ammonia-hydrogen" power sources, and comprehensively considering various economic and reliability indicators, a method for optimizing the allocation of "ammonia-hydrogen" power systems and distributed resources is derived through multi-objective optimization.

[0176] Figure 5 This is a schematic diagram of the two-layer optimization model structure of the wind-solar-thermal-ammonia coupled integrated energy system of the present invention, provided in an embodiment of the present invention. Figure 5 As shown, a mathematical model of "ammonia-hydrogen" power source, a mathematical model of distributed power source, and a mathematical model of load are constructed. The proposed method can be verified by power balance analysis to meet the power supply requirements under various typical scenarios. The proposed method is optimized using mathematical models, and finally the optimal method for the coordinated optimization configuration of "ammonia-hydrogen" power system and distributed resources of distribution network is obtained.

[0177] Based on actual conditions, the proposed methods and strategies are tested, and mathematical models are used to verify whether the proposed methods can meet the power supply requirements under various typical scenarios. If they do not meet the requirements, they are re-optimized. This ensures comprehensive optimization of the distribution network in terms of safety, reliability, and economy.

[0178] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of embodiments.

[0179] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for coordinated control and optimized configuration of ammonia-hydrogen power supply connected to a power distribution network, characterized in that, Includes the following steps: A mathematical model and operational constraints for an ammonia-hydrogen power system covering the entire energy conversion process from electricity to hydrogen to ammonia to hydrogen to electricity are constructed. The model uses hydrogen storage devices as short-cycle fast-response energy storage and ammonia storage devices as long-cycle large-scale energy storage, forming a dual-energy storage complementary architecture, and includes operational constraints on the uncertainty of hydrogen production. A multi-time-scale hierarchical collaborative control architecture matching the response characteristics of each link in the ammonia-hydrogen system is constructed. The energy storage scheduling, co-firing regulation, and disturbance suppression levels are respectively arranged according to the response speed from slow to fast, so as to realize the collaborative scheduling of ammonia-hydrogen power supply and distributed resources of the distribution network. A distributed iterative solution framework based on boundary consistency constraints is adopted, and multi-region parallel solution is completed by combining distribution network power flow constraints; A two-layer optimization scheduling model coupling wind-solar-thermal-ammonia is constructed, and a collaborative optimization configuration scheme of ammonia-hydrogen power source and distributed resources is output through iterative output of upper and lower layers; Output the aforementioned collaborative optimization configuration scheme.

2. The method for coordinated control and optimized configuration of ammonia-hydrogen power supply connected to the power distribution network according to claim 1, characterized in that: The hydrogen power generation and ammonia power generation stages of the ammonia-hydrogen power system both employ a blending ratio coefficient to decouple the energy contributions of the base fuel and alternative fuels. The uncertainty constraint on hydrogen production is implemented using a three-layer structure, which includes, in turn, the calculation of the baseline conversion efficiency, the correction of catalyst dynamic decay, and the characterization of random fluctuations in hydrogen production.

3. The method for coordinated control and optimized configuration of ammonia-hydrogen power supply connected to the power distribution network according to claim 1, characterized in that: In the multi-timescale hierarchical collaborative control architecture, the energy storage scheduling level adopts a model predictive control method to continuously optimize the power output plan, the co-firing regulation level adopts a reinforcement learning method to output adaptive decisions, and the disturbance smoothing level adopts a robust control method to suppress power fluctuations.

4. The method for coordinated control and optimized configuration of ammonia-hydrogen power supply connected to the power distribution network according to claim 3, characterized in that: The reward function of the reinforcement learning method includes an operating cost term, a power distribution network operating pressure penalty term, and an equipment adjustment smoothness penalty term.

5. The method for coordinated control and optimized configuration of ammonia-hydrogen power supply connected to the power distribution network according to claim 1, characterized in that: The distributed iterative solution framework employs the alternating direction multiplier method to construct an augmented Lagrangian function containing a dual term and a quadratic penalty term, and uses the original residual and the dual residual as convergence criteria to complete the iterative solution.

6. The method for coordinated control and optimized configuration of ammonia-hydrogen power supply connected to the power distribution network according to claim 1, characterized in that: The upper layer of the dual-layer optimization scheduling model aims to maximize the coordinated operation indicators of the system, and outputs the net output power of wind and solar power and the total output of coal-fired units as the power boundary for the lower layer optimization; the lower layer aims to minimize the operating cost throughout the entire period, and coordinates the material consumption and equipment operating cost of the ammonia-hydrogen system.

7. The method for coordinated control and optimized configuration of ammonia-hydrogen power supply connected to the power distribution network according to claim 1, characterized in that: In the mathematical model of the ammonia-hydrogen power supply system, the ammonia synthesis stage adopts a quasi-steady-state regulation model, which uses a first-order inertial transient process to characterize the dynamic characteristics of the operating condition switching, and sets constraints on the adjustable range of the operating condition and the ramp rate constraint.

8. The method for coordinated control and optimized configuration of ammonia-hydrogen power supply connected to the power distribution network according to claim 1, characterized in that: Differentiated configuration strategies are adopted for three types of power distribution network scenarios: rural, mountainous, urban, and island. In rural and mountainous scenarios, ammonia synthesis and ammonia decomposition devices are configured to achieve medium- and long-term energy transfer. In urban scenarios, the interaction capability between ammonia hydrogen equipment and the power grid is optimized. In island scenarios, electrolysis hydrogen production and fuel cell capacity are matched to ensure off-grid power supply.

9. A collaborative control and optimized configuration system for ammonia-hydrogen power supply connected to a power distribution network, characterized in that, include: The model building module is used to construct a mathematical model and operational constraints for an ammonia-hydrogen power system that covers the entire energy conversion process from electricity to hydrogen to ammonia to hydrogen to electricity, forming a dual-energy storage complementary architecture that combines short-cycle hydrogen storage with long-cycle ammonia storage, and generating operational constraints for uncertain hydrogen production. The collaborative control module is used to construct a multi-time-scale hierarchical collaborative control architecture that matches the response characteristics of each link in the ammonia-hydrogen system, and to realize the collaborative scheduling of ammonia-hydrogen power supply and distributed resources of the distribution network by combining a distributed iterative solution framework. The optimization configuration module is used to construct a two-layer optimization scheduling model that couples wind, solar, thermal and ammonia, and generates a collaborative optimization configuration scheme through iterative generation of upper and lower layers; The output module is used to output the collaborative optimization configuration scheme.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the collaborative control and optimized configuration method for connecting ammonia-hydrogen power supply to the power distribution network as described in any one of claims 1 to 8.

Images