Dual-threshold value regulation based multi-stack fuel cell cogeneration system optimization method and system and storage medium

Abstract

This application discloses an optimization method, system, and storage medium for a multi-stack fuel cell cogeneration system based on dual-threshold control, belonging to the field of fuel cell control technology. The method includes: establishing an optimal power mapping relationship between the number of fuel cell stacks and the heat recovery state; starting and stopping the heat recovery loop based on a dual-threshold liquid level; dynamically determining power input and output thresholds according to real-time thermal state and the number of fuel cell stacks, forming a hysteresis interval to match the load; deciding on the addition or reduction of fuel cell stacks based on dual-threshold power, and introducing a minimum duration constraint to suppress frequent switching; establishing a power safety boundary, and forcibly protecting against overload or high-potential risks by exceeding time constraints; and finally, coordinating the execution of heat and electricity control commands. This application effectively reduces the start-up and shutdown frequency of fuel cell stacks through dynamic dual-threshold coupled thermoelectric management, improving the overall energy efficiency, economy, and operational safety of the system.

Description

Technical Field

[0001] This application relates to the field of fuel cell system control technology, specifically to an optimization method, system, and storage medium for multi-stack fuel cell cogeneration systems based on dual-threshold regulation. Background Technology

[0002] Hydrogen energy, as a clean and efficient secondary energy carrier, has become an important direction for energy transformation. Proton exchange membrane fuel cells (PEMFCs) are considered promising power devices for transportation, distributed generation, and backup power due to their advantages such as rapid low-temperature start-up, high efficiency, and zero emissions. However, in megawatt-level high-power applications such as industrial power supply, district power supply, and large ship propulsion, the limited power density, cost, and reliability of a single PEMFC make parallel operation of multiple PEMFCs an inevitable choice to meet high power demands.

[0003] One of the core control challenges of multi-stack fuel cell systems lies in matching the number of operating stacks with the load power, i.e., "stack control." Traditional strategies mainly include power-sharing and sequential start-stop strategies. Power-sharing distributes the load evenly across all online stacks, resulting in uniform power distribution and high system stability under moderate loads. However, under low load conditions, all stacks operate in the high-potential range, accelerating the electrochemical corrosion of the catalyst carbon support and bipolar plates. Sequential start-stop strategies start and stop stacks sequentially according to load size, allowing some stacks to operate at their high-efficiency points or shut down, improving low-load efficiency. However, under medium to high loads, this may cause some stacks to operate at full load for extended periods while others remain idle, failing to fully utilize the power regulation potential of multiple stacks, leaving room for improvement in overall energy efficiency.

[0004] To integrate the advantages of the above strategies, existing technologies have proposed some adaptive methods. For example, by calculating the optimal system load power for different numbers of fuel cell stacks, a set of discrete load power thresholds is set, each threshold directly corresponding to a specific optimal number of fuel cell stacks. When the system load exceeds a certain threshold, a corresponding switch in the number of fuel cell stacks is triggered, and a power-sharing strategy is used to allocate the load among the currently operational stacks. This method achieves a certain degree of efficiency trade-off. However, in actual operation, especially when applied to combined heat and power systems, this type of single-threshold-based control method reveals significant drawbacks:

[0005] First, the problem of frequent start-ups and shutdowns of fuel cell stacks is exacerbated. User-side electrical and thermal loads often fluctuate, and even small load fluctuations near thresholds can easily cause the fuel cell stack to repeatedly switch between "on" and "off" states. Frequent start-ups and shutdowns not only generate additional energy losses but also accelerate the mechanical and chemical degradation of core components such as proton exchange membranes and catalysts due to the drastic changes in temperature, humidity, and potential accompanying each start-up and shutdown. This severely shortens the fuel cell stack's lifespan and increases system maintenance costs.

[0006] Secondly, there is insufficient coordination between thermoelectricity and fuel cells. Fuel cells generate a significant amount of waste heat while generating electricity. Combined heat and power (CHP) systems can utilize this heat through heat recovery devices (such as waste heat boilers and absorption chillers), significantly improving overall energy efficiency. However, most existing fuel cell stack control strategies focus only on power tracking and efficiency, failing to deeply couple with the state of the heat recovery system (such as the water level in the storage tank and user heat demand). The start and stop of heat recovery alter the optimal operating point of the fuel cell stack, and changes in the number of operating stacks directly affect heat generation. If fuel cell stack control and thermal management lack coordination, it is difficult to achieve global optimization of the system's electrical and thermal energy output.

[0007] Secondly, there is a lack of effective protection over the safe operating boundaries of the fuel cell stack. The durability of a PEMFC stack is closely related to its operating conditions. Prolonged operation at excessively high power (overload) will exacerbate voltage decay, while operation at excessively low power (underload) will accelerate cathode catalyst corrosion due to increased operating potential. Existing stack control strategies are mostly aimed at achieving optimal efficiency. Under drastic load changes or prolonged deviations from typical operating conditions, the stack may operate in the aforementioned dangerous ranges. There is a lack of rapid and effective protective intervention mechanisms, threatening the long-term operational safety of the system.

[0008] Finally, the robustness of the strategy needs to be strengthened. Single threshold control is sensitive to load fluctuations, while a simple fixed dual threshold, although capable of creating hysteresis, has a static threshold range that cannot adapt to changes in heat recovery status, ambient temperature, and stack health status. This results in poor adaptability of the control strategy under different operating conditions, and the overall efficiency cannot be maintained at a high level consistently.

[0009] Therefore, there is an urgent need in this field for an innovative optimization method for multi-stack fuel cell cogeneration systems that can fundamentally suppress frequent start-ups and shutdowns of the stacks, deeply couple thermoelectric management, embed safety protection logic, and possess dynamic adaptive capabilities, thereby comprehensively improving the system's economy, durability, and operational reliability. Summary of the Invention

[0010] To address the aforementioned deficiencies in existing technologies, this application aims to provide an optimization method for multi-stack fuel cell cogeneration systems based on dual-threshold control. This method constructs a dynamic power dual-threshold control mechanism coupled with heat recovery status, and combines minimum time constraints and safety over-limit protection to achieve intelligent coordination between stack switching and heat recovery start-up and shutdown. Ultimately, this achieves a comprehensive effect of reducing stack switching frequency, improving overall system energy efficiency, ensuring safe equipment operation, and extending system lifespan. To achieve the above objectives, the technical solution adopted in this application is as follows:

[0011] The optimization method for a multi-stack fuel cell cogeneration system based on dual-threshold control in this application includes the following steps:

[0012] Step S1: Based on the simulation or historical operation optimization results of the multi-stack fuel cell cogeneration system, establish the mapping relationship between the operating power setpoints corresponding to different combinations of the number of fuel cell stacks in operation and different heat recovery states of the hot water storage tank.

[0013] Step S2: Monitor the heat storage status parameters of the hot water storage tank in real time, compare them with the preset start liquid level threshold and stop liquid level threshold, and generate a start command or stop command for the heat recovery circuit of the hot water storage tank according to the comparison result, while updating the current heat recovery status identifier.

[0014] Step S3: Based on the current heat recovery status identifier updated in step S2 and the real-time number of fuel cell stacks, query the system's optimal operating power setpoint mapping relationship established in step S1, and dynamically determine the fuel cell stack input power threshold and fuel cell stack cut-out power threshold for fuel cell stack quantity control under the current operating conditions, wherein the fuel cell stack input power threshold is higher than the fuel cell stack cut-out power threshold, and a hysteresis interval is formed between the two.

[0015] Step S4: Collect real-time electrical load data from the user side and compare it with the stack input power threshold and stack cut-off power threshold dynamically determined in step S3 to generate control commands to increase or decrease the number of stacks;

[0016] Step S5: Set the minimum duration threshold for switching the number of fuel cells. When a control command to increase or decrease the number of fuel cells is received from step S4, determine whether the time interval between the current time and the last effective fuel cell number switching time is less than the minimum duration threshold. If the condition is met, block the control command and maintain the current number of fuel cells unchanged. If the condition is met, mark the control command as an instruction to be executed.

[0017] Step S6: Preset the upper and lower limits of the power safety under the current number of operating fuel cells, and monitor the real-time power load data on the user side in real time. If the power safety upper limit is exceeded or the power safety lower limit is lowered, a forced execution instruction is generated. The forced execution instruction is to perform the increase or decrease operation of the number of fuel cells corresponding to the current overload or underload risk.

[0018] Step S7: Execute the start or stop command from the heat recovery loop in step S2, and the stack quantity control command finally determined after the constraints in step S5 and the over-authorization judgment in step S6, to perform coordinated control of the multi-stack fuel cell cogeneration system.

[0019] In one embodiment, step S1 specifically includes:

[0020] Based on the simulation or historical operation optimization results of the multi-stack fuel cell cogeneration system, the optimal stack operating power setpoints corresponding to different stack operating numbers under the two states of starting heat recovery and stopping heat recovery are established.

[0021] The optimal operating power setpoint mapping relationship is stored in the form of a table or function. The input is the number of fuel cell stacks in operation and the heat recovery status identifier, and the output is the corresponding optimal total operating power value of the fuel cell stack.

[0022] The optimal total operating power value of the fuel cell stack is obtained through multi-objective optimization calculation with the goal of maximizing the overall benefits of the multi-stack fuel cell cogeneration system.

[0023] In one embodiment, step S2 specifically includes:

[0024] The liquid level and operating status of the hot water storage tank are monitored in real time, and the judgment is made by comparing the preset stop liquid level threshold and start liquid level threshold.

[0025] If the liquid level is higher than the preset stop liquid level threshold, a stop heat recovery command is generated and output to shut down the heat recovery circuit;

[0026] If the liquid level is lower than the preset start-up liquid level threshold, a start-up heat recovery command is generated and output to activate the heat recovery circuit.

[0027] In one embodiment, the dynamic determination of the stack input power threshold and the stack cut-out power threshold specifically includes:

[0028] Real-time monitoring of the heat storage status of the hot water storage tank and the real-time number of electric stacks in operation;

[0029] Using the current number of operating fuel cells and the current heat recovery status identifier as indexes, the corresponding optimal operating power value is retrieved from the mapping relationship;

[0030] Based on the optimal operating power value and combined with the preset hysteresis interval width, the stack input power threshold and stack cut-out power threshold are calculated.

[0031] In one embodiment, step S4 specifically includes:

[0032] Real-time electrical load data from the user side is collected and compared with the dynamically updated stack input power threshold and stack cut-out power threshold.

[0033] If the real-time electrical load on the user side continues to be higher than the power threshold of the fuel cell stack, a first control command is generated to increase the current number of fuel cell stacks.

[0034] If the real-time electrical load on the user side remains below the stack cut-off power threshold, a second control command is generated to reduce the current number of stacks.

[0035] If the real-time electrical load on the user side is between the stack input power threshold and the stack cut-out power threshold, a third control command is generated to maintain the current number of stacks.

[0036] In one embodiment, the minimum duration threshold for switching the number of fuel cells is set based on the physical time constant of the fuel cell startup / shutdown process and the impact model of frequent start-ups and shutdowns on fuel cell lifetime degradation, so as to ensure that the fuel cell has sufficient stable operating time after each start-up and shutdown.

[0037] In one embodiment, step S6 specifically includes:

[0038] When the real-time electrical load on the user side exceeds the upper limit of power safety or falls below the lower limit of power safety, a forced execution command is generated to adjust the number of fuel cells. The forced execution command is to perform an increase or decrease operation of the number of fuel cells corresponding to the current overload or underload risk.

[0039] The power safety limit is the maximum total power that all fuel cells can safely and continuously bear given the current number of operating fuel cells;

[0040] The power safety lower limit is the minimum total power allowed to avoid a single fuel cell stack operating in a high-potential corrosion risk zone, given the current number of operating fuel cell stacks.

[0041] In one embodiment, the coordinated control further includes:

[0042] When executing a control command to increase the number of fuel cell stacks, priority is given to starting fuel cell stacks that are currently in standby mode and have accumulated less operating time.

[0043] When executing the control command to reduce the number of fuel cell stacks, priority is given to shutting down fuel cell stacks that are currently in operation and have accumulated a large amount of operating time, so as to achieve balanced management among the fuel cell stacks.

[0044] In one embodiment, the multi-stack fuel cell cogeneration system includes multiple parallel proton exchange membrane fuel cell modules, a shared thermal storage module, and a user-side coupling interface; each proton exchange fuel cell module includes a stack, an air supply subsystem, a thermal management subsystem, and a power conversion unit; the heat recovery loop, through valve switching, can selectively introduce waste heat generated by the stack and auxiliary components into the thermal storage module or dissipate it to the environment.

[0045] Therefore, this application has the following beneficial effects:

[0046] This patent application aims to address the main shortcomings in the field of multi-stack fuel cell cogeneration: first, frequent start-ups and shutdowns of the stack caused by load fluctuations near a single threshold lead to accelerated equipment lifespan degradation; second, existing strategies suffer from disconnected electrical and thermal control, making it difficult to achieve optimal overall energy efficiency under dynamic operating conditions; and third, there is a lack of rapid safety protection mechanisms for dangerous conditions such as overload and low-load high-potential. This solution achieves the following triple optimization effects:

[0047] By introducing dynamic power dual thresholds coupled with thermal state and minimum time constraints, the number of fuel cell stack start-ups and shutdowns is reduced by approximately 46.8% in typical scenarios, effectively mitigating the degradation of core components.

[0048] By using heat recovery status as the core variable for power regulation thresholds, the system prioritizes tracking electrical load and utilizing waste heat when heat is needed, and operates in a higher electrical efficiency range when heat is not needed, thus achieving global energy optimization. Simulations show that the levelized cost of electricity (LCOE) is reduced by approximately 12.42%, and the overall efficiency is improved by approximately 4.34%.

[0049] A power safety boundary overriding mechanism with the highest priority is established to ensure that protective switching can be performed immediately under any extreme operating conditions, preventing equipment overload or high-potential corrosion, and fundamentally improving the operational safety and reliability of the system. Attached Figure Description

[0050] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0051] Figure 1 This is a flowchart of an optimization method for a multi-stack fuel cell cogeneration system based on dual-threshold regulation.

[0052] Figure 2 A schematic diagram of a multi-stack fuel cell cogeneration system based on multi-end recovery technology provided in this application embodiment;

[0053] Figure 3 This is a schematic diagram of an optimized control strategy for a multi-stack fuel cell cogeneration system based on dual thresholds, provided in an embodiment of this application.

[0054] Figure 4 A comparison of the number of fuel cell start-stop switching operations under dual-threshold and single-threshold control schemes provided in the embodiments of this application;

[0055] Figure 5 The results of dual-threshold stacking optimization and control are provided in a typical scenario according to the embodiments of this application. Detailed Implementation

[0056] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0057] It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of this application.

[0058] To address the shortcomings of existing technologies, this application provides an optimization method for multi-stack fuel cell cogeneration systems based on dual-threshold regulation. The core of this method lies in constructing a dynamic dual-threshold collaborative regulation mechanism that is deeply coupled with the heat recovery state.

[0059] First, during the offline phase, an optimal operating power mapping library is established based on simulation or historical data optimization, under different combinations of the number of operating fuel cell stacks and heat recovery status (on / off), providing a benchmark for online control. During online operation, this method executes two parallel closed loops: one is an independent heat recovery loop control based on dual liquid level thresholds (start threshold and stop threshold), which avoids frequent actuator movements through hysteresis logic and outputs the system thermal status indicator in real time; the other is fuel cell stack number optimization control based on dynamic power dual thresholds. These power dual thresholds are not fixed values, but rather, based on the real-time thermal status indicator and the current number of operating fuel cell stacks, the corresponding optimal power point is queried from the mapping library, and an input threshold and a cut-off threshold with a preset hysteresis width are dynamically generated around it, thereby achieving precise matching between the control strategy and the real-time thermoelectric coupling condition.

[0060] Based on this dynamic threshold, the system compares user electrical loads in real time and decides whether to increase or decrease the number of fuel cell stacks. To completely suppress frequent switching, a minimum duration constraint is introduced, forcing the fuel cell stacks to operate stably for a period of time after each state change. Simultaneously, a highest-priority power safety boundary protection mechanism is established. When the load exceeds the upper and lower limits of safe operation of the fuel cell stack, the system will override the time constraint and forcibly execute protective switching to ensure the intrinsic safety of the equipment. Finally, the controller coordinates the execution of thermal management commands and fuel cell stack control commands, and can integrate fuel cell stack lifetime balancing management strategies.

[0061] Through the aforementioned mechanisms, this method fundamentally addresses the shortcomings of existing technologies: the dynamic hysteresis interval and time constraints significantly reduce the stack switching frequency by approximately 46.8%; thermoelectric state coupling improves the overall system efficiency by approximately 4.34% and reduces the levelized cost of power generation by approximately 12.42%; and the built-in safety overriding mechanism greatly enhances the system's robustness. This method provides an effective optimization solution for achieving efficient, long-lasting, safe, and economical operation of multi-stack fuel cell cogeneration systems.

[0062] This application provides an optimization method for a multi-stack fuel cell cogeneration system based on dual-threshold control, including steps S1-S7, as described above. Figure 1 , Figure 1 This is a flowchart of an optimization method for a multi-stack fuel cell cogeneration system based on dual-threshold control. The specific steps are as follows:

[0063] Step S1: Based on the simulation or historical operation optimization results of the multi-stack fuel cell cogeneration system, establish the mapping relationship between the operating power setpoints corresponding to different combinations of the number of fuel cell stacks in operation and different heat recovery states of the hot water storage tank.

[0064] Step S2: Monitor the heat storage status parameters of the hot water storage tank in real time, compare them with the preset start liquid level threshold and stop liquid level threshold, and generate a start command or stop command for the heat recovery circuit of the hot water storage tank according to the comparison result, while updating the current heat recovery status identifier.

[0065] Step S3: Based on the current heat recovery status identifier updated in step S2 and the real-time number of fuel cell stacks, query the system's optimal operating power setpoint mapping relationship established in step S1, and dynamically determine the fuel cell stack input power threshold and fuel cell stack cut-out power threshold for fuel cell stack quantity control under the current operating conditions, wherein the fuel cell stack input power threshold is higher than the fuel cell stack cut-out power threshold, and a hysteresis interval is formed between the two.

[0066] Step S4: Collect real-time electrical load data from the user side and compare it with the stack input power threshold and stack cut-off power threshold dynamically determined in step S3 to generate control commands to increase or decrease the number of stacks;

[0067] Step S5: Set the minimum duration threshold for switching the number of fuel cells. When a control command to increase or decrease the number of fuel cells is received from step S4, determine whether the time interval between the current time and the last effective fuel cell number switching time is less than the minimum duration threshold. If the condition is met, block the control command and maintain the current number of fuel cells unchanged. If the condition is met, mark the control command as an instruction to be executed.

[0068] Step S6: Preset the upper and lower limits of the power safety under the current number of operating fuel cells, and monitor the real-time power load data on the user side in real time. If the power safety upper limit is exceeded or the power safety lower limit is lowered, a forced execution instruction is generated. The forced execution instruction is to perform the increase or decrease operation of the number of fuel cells corresponding to the current overload or underload risk.

[0069] Step S7: Execute the start or stop command from the heat recovery loop in step S2, and the stack quantity control command finally determined after the constraints in step S5 and the over-authorization judgment in step S6, to perform coordinated control of the multi-stack fuel cell cogeneration system.

[0070] Specifically, this embodiment provides an optimization method for a multi-stack fuel cell cogeneration system based on dual-threshold control, referring to... Figure 1 As shown, this method achieves intelligent coordination and optimization of stack switching and heat recovery start-up and shutdown in multi-stack fuel cell cogeneration systems by constructing a dynamic, coupled dual-threshold control mechanism. The specific implementation process includes the following steps S1 to S7:

[0071] Step S1: Preset the optimal power mapping relationship of the system

[0072] In this step, the basic database is constructed during the offline phase of the multi-stack fuel cell cogeneration system. Based on the high-precision simulation model or long-term historical operating data of the multi-stack fuel cell cogeneration system, multi-objective optimization calculations are performed with the goal of maximizing the overall system benefits (such as comprehensive energy efficiency, economics, carbon emission reduction, and equipment health status). For all possible combinations of "number of operating stacks" and "heat recovery status" (on or off), the corresponding optimal total power operating setpoint of the system is solved and determined. Subsequently, these correspondences are established as mapping tables or function models and pre-stored in the multi-stack fuel cell cogeneration system as the reference parameters for online real-time control.

[0073] Optimization calculations are performed on multi-stack fuel cell combined heat and power systems through high-precision simulation or analysis of their long-term historical operating data. A core parameter mapping table is established, which clearly defines the stack operating power setpoint P_opt obtained through multi-objective optimization under different numbers of operating stacks (e.g., 1, 2...N stacks) and two states where the heat recovery system is "on" or "off". This mapping table will be pre-stored in the system as a core database, providing a reference parameter basis for online real-time control.

[0074] Step S2: Regulation of the thermal storage system based on dual threshold liquid level

[0075] This step independently manages the heat recovery loop. The system monitors the heat storage status parameters (such as liquid level H) of the hot water tank in real time. A stop liquid level threshold H_high and a start liquid level threshold H_low are preset, with H_high > H_low, forming a hysteresis range for liquid level control. The system maintains a current heat recovery status identifier A (e.g., A > 0 indicates on, A ≤ 0 indicates off). The control logic is as follows: if the current heat recovery is on (A=1) and H > H_high is detected, a "stop heat recovery" command is generated, the heat recovery loop is closed, and A is updated to 0; if the current heat recovery is off (A=0) and H < H_low is detected, a "start heat recovery" command is generated, the heat recovery loop is opened, and A is updated to 1. This dual-threshold mechanism effectively avoids frequent actuator actions caused by small fluctuations in liquid level at a single threshold point.

[0076] Step S3: Dynamic generation of dual thresholds for power regulation

[0077] This step is the core link in connecting thermoelectric management. In each control cycle, the system, based on the status parameter A corresponding to the real-time heat recovery status identifier updated in step S2, and the real-time number of fuel cell stacks n obtained from system feedback, queries the preset mapping relationship in step S1 to obtain the optimal operating power setpoint P_opt under the current operating conditions. Based on this, and combined with the preset hysteresis interval width (which is set according to the typical load fluctuation amplitude and the minimum start-stop time constraint of the fuel cell stacks), it dynamically calculates two thresholds for fuel cell stack quantity control: the fuel cell stack input power threshold P_in and the fuel cell stack cut-off power threshold P_re, forming a dynamic power hysteresis interval. This interval adaptively adjusts with changes in heat recovery status and the number of fuel cell stacks, achieving precise matching between the control strategy and real-time operating conditions.

[0078] The core of dynamic determination lies in the following: when heat recovery is activated, the adjustable range of the system's thermoelectric output is wider, and the power thresholds P_in and P_re for the fuel cell stack can be adjusted accordingly to prioritize electrical load tracking and utilize waste heat; when heat recovery is deactivated, the power operating range needs to be tightened to ensure that the fuel cell stack operates in the high-efficiency zone. This achieves precise matching of thermoelectric decoupling requirements and effectively suppresses frequent start-stop cycles caused by load fluctuations.

[0079] Step S4: Stack Quantity Decision Based on Power Dual Thresholds

[0080] The system collects the user-side electrical load demand P_load in real time. P_load is compared with P_in and P_re, which are dynamically generated in step S3, and a preliminary decision instruction is generated: if P_load > P_in, an instruction to "increase the number of fuel cells" is generated; if P_load < P_re, an instruction to "decrease the number of fuel cells" is generated; if P_load is between P_re and P_in, an instruction to "maintain the current number of fuel cells" is generated. This step makes a logical judgment based on the instantaneous power relationship.

[0081] Step S5: Minimum Duration Constraint

[0082] To suppress frequent start-ups and shutdowns of fuel cell stacks caused by short-term large load fluctuations, the system sets a minimum duration threshold T_min (e.g., 60 minutes). When a command to add or remove fuel cell stacks is received from step S4, the system calculates the time interval Δt between the current moment and the moment when the number of fuel cell stacks actually changed. If Δt < T_min, the switching request is deemed too frequent and is blocked, maintaining the current number of fuel cell stacks. If Δt ≥ T_min, the command is allowed to be transmitted and marked as a "pending command". This constraint forces the fuel cell stacks to operate stably for a period of time after each state change, enhancing operational stability.

[0083] This constraint forces the fuel cell stack to operate stably for at least T_min after each start-up and shutdown, avoiding excessively frequent start-ups and shutdowns caused by large load fluctuations around the two thresholds.

[0084] Step S6: Unauthorized regulation of power safety boundary

[0085] This is the highest priority protection mechanism. The system presets power safety operating boundaries for each number of fuel cells: an upper safety power limit P_safe_max (overload protection) and a lower safety power limit P_safe_min (low-load high-potential corrosion protection). The system continuously monitors P_load and compares it with the safety boundaries corresponding to the number of currently operating fuel cells. If P_load ≥ P_safe_max or P_load ≤ P_safe_min, it means that the system faces the risk of overload or high-potential corrosion. At this time, the system will immediately generate a "forced execution command" (corresponding to adding or removing fuel cells). This command will directly bypass the minimum duration constraint of step S5, ensuring that equipment safety is given priority.

[0086] If the electrical load P_load is greater than or equal to the safe power limit under the current operating stack configuration, the system will immediately bypass the minimum duration constraint in step five and directly execute the "Stack Quantity Control Scheme" command. If the electrical load P_load is less than the high-efficiency power limit under the current operating stack configuration, the system will also immediately bypass the minimum duration constraint and directly execute the "Stack Quantity Control Scheme" command. This scheme ensures that under any circumstances, the operating power of the stack does not exceed the safe and high-efficiency range, thus guaranteeing equipment safety and lifespan.

[0087] Step S7: Cooperative Control Execution

[0088] In this final step, the system integrates and executes various commands: First, it executes the start / stop command for the heat recovery loop generated in step S2. Second, it arbitrates the commands for adjusting the number of fuel cell stacks: if a "forced execution command" generated in step S6 exists, it is executed first; otherwise, the "pending execution command" that passed the judgment in step S5 is executed. Finally, based on the finally determined commands for adjusting the number of fuel cell stacks, specific fuel cell stacks are selected for start / stop operations (which can be combined with a balancing strategy), and the power output of all online fuel cell stacks is adjusted to meet real-time electrical load requirements. Through step S7, the thermal management subsystem and the fuel cell stack power generation system are unified and coordinated, jointly achieving comprehensive optimized operation of the multi-stack fuel cell cogeneration system under the premise of safety, durability, and efficiency.

[0089] In one embodiment, step S1 specifically includes:

[0090] Based on the simulation or historical operation optimization results of the multi-stack fuel cell cogeneration system, the optimal stack operating power setpoints corresponding to different stack operating numbers under the two states of starting heat recovery and stopping heat recovery are established.

[0091] The optimal operating power setpoint mapping relationship is stored in the form of a table or function. The input is the number of fuel cell stacks in operation and the heat recovery status identifier, and the output is the corresponding optimal total operating power value of the fuel cell stack.

[0092] The optimal total operating power value of the fuel cell stack is obtained by multi-objective optimization calculation with the goal of maximizing the comprehensive benefits of the multi-stack fuel cell cogeneration system.

[0093] Specifically, in this embodiment, to construct a highly reliable benchmark parameter library, high-fidelity system modeling and simulation analysis of the multi-stack fuel cell cogeneration system are required, or in-depth optimization driven by its long-term historical operating data is necessary. The simulation model or data analysis model must accurately cover the electrochemical and thermodynamic characteristics of the fuel cell stack, the power consumption characteristics of auxiliary components (such as air compressors, coolant pumps, and DC / DC converters), the heat transfer efficiency of heat recovery modules (such as plate heat exchangers), the dynamic heat storage / release characteristics of the heat storage module (hot water tank), and the coupling relationships among the various components of the system. This is the foundation for obtaining accurate optimization results.

[0094] Secondly, the multi-objective optimization function of "system comprehensive benefit" is defined and quantified. This function is not a single efficiency index, but a comprehensive evaluation system integrating multiple dimensions such as economy, energy efficiency, carbon emission reduction, and equipment health status. Specifically, the optimization objective can be expressed as maximizing a weighted comprehensive benefit index, which is typically composed of sub-items such as power generation efficiency (directly related to hydrogen consumption rate), waste heat recovery efficiency, equipment depreciation cost converted to unit energy (closely linked to the life decay model related to the number of stack start-ups and shutdowns and operating conditions), system operation and maintenance costs, and environmental benefits from carbon emission reduction. By setting reasonable weighting coefficients, the multi-objective problem is transformed into a single-objective optimization problem or solved using Pareto front theory.

[0095] Then, for each specific operating condition combination—namely, the "number of fuel cell stacks in operation N" (N=1, 2, ..., N_max, where N_max is the total number of fuel cell stacks in the system) and the "heat recovery status identifier A" (A=0 represents "heat recovery off", A=1 represents "heat recovery on")—the above optimization algorithm is run. For each (N, A) combination, under the constraint of system safety operation, the algorithm searches for the key control variables that maximize the aforementioned comprehensive benefit indicators. Its core output is the optimal total fuel cell stack operating power setpoint P_opt(N, A) for the system under the current combination. This setpoint represents the optimal balance point for the overall system operation under this specific fuel cell stack configuration and thermal management mode.

[0096] Finally, all calculated mapping relationships (N, A) -> P_opt(N, A) are systematically stored. In specific implementations, it is preferable to store them in the system's non-volatile memory in the form of a two-dimensional lookup table. The horizontal axis index of this lookup table is the number of fuel cell stacks in operation N, and the vertical axis index is the heat recovery state A. The table content is the corresponding P_opt value. Alternatively, it can be fitted into a function model with N and A as input parameters based on the data characteristics. This mapping relationship is the core knowledge base for the system to achieve dynamic optimization operation, ensuring that the system can quickly look up the theoretically optimal power operating benchmark based on real-time operating conditions.

[0097] In one embodiment, step S2 specifically includes:

[0098] The liquid level and operating status of the hot water storage tank are monitored in real time, and the judgment is made by comparing the preset stop liquid level threshold and start liquid level threshold.

[0099] If the liquid level is higher than the preset stop liquid level threshold, a stop heat recovery command is generated and output to shut down the heat recovery circuit;

[0100] If the liquid level is lower than the preset start-up liquid level threshold, a start-up heat recovery command is generated and output to activate the heat recovery circuit.

[0101] Specifically, in this embodiment, the start-up and shutdown of the heat recovery loop are managed independently and stably based on the real-time status of the thermal storage system, and key thermal status inputs are provided for subsequent stack power regulation.

[0102] First, the system monitors the liquid level (denoted as H) in the hot water storage tank in real time using a level sensor installed on the tank, which serves as the main parameter characterizing the system's heat storage status. In other embodiments, the heat storage status parameter can also be the equivalent heat storage capacity calculated based on the temperature distribution and total volume within the tank. Simultaneously, the system maintains an internal heat recovery loop operation status flag to record whether the loop is currently in an "on" (open) or "off" (closed) state.

[0103] To achieve reliable control and prevent actuators (such as circulating pumps and three-way valves) from frequently operating near critical points due to signal fluctuations, this application presets two hysteresis-related liquid level thresholds: a stop liquid level threshold H_high and a start liquid level threshold H_low, satisfying that H_high > H_low. These two thresholds are set comprehensively based on the design volume of the hot water tank, the user's heat load characteristics, and the system's thermal inertia. For example, H_high can be set to 85% of the total capacity, and H_low can be set to 25% of the total capacity.

[0104] The specific comparison and instruction generation logic is as follows:

[0105] When the heat recovery loop is currently in "Running" (A = "ON"): The system continuously compares the real-time liquid level H with H_high. If H is detected to be higher than H_high, it is determined that the heat storage is sufficient, and continuing to recover waste heat may lead to heat energy waste or system overpressure. At this time, the system generates and outputs a "Stop Heat Recovery" command. This command is sent to the actuator, which will shut down the water pump in the heat recovery loop and switch the valve to the heat dissipation path. At the same time, the system updates its internal status flag to "OFF".

[0106] When the current state of the heat recovery loop is "Stop" (A = "OFF"): The system continuously compares the real-time liquid level H with H_low. If H is detected to be lower than H_low, it is determined that the heat storage is insufficient, and the recovery process needs to be started to supplement the heat source. At this time, the system generates and outputs a "Start Heat Recovery" command. This command will drive the actuator to start the water pump in the heat recovery loop and switch the valve to the heat storage passage. At the same time, the system updates the internal status flag to "ON".

[0107] By setting H_high > H_low, a level control hysteresis range is established between the two. As long as the level fluctuates within this range, the state of the heat recovery loop will remain unchanged. This design effectively filters out interference from level sensor noise or short-term small fluctuations, significantly reduces the operating frequency of pumps and valves, and improves equipment lifespan and system operational stability. The status indicator output in step S2, as a key parameter of the system's current thermal state, will be transmitted in real time to the subsequent step S3 for dynamic decision-making on the fuel cell power regulation threshold, thereby achieving synergy between thermal management and electrical power management.

[0108] In one embodiment, the dynamic determination of the stack input power threshold and the stack cut-out power threshold specifically includes:

[0109] Real-time monitoring of the heat storage status of the hot water storage tank and the real-time number of electric stacks in operation;

[0110] Using the current number of operating fuel cells and the current heat recovery status identifier as indexes, the corresponding optimal operating power value is retrieved from the mapping relationship;

[0111] Based on the optimal operating power value and combined with the preset hysteresis interval width, the stack input power threshold and stack cut-out power threshold are calculated.

[0112] Specifically, in this embodiment, this step is the core link connecting thermal management and electric power control, and realizing dynamic adaptive strategy. Its implementation is based on real-time parameters provided by the preceding steps: the status parameter corresponding to the current heat recovery status identifier from step S2 (denoted as A, for example, A greater than 0 indicates that the water tank is in heat recovery operation state, and A less than or equal to 0 indicates that the water tank is in heat recovery stop state) and the real-time number of fuel cell stacks in operation (denoted as n) directly from the system status feedback.

[0113] First, the system uses the current (n, A) combination as an exact index to query the "optimal operating power setpoint mapping relationship" pre-established and stored in step S1. This mapping relationship typically exists in the form of a two-dimensional lookup table or a function. Through the query, the system can obtain the optimal operating power value P_opt(n, A) that perfectly matches the current operating conditions. This value represents the theoretically optimal total output power setpoint from the perspective of overall system benefits, given that n fuel cell stacks are currently operating and the heat recovery state is A.

[0114] After obtaining P_opt(n, A), it is not directly used as a single switching threshold, but rather as a benchmark to construct a power control range with hysteresis characteristics. For this purpose, the system presets a key control parameter—the hysteresis range width ΔP. The setting of ΔP has clear engineering basis: First, it must be greater than the typical random fluctuation amplitude of the user-side electrical load over a short time scale (such as several minutes). For example, for a megawatt-level system, ΔP can be set to 100-200kW. Second, its size needs to be coordinated with the "minimum duration threshold" in step S5 to ensure that when the load changes at a normal rate, the time required to cross the hysteresis range is greater than the minimum duration constraint, thereby avoiding frequent switching from a mechanism perspective.

[0115] The resulting P_in and P_re satisfy P_in > P_re, forming dynamic "thresholds" at both ends of an interval centered on P_opt and with a width of ΔP. These thresholds are not fixed values ​​but adaptively adjust with changes in n and A. When heat recovery is enabled, P_opt is typically higher, causing the overall threshold interval to shift upwards, allowing the system to respond more proactively to electrical loads. When heat recovery is disabled, the interval shifts downwards, and the operating strategy becomes more conservative to ensure efficiency. This dynamic determination mechanism ensures that the control strategy is always precisely matched to the real-time thermoelectric coupling state, a key design feature for suppressing frequent start-stop cycles caused by load fluctuations while simultaneously improving the overall system performance.

[0116] In one embodiment, step S4 specifically includes:

[0117] Real-time electrical load data from the user side is collected and compared with the dynamically updated stack input power threshold and stack cut-out power threshold.

[0118] If the real-time electrical load on the user side continues to be higher than the power threshold of the fuel cell stack, a first control command is generated to increase the current number of fuel cell stacks.

[0119] If the real-time electrical load on the user side remains below the stack cut-out power threshold, a second control command is generated to reduce the current number of stacks.

[0120] If the real-time electrical load on the user side is between the stack input power threshold and the stack cut-out power threshold, a third control command is generated to maintain the current number of stacks.

[0121] Specifically, in this embodiment, this step is the direct decision-making stage for adjusting the number of fuel cell stacks. It involves comparing the dynamic power threshold with the real-time load demand to generate preliminary logic commands for adding or removing stacks. The system collects real-time electrical load data from the user side at a fixed sampling period (e.g., once per second). This data is typically the total electrical power demand of the system, denoted as P_load. The decision is based on two key thresholds dynamically generated and updated in real-time from step S3: the stack input power threshold P_in and the stack output power threshold P_re.

[0122] The system continuously compares P_load with the current P_in. If P_load is consistently higher than P_in (to enhance anti-interference capabilities, a sustained period of time, such as 10 seconds, above the threshold can be used as the judgment criterion, rather than a single instantaneous value), it indicates that the current fuel cell stack configuration can no longer meet the increasing load demand within the efficient or safe range. At this time, the system generates a first control command to increase the number of current fuel cell stacks. This command is a logical signal that means "one additional operating unit (one fuel cell stack or a group of fuel cell stacks) is required."

[0123] The system continuously compares P_load with the current P_re. If P_load is consistently lower than P_re (this can also be determined using a continuous criterion), it indicates that the current configuration is redundant, and some fuel cells operating in the low-load region may lead to efficiency degradation or high-potential risks. At this time, the system generates a second control command to reduce the number of current fuel cells, which means "one operating unit can be reduced".

[0124] If the real-time electrical load P_load is between P_re and P_in, i.e., P_re ≤ P_load ≤ P_in, then the current load is within the "hysteresis interval" or "stable operating interval" constructed by the two thresholds. Within this interval, the system operates in the neighborhood of the optimal power setpoint P_opt, without needing to change the number of fuel cells. The system generates a third control command (or no action command) to maintain the current number of fuel cells.

[0125] The "first control command," "second control command," or "third control command" generated in this step are only preliminary logical decision results and have not yet considered time constraints and safety protection. These commands will be immediately sent to the subsequent S5 step (minimum duration constraint) for filtering. By transforming explicit power comparisons into clear increase / decrease / maintenance commands, the S4 step provides a direct basis for the adaptive reconfiguration of the entire multi-reactor system, and the accuracy of its decision directly depends on the dynamic and precise thresholds provided by the S3 step.

[0126] In one embodiment, the minimum duration threshold for switching the number of fuel cells is set based on the physical time constant of the fuel cell startup / shutdown process and the impact model of frequent start-ups and shutdowns on fuel cell lifetime degradation, so as to ensure that the fuel cell has sufficient stable operating time after each start-up and shutdown.

[0127] Specifically, in this embodiment, the setting of the minimum duration threshold must fully consider the physical time constants of the stack startup and shutdown processes. The startup of a proton exchange membrane fuel cell is not instantaneous; it involves multiple sequential or parallel sub-processes, such as the introduction of reactant gases and pressure establishment, the stack body rising from ambient temperature to its optimal operating temperature, and the thorough wetting of the membrane electrodes. The entire process typically takes several minutes to tens of minutes. Similarly, the normal shutdown process includes steps such as load unloading and gas purging to remove residual reactants to prevent corrosion after shutdown, which also requires a certain amount of time. Therefore, the minimum duration threshold (T_min) must be significantly greater than the time required for a single complete startup and shutdown process to ensure that the stack can truly enter a stable operating or completely shut-down state after a state switch, avoiding receiving a reverse command before the previous switching action is fully completed.

[0128] Secondly, the threshold setting needs to be based on the impact model of frequent start-stop cycles on fuel cell stack lifespan degradation. The durability of a fuel cell stack is closely related to its operating history. Each start-up is accompanied by drastic changes in humidity and temperature, generating mechanical and chemical stresses on the membrane electrode assembly, accelerating its performance degradation. By analyzing historical data or accelerated aging test data, a quantitative relationship model between the number of start-stop cycles and lifespan degradation can be established. Based on this model, combined with the system's expected fuel cell stack lifespan, an acceptable upper limit for start-stop frequency can be derived. The setting of T_min directly determines this upper limit (frequency ≤ 1 / T_min), thereby controlling the lifespan reduction caused by start-stop cycles within the design allowable range. The minimum duration threshold T_min is a key parameter that integrates process physical constraints and long-term lifespan management requirements. For example, it can be set to 30 minutes or 1 hour, which ensures sufficient execution and stabilization time for each start-stop process while fundamentally limiting the maximum number of start-stop cycles per unit time. This is an important guarantee for achieving long-life, high-reliability operation of the fuel cell stack.

[0129] In one embodiment, step S6 specifically includes:

[0130] When the real-time electrical load on the user side exceeds the upper limit of power safety or falls below the lower limit of power safety, a forced execution command is generated to adjust the number of fuel cells. The forced execution command is to perform an increase or decrease operation of the number of fuel cells corresponding to the current overload or underload risk.

[0131] The power safety limit is the maximum total power that all fuel cells can safely and continuously bear given the current number of operating fuel cells;

[0132] The power safety lower limit is the minimum total power allowed to avoid a single fuel cell stack operating in a high-potential corrosion risk zone, given the current number of operating fuel cell stacks.

[0133] Specifically, in this embodiment, this step serves as the highest-priority protection layer in the entire control architecture. Its design goal is to ensure that, under any circumstances, the operating conditions of the fuel cell stack do not exceed its safety and durability boundaries, even if this means temporarily deviating from conventional optimization and control rules. Its implementation relies on two predefined hard safety boundaries that are related to the number of fuel cell stacks.

[0134] First, define and store the upper and lower power safety limits. The upper power safety limit (P_safe_max) is defined as the maximum total power that all online fuel cell stacks can safely and continuously operate without causing overheating, overvoltage, or irreversible damage, given the current number of operating stacks. This value is typically lower than the sum of the maximum power of all stacks, taking into account system heat dissipation capacity, hydrogen supply stability, and necessary power margin. The lower power safety limit (P_safe_min) is defined as the minimum total power that must be maintained to ensure that no single stack operates in a high-potential region (typically corresponding to extremely low load rates) for an extended period, thereby preventing severe corrosion and degradation of the cathode catalyst, given the current number of operating stacks. This lower limit ensures that the output power of a single stack is not lower than the minimum required for its chemical durability.

[0135] Secondly, during operation, the system monitors the real-time electrical load data (P_load) on the user side and simultaneously obtains the number of currently operating fuel cells (n). The control logic is as follows:

[0136] Overload risk assessment and unauthorized access: Compare P_load with P_safe_max(n) corresponding to the current n. If P_load ≥ P_safe_max(n), the system is determined to face an immediate overload risk. At this time, the system will immediately generate a "forced execution instruction: increase the number of fuel cell stacks". This instruction means that the number of operating fuel cell stacks must be increased to distribute the load, reduce the load of each fuel cell stack, and bring it back to the safe range.

[0137] Low-load high-potential risk assessment and overreach: Compare P_load with P_safe_min(n) corresponding to the current n. If P_load ≤ P_safe_min(n), the system is deemed to face a high-potential corrosion risk. At this time, the system will immediately generate a "forced execution instruction: reduce the number of fuel cell stacks". This instruction means that the number of operating fuel cell stacks must be reduced to increase the load rate of the remaining online fuel cell stacks and remove them from the high-potential danger zone.

[0138] The key is that, regardless of the system's current state or whether the "minimum duration constraint" in step S5 is met, once the aforementioned mandatory execution instruction is generated, it will have the highest priority and directly bypass the constraint logic of step S5, being passed to the final execution unit. This "overriding authority" mechanism ensures the real-time and mandatory nature of the safety response, placing equipment protection above operational smoothness, fundamentally enhancing the system's robustness and safety in the face of sudden extreme loads or fault conditions.

[0139] In one embodiment, the coordinated control further includes:

[0140] When executing a control command to increase the number of fuel cell stacks, priority is given to starting fuel cell stacks that are currently in standby mode and have accumulated less operating time.

[0141] When executing the control command to reduce the number of fuel cell stacks, priority is given to shutting down fuel cell stacks that are currently in operation and have accumulated a large amount of operating time, so as to achieve balanced management among the fuel cell stacks.

[0142] Specifically, in this embodiment, the strategy is implemented in the selection of the specific execution target of the fuel cell stack quantity control command. It aims to smooth out the differences in operating losses between fuel cell stacks through intelligent fuel cell stack scheduling, thereby extending the overall service life of the entire fuel cell stack array.

[0143] When the system determines that it needs to execute a control command to increase the number of fuel cell stacks, it must select one of the multiple fuel cell stacks currently in standby (shutdown or standby) state to start. At this time, the lifespan balancing management strategy is activated: the system accesses its maintained "fuel cell stack operation history database," which records health status parameters such as the cumulative operating time, historical start-stop count, and average operating power of each fuel cell stack in the system. Based on the principle of "prioritizing the start of fuel cell stacks with shorter cumulative operating time," the algorithm sorts all standby fuel cell stacks and prioritizes the fuel cell stack with the shortest cumulative operating time as the target for this start. This selection logic ensures that the operating load is more evenly distributed among the fuel cell stacks, avoiding the problem of a few fuel cell stacks being "overworked" due to frequent calls, while other fuel cell stacks are idle for a long time, resulting in uneven performance degradation.

[0144] Conversely, when the system determines that it needs to execute a control command to reduce the number of fuel cell stacks, it must select one of the currently running stacks to shut down. In this case, the management strategy follows the principle of "prioritizing the shutdown of stacks with longer cumulative operating time." The system sorts all online running stacks by their cumulative operating time and prioritizes the stack with the longest cumulative operating time as the target for shutdown. The purpose of this is to allow stacks with longer operating times to have more opportunities for "rest" or "low-load maintenance," while allowing stacks with shorter operating times to continue performing tasks, thus achieving a dynamic "rotation" mechanism.

[0145] This intelligent selection mechanism based on operational history is the core of achieving balanced management among fuel cell stacks. It goes beyond simple sequential start-up / shutdown or random selection, combining long-term lifetime prediction with real-time operational decision-making. Long-term implementation of this strategy can effectively suppress the widening trend of stack performance parameter dispersion caused by uneven operating time, keeping the performance degradation of the entire stack array synchronized. This not only helps maintain system stability but also makes future maintenance and replacement plans more predictable, further reducing system operating costs from a life-cycle perspective.

[0146] In one embodiment, the multi-stack fuel cell cogeneration system includes multiple parallel proton exchange membrane fuel cell modules, a shared thermal storage module, and a user-side coupling interface; each proton exchange fuel cell module includes a stack, an air supply subsystem, a thermal management subsystem, and a power conversion unit; the heat recovery loop, through valve switching, can selectively introduce waste heat generated by the stack and auxiliary components into the thermal storage module or dissipate it to the environment.

[0147] Specifically, in this embodiment, the optimization method for a multi-stack fuel cell cogeneration system based on dual-threshold control provided in this application is applicable to systems such as... Figure 2 The diagram shows a multi-stack fuel cell cogeneration system based on multi-terminal recovery technology. A single stack module of the multi-stack fuel cell cogeneration system includes: a PEMFC heat exchanger (1), a PEMFC radiator (2), a first three-way valve (3), a first coolant pump (4), an air filter (5), an air compressor (6), an expander (7), an intercooler (8), a PEMFC stack (9), a second coolant pump (10), a DC / DC converter (11), a second three-way valve (12), an auxiliary radiator (13), an auxiliary heat exchanger (14), and an exhaust heat exchanger (15). The thermal storage module includes a first water pump (16), a cold water tank (17), a hot water tank (18), a second water pump (19), and a third three-way valve (20). A single container integrates the multi-stack fuel cell system module, the stack heat recovery module, the stack heat dissipation module, the auxiliary heat recovery module, the auxiliary heat dissipation module, and the exhaust heat recovery module. The connection methods between the modules are clearly shown in the detailed schematic diagram of a single PEMFC system.

[0148] See Figure 2 The megawatt-level multi-stack fuel cell cogeneration system in this application consists of four fuel cell system containers and one energy storage and distribution container. Each fuel cell system container houses six fuel cell stacks and five heat exchange modules. When heat recovery is initiated, the fuel cell stack and auxiliary component heat recovery modules, the fuel cell stack exhaust heat recovery module, and the cold water tank and hot water tank circulation are simultaneously activated. The connection and operation of each module are as follows: First, the first water pump (16) delivers the cold water in the cold water tank (17) to the fuel cell power generation system container, and realizes system heat recovery through three parallel branches: In the first branch, the first coolant pump (4) drives the coolant to deliver the heat generated by the intercooler (8) and the PEMFC stack (9) to the PEMFC heat exchanger (1) to recover the heat of the stack operation; In the second branch, the second coolant pump (10) drives the coolant to flow through the air compressor (6) and the DC / DC converter (11) in sequence, and finally enters the auxiliary heat exchanger (14) to recover the heat of the above equipment operation; In the third branch, the tail heat exchanger (15) recovers the waste heat of the humid air, and then flows into the hot water tank (18). The three branches finally converge to complete the multi-port heat recovery of the system. In addition, after being filtered by the air filter (5), the air is pressurized by the air compressor (6) and delivered to the fuel cell. The fuel cell exhaust first recovers some kinetic energy through the expander (7) to reduce the power consumption of the air compressor (6), and then enters the exhaust heat exchanger (15) to recover the latent heat in the humid air. After heat recovery is completed, hot water is delivered to the user side by the second water pump (19) to meet the heat load demand. When heat recovery is turned off, the system switches the circuit to the PEMFC radiator (2) and the auxiliary radiator (13) by adjusting the first three-way valve (3) and the second three-way valve (12) to dissipate heat from the system equipment. If the water temperature in the hot water tank (18) does not reach the set requirement, the low-temperature hot water in the hot water tank (18) is mixed with the cold water in the cold water tank (17) by adjusting the third three-way valve (20) to increase the temperature after heat recovery.

[0149] See Figure 3 , Figure 3This application's embodiment of the dual-threshold multi-stack fuel cell system optimization and control strategy establishes a matching relationship between the power demand range and the optimal number of operating stacks, and introduces a power threshold with dynamic hysteresis characteristics to achieve comprehensive system benefit optimization. The input threshold (upper limit) and cut-off threshold (lower limit) are dynamically adjusted based on real-time heat recovery efficiency and heat load status: when the thermal storage system starts heat recovery, the input and cut-off thresholds can be increased; conversely, they can be decreased, allowing the multi-stack fuel cell system to operate near its optimal condition under different heat recovery states. The interval between the two thresholds constitutes the "hysteresis zone," the width of which can be set according to the minimum start-up and shutdown time constraints of the stacks and load fluctuation characteristics (e.g., the small-range fluctuation range of megawatt-level loads typically does not exceed 200kW). This design allows the system to trigger stack addition when the load exceeds the dynamic input threshold, and only allows stack reduction after the load drops to the dynamic cut-off threshold. This hysteresis control mechanism based on dynamic adjustment of thermal state effectively avoids the impact of small-range load fluctuations (as shown in the attached figure) on the system's performance. Figure 4 The frequent mode switching caused by the load curve fluctuating within the hysteresis region, as shown in the detailed diagram, improves the system's operational stability and equipment durability.

[0150] See Figure 4 , Figure 4 This application compares the number of fuel cell start-up and shutdown switching operations under the dual-threshold and single-threshold control schemes provided in the embodiments of this application. The study conducts energy consumption analysis based on a typical monthly electricity load scenario in Guangzhou, and constructs a 2.5MW-scale electricity load curve using user data provided by the project demonstration unit, coupled with a domestic hot water load of the same scale for simulation. Simulation results show that within a 30-day operating cycle, the single-threshold control scheme triggers 205 fuel cell number switching operations, while the dual-threshold control scheme of this application triggers only 109, reducing the number of switching operations by 46.8%. Further observation is made during the typical fluctuation period from 324h to 348h (e.g., ...). Figure 5 (As shown in the enlarged view), the single-threshold scheme triggers 11 switching operations, while the dual-threshold scheme triggers only 4, reducing the number of switching operations by 63.6%. These results demonstrate that the dual-threshold control mechanism proposed in this application can effectively suppress frequent switching of the fuel cell stack caused by small-range load fluctuations, significantly reducing the switching frequency of the number of operating fuel cell stacks, thereby helping to enhance the stability of fuel cell stack operation and improve the durability of key equipment.

[0151] See Figure 5 , Figure 5This application presents the dual-threshold stack optimization control results under a typical scenario provided in the embodiments of this application. To simultaneously consider heating and domestic hot water loads, the simulation uses Shanghai's electricity load data as a typical scenario, constructing a 2.5MW-scale electricity load curve, and coupling it with an equivalent-scale heat load for comprehensive benefit analysis. The system stack control effect is determined based on an integrated evaluation system and its weights formed by coupling multiple dimensions such as economy, energy efficiency, power efficiency, environment and health. The comprehensive evaluation result of this system is expressed as a comprehensive benefit value, with a baseline value of 1; when the comprehensive benefit value is greater than 1, it represents positive benefits. Simulation results show that during 24 hours of operation, the average comprehensive benefit of the system is approximately 1.0434, indicating that the stack control strategy achieves an improvement of approximately 4.34% in comprehensive indicators compared to the non-stack control method. In terms of economy, the levelized cost of electricity (LCOE) is calculated based on net power generation or the equivalent value of net power generation and heat recovery, depending on the heat recovery status. Since the lifespan and total lifecycle cost of fuel cells change dynamically with the number of fuel cells put into operation, simulation results show that the 24-hour average LCOE is optimized from 0.4974 $ / kWh before the optimization and control to 0.4356 $ / kWh, achieving a cost saving of 12.42%.

[0152] Other embodiments or specific implementations of this application can be found in the above-described method embodiments, and will not be repeated here.

[0153] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0154] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0155] It should be particularly noted that, 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, or of course, by hardware. Based on this understanding, the above technical solutions, in essence or the parts that contribute 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., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0156] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application 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. Such 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 this application.

Claims

1. An optimization method for a multi-stack fuel cell cogeneration system based on dual-threshold control, characterized in that, Includes the following steps: Step S1: Based on the simulation or historical operation optimization results of the multi-stack fuel cell cogeneration system, establish the mapping relationship between the operating power setpoints corresponding to different combinations of the number of fuel cell stacks in operation and different heat recovery states of the hot water storage tank. Step S2: Monitor the heat storage status parameters of the hot water storage tank in real time, compare them with the preset start liquid level threshold and stop liquid level threshold, and generate a start command or stop command for the heat recovery circuit of the hot water storage tank according to the comparison result, while updating the current heat recovery status identifier. Step S3: Based on the current heat recovery status identifier updated in step S2 and the real-time number of fuel cell stacks, query the system's optimal operating power setpoint mapping relationship established in step S1, and dynamically determine the fuel cell stack input power threshold and fuel cell stack cut-out power threshold for fuel cell stack quantity control under the current operating conditions, wherein the fuel cell stack input power threshold is higher than the fuel cell stack cut-out power threshold, and a hysteresis interval is formed between the two. Step S4: Collect real-time electrical load data from the user side and compare it with the stack input power threshold and stack cut-off power threshold dynamically determined in step S3 to generate control commands to increase or decrease the number of stacks; Step S5: Set the minimum duration threshold for switching the number of fuel cells. When a control command to increase or decrease the number of fuel cells is received from step S4, determine whether the time interval between the current time and the last effective fuel cell number switching time is less than the minimum duration threshold. If the conditions are met, the control command is blocked, and the current number of fuel cell stacks remains unchanged; If the conditions are met, the control command is marked as a command to be executed; Step S6: Preset the upper and lower limits of the power safety under the current number of operating fuel cells, and monitor the real-time power load data on the user side in real time. If the power safety upper limit is exceeded or the power safety lower limit is lowered, a forced execution instruction is generated. The forced execution instruction is to perform the increase or decrease operation of the number of fuel cells corresponding to the current overload or underload risk. Step S7: Execute the start or stop command from the heat recovery loop in step S2, and the stack quantity control command finally determined after the constraints in step S5 and the over-authorization judgment in step S6, to perform coordinated control of the multi-stack fuel cell cogeneration system.

2. The method according to claim 1, characterized in that, The S1 step specifically includes: Based on the simulation or historical operation optimization results of the multi-stack fuel cell cogeneration system, the optimal stack operating power setpoints corresponding to different stack operating numbers under the two states of starting heat recovery and stopping heat recovery are established. The optimal operating power setpoint mapping relationship is stored in the form of a table or function. The input is the number of fuel cell stacks in operation and the heat recovery status identifier, and the output is the corresponding optimal total operating power value of the fuel cell stack. The optimal total operating power value of the fuel cell stack is obtained by multi-objective optimization calculation with the goal of maximizing the overall benefits of the multi-stack fuel cell cogeneration system.

3. The method according to claim 1, characterized in that, The S2 step specifically includes: The liquid level and operating status of the hot water storage tank are monitored in real time, and the judgment is made by comparing the preset stop liquid level threshold and start liquid level threshold. If the liquid level is higher than the preset stop liquid level threshold, a stop heat recovery command is generated and output to shut down the heat recovery circuit; If the liquid level is lower than the preset start-up liquid level threshold, a start-up heat recovery command is generated and output to activate the heat recovery circuit.

4. The method according to claim 1, characterized in that, In step S3, the dynamic determination of the stack input power threshold and the stack cut-out power threshold specifically includes: Real-time monitoring of the heat storage status of the hot water storage tank and the real-time number of electric stacks in operation; Using the current number of operating fuel cells and the current heat recovery status identifier as indexes, the corresponding optimal operating power value is retrieved from the mapping relationship; Based on the optimal operating power value and combined with the preset hysteresis interval width, the stack input power threshold and stack cut-out power threshold are calculated.

5. The method according to claim 1, characterized in that, The S4 step specifically includes: Real-time electrical load data from the user side is collected and compared with the dynamically updated stack input power threshold and stack cut-out power threshold. If the real-time electrical load on the user side continues to be higher than the power threshold of the fuel cell stack, a first control command is generated to increase the current number of fuel cell stacks. If the real-time electrical load on the user side remains below the stack cut-off power threshold, a second control command is generated to reduce the current number of stacks. If the real-time electrical load on the user side is between the stack input power threshold and the stack cut-out power threshold, a third control command is generated to maintain the current number of stacks.

6. The method according to claim 1, characterized in that, In step S5, the minimum duration threshold for switching the number of fuel cells is set based on the physical time constant of the fuel cell startup / shutdown process and the impact model of frequent startups and shutdowns on fuel cell lifetime decay, so as to ensure that the fuel cell has sufficient stable operating time after each startup and shutdown.

7. The method according to claim 1, characterized in that, Step S6 specifically includes: When the real-time electrical load on the user side exceeds the upper limit of power safety or falls below the lower limit of power safety, a forced execution command is generated to adjust the number of fuel cells. The forced execution command is to perform an increase or decrease operation of the number of fuel cells corresponding to the current overload or underload risk. The power safety limit is the maximum total power that all fuel cells can safely and continuously bear given the current number of operating fuel cells; The power safety lower limit is the minimum total power allowed to avoid a single fuel cell stack operating in a high-potential corrosion risk zone, given the current number of operating fuel cell stacks.

8. The method according to claim 1, characterized in that, In step S7, the cooperative control further includes: When executing a control command to increase the number of fuel cell stacks, priority is given to starting fuel cell stacks that are currently in standby mode and have accumulated less operating time. When executing the control command to reduce the number of fuel cell stacks, priority is given to shutting down fuel cell stacks that are currently in operation and have accumulated a large amount of operating time, so as to achieve balanced management among the fuel cell stacks.

9. A multi-stack fuel cell cogeneration system based on dual-threshold regulation, characterized in that, The multi-stack fuel cell cogeneration system includes multiple parallel proton exchange membrane fuel cell modules, a shared thermal storage module, and a user-side coupling interface. Each proton exchange fuel cell module includes a stack, an air supply subsystem, a thermal management subsystem, and a power conversion unit. The heat recovery loop can selectively introduce waste heat generated by the stack and auxiliary components into the thermal storage module or dissipate it to the environment through valve switching.

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 optimization method for a multi-stack fuel cell cogeneration system based on dual threshold control as described in any one of claims 1 to 8.

Images