A power distribution control method for a hydrogen fuel cell power generation system

Abstract

The application discloses a power distribution control method of a hydrogen fuel cell power generation system, which is suitable for multiple parallel electric pile modules. The method acquires system power demand and operating parameters of each electric pile module, judges the abnormality of the power demand, and divides the electric pile modules into healthy working modules, healthy shutdown modules, fault limited modules and invalid modules. The number of working modules is adjusted according to the power demand, the target power is distributed in stages in the order of minimum power, lower limit of comfort zone, highest efficiency point, upper limit of comfort zone, upper limit of efficiency interval and maximum power, and the distribution amount of each stage is determined in combination with the health degree. The modules reaching the current maximum allowable power are fixed and the remaining power is redistributed, the power deviation of the start-stop transition module is compensated, and finally the target power is output and the health degree and the module state are updated. The application helps to improve the coordination of multi-pile power distribution and reduce the power fluctuation under the start-stop and fault limited working conditions.

Description

Technical Field

[0001] This invention relates to the field of energy management and operation control technology for hydrogen fuel cell power generation systems, and specifically to a power distribution control method for hydrogen fuel cell power generation systems. Background Technology

[0002] Stationary hydrogen fuel cell power generation systems are evolving from single-stack power supply to parallel power supply of multiple stack modules. Multi-stack architecture can improve system output capacity and power supply redundancy, but due to differences in manufacturing, cumulative operating time, performance degradation, and fault conditions, the maximum allowable power, optimal efficiency point, and suitable operating range of each stack often differ.

[0003] Existing power allocation methods mostly employ average allocation, capacity ratio allocation, or step-by-step switching strategies, making it difficult to simultaneously balance the number of fuel cell stacks started / stopped, operating efficiency, health status, and lifespan. During load changes, module start / stop, or partial fuel cell stack derating, issues such as power deficit, power overshoot, frequent start / stop, and some fuel cell stacks operating in unsuitable power ranges for extended periods can easily arise. Furthermore, when system demand is too low or exceeds system capacity limits, the lack of appropriate judgment and protection may lead to inefficient system operation or excessive output commands.

[0004] Therefore, there is an urgent need for a power distribution control method for hydrogen fuel cell power generation systems that can coordinate the switching and target power allocation of multiple fuel cell modules by combining stack health, comfortable operating range, real-time allowable power boundary and start-stop transition state, and has the capabilities of demand anomaly protection, fault derating tolerance and transient power compensation.

[0005] A search revealed a Chinese patent document disclosing a power control method for a multi-stack fuel cell system based on power ranges [Application No.: 202111505676.2, Publication No.: CN114188575B]. This comparative patent, based on other power control methods, pre-determines the number of fuel cell stacks to be operated and the target power for each stack. It divides the output power range of a single fuel cell stack into multiple power ranges, records the cumulative operating time of each stack within different power ranges, and allocates the target power to the stack with the shortest cumulative operating time within the corresponding power range. When multiple target power values ​​exist, this patent also determines the priority allocation order of the target power based on the maximum difference in operating time between different fuel cell stacks within each power range, in order to balance the commissioning time and power conditions of each fuel cell stack.

[0006] While the comparative patent achieves a similar goal of balancing the operational time of each fuel cell stack, mitigating the rapid degradation of some stacks, and improving the lifespan and performance of multi-stack systems by allocating target power to the stack with the shortest operating time within the corresponding power range, and features the characteristic of recording cumulative operating time according to power range and selecting specific operating stacks accordingly, this invention has complete closed-loop control features such as demand anomaly identification, three-state classification of healthy working modules, healthy shutdown modules, and fault-limited modules, dynamic addition and reduction of working modules, stepwise allocation from minimum power through the lower limit of the comfort zone, the highest efficiency point, the upper limit of the comfort zone and the upper limit of the efficiency range to the maximum power, fine-tuning of start-stop transition power, fixed derating of fault modules, and re-compensation for power deficit. From the core claims and specification of the comparative patent, it mainly solves the problem of balancing the operating time of a given target power to specific fuel cell stacks, and does not show the collaborative control mechanism based on health, comfortable working zone, highest efficiency point, current allowable maximum power, and start-stop transient state adopted by this invention. Therefore, this invention has more complete technical means in terms of complex load adaptation, fault derating fault tolerance, and power stability control during start-stop processes. Summary of the Invention

[0007] To address the problems existing in the prior art, the purpose of this invention is to provide a power distribution control method for a hydrogen fuel cell power generation system.

[0008] A power distribution control method for a hydrogen fuel cell power generation system, the system comprising multiple parallel fuel cell stack modules, characterized in that it includes: S101, obtain power requirements and operating parameters of each fuel cell stack module; S102, perform demand anomaly judgment, and classify each fuel cell stack module into healthy working module, healthy shutdown module, fault-limited module and failed module according to the failure flag, operating status and current maximum allowable power of each fuel cell stack module; S103, add or remove working modules according to power requirements; S104: Starting from the minimum power, the target power of the working module is allocated sequentially according to the lower limit of the comfort zone, the highest efficiency point, the upper limit of the comfort zone, the upper limit of the efficiency zone, and the maximum power, and the allocation amount for each stage is determined based on the health status; the target power of the module that has reached the current maximum allowable power is fixed, and the remaining power is redistributed. S105, compensate for power fluctuations based on the difference between the current power and the target power of the start-stop transition module; S106 outputs the target power and updates the health and module status.

[0009] Preferably, the operating parameters include: current power p0, target power p1, comfort zone consisting of lower comfort zone limit p2 and upper comfort zone limit p3, maximum power p4, minimum power p5, current maximum allowable power p6, health status p7, highest efficiency point power p8, and efficiency range consisting of lower efficiency range limit p9 and upper efficiency range limit p10; the parameters of the hydrogen fuel cell power generation system include power demand a0, actual power a1, working module list a2, and shutdown module list a3, wherein the actual power a1 is the sum of the current power p0 of each stack module.

[0010] The above technical solutions can describe each fuel cell stack module in terms of power status, efficiency characteristics, output boundaries and health status. System-level parameters can reflect the current load demand, actual output and module start-up and shutdown status of the power generation system, providing a parameter basis for subsequent module status judgment, working module adjustment and target power allocation.

[0011] Specifically, the current power p0 represents the actual output state of the fuel cell stack module, the target power p1 represents the power command issued by the controller to the fuel cell stack module, the minimum power p5 and maximum power p4 define the normal power range of the module, and the current maximum allowable power p6 reflects the actual output limit of the module under the current performance state. The comfort zone, the highest efficiency point power, and the efficiency range characterize the operating characteristics of the fuel cell stack module at different power stages, while the health status p7 reflects the module's accumulated operating conditions, degradation status, or overall health condition.

[0012] In practical applications, the system controller can determine whether power reallocation is needed based on the power demand a0 and the actual power a1, and determine the current participation status of each module based on the working module list a2 and the shutdown module list a3. By combining module parameters with system parameters, the power allocation for each control cycle can be based on the real-time operating capabilities of each fuel cell module and the overall system requirements.

[0013] Preferably, step S102 includes: S201, determine whether the power demand is within the system's allowed power range. If the power demand exceeds the power range, perform a protective return and do not output the target power that exceeds the system's capacity. S202, If the deviation between the actual power of the system and the power demand is within the preset allowable deviation range, then the current power allocation result is maintained; S203, if the power demand is lower than the minimum start-up threshold, the target power of the healthy working module is set to zero, and the healthy working module is transferred to the healthy shutdown state; S204 sets the target power of the fuel cell stack module with the failure flag to zero.

[0014] The above technical solution enables the system power demand, actual output deviation, minimum power demand, and module failure status to be pre-judged before module start-up and shutdown and target power allocation, thus avoiding unnecessary power adjustments by the controller when demand is abnormal, output is basically stable, or the module does not have normal output conditions.

[0015] Specifically, when the power demand exceeds the system's allowable range, a protective return mechanism prevents the controller from generating power commands that exceed the system's output capacity. When the deviation between the actual power and the power demand is within the allowable range, the existing allocation results are maintained to reduce redundant adjustments caused by small power fluctuations. When the power demand is insufficient to support the stable operation of the fuel cell stack module, healthy operating modules are switched to a shutdown state, and zero-power processing is performed on failed modules.

[0016] In practical applications, this anomaly identification process ensures that subsequent control procedures are initiated only when power demand is valid and adjustments are truly necessary. This reduces frequent adjustments to modules within critical demand regions and prevents fuel cell modules from continuously operating below their optimal power levels, while also allowing failed modules to exit the power allocation process promptly.

[0017] Preferably, step S102 further includes: S211, classify fuel cell stack modules that do not have a failure flag, are in operation, and whose current maximum allowed power p6 is equal to the maximum power p4 as healthy working modules; S212, classify fuel cell stack modules that do not have a failure flag and are in a shutdown state as healthy shutdown modules; S213, classify fuel cell stack modules that do not have a failure indicator, are in operation, have a current maximum allowable power p6 lower than the maximum power p4, and still have power output capability as fault-limited modules; S214, the fuel cell stack module with failure flag is classified as a failure module, and the target power p1 of the failure module is set to zero; S215, when the target power p1 of the fault-limited module reaches or exceeds the current maximum allowable power p6, its target power p1 is limited to p6, and its fixed flag isFixed is set to true.

[0018] The above technical solution enables the classification and management of each fuel cell module based on its failure flags, operating status, and actual allowable output capacity, allowing modules that are operating normally, shut down normally, or have limited performance but can still output power to enter the corresponding control processes.

[0019] Specifically, the healthy operating module is used to handle the current system power demand, the healthy shutdown module serves as a candidate module for subsequent startup, and the fault-limited module continues to operate within its current maximum allowable power range. When the fault-limited module reaches its actual allowable output limit, its target power is restricted from further increasing by a fixed flag (isFixed), allowing subsequent power allocation to be transferred to other operating modules that still have adjustment margins.

[0020] In practical applications, this classification method avoids directly removing fuel cell stack modules that have degraded in performance but have not yet completely failed from the system, allowing them to continue providing power within the current allowable range. Simultaneously, by setting a target power limit and a fixed flag, it prevents the controller from issuing power commands exceeding the current capacity of restricted modules and provides a clear basis for subsequent power deficit compensation based on module status.

[0021] Preferably, step S103 includes: S301 records the number of currently working modules; S302, determine whether an additional working module is needed based on power requirements. If so, select a module to be started from the health shutdown module and transfer the module to be started to the working module list. S303, when there is no need to add working modules, determine whether it is necessary to reduce working modules. If so, select the module to be stopped from the working module list and transfer the module to be stopped to the healthy shutdown module. S304. Compare the number of working modules before and after the adjustment. If the number of working modules changes, repeat steps S301 to S303 until the number of working modules no longer changes or the number of iterations reaches ten.

[0022] The above technical solution enables the number of fuel cell stack modules participating in operation to be adjusted according to the current power demand before the target power allocation. This allows the system to first determine how many fuel cell stack modules will bear the load, and then allocate the target power to each working module separately, thereby distinguishing between module switching control and inter-module power allocation.

[0023] Specifically, when the output capacity of the existing working modules is insufficient to meet the power demand, modules that meet the startup conditions are selected from the healthy shutdown modules and added to the working module list. When the number of working modules is large and the power demand can still be met after reducing some modules, modules that meet the shutdown conditions are moved to the healthy shutdown state. After each addition or removal of modules, it is reassessed whether the number of working modules still needs to be adjusted.

[0024] In practical applications, iterative judgment can handle situations where the available power range of the system changes after adding or removing a single module, gradually bringing the number of working modules closer to the operating combination corresponding to the current power demand. Setting an iteration limit can also prevent the adjustment process from continuously looping due to abnormal parameters or repeated changes in critical conditions.

[0025] Preferably, when selecting a module to be started, healthy shutdown modules with failure flags are excluded, and the remaining healthy shutdown modules are sorted in ascending order using 1-p6 / p4, p2 / p4, p7, and module identifier as the first to fourth sorting keys. When the working module list is empty, the minimum power p5 of the candidate module is used as the power required for startup. When the working module list is not empty, the lower limit of the comfort zone p2 of the candidate module is used as the power required for startup, and candidate modules that meet the power requirements are selected according to the sorting results. When selecting a module to be shut down, working modules with failure flags are excluded, and the remaining working modules are sorted in ascending order using p6 / p4, -p2 / p4, -p7, and module identifier as the first to fourth sorting keys. Based on the minimum power or the sum of the lower limits of the comfort zone of the remaining working modules after the candidate modules exit, the power requirements, and the preset hysteresis, candidate modules that meet the shutdown conditions are selected according to the sorting results.

[0026] The above technical solution can comprehensively consider the actual available output capacity, lower limit of comfort zone, health status and module identification of fuel cell module, and orderly select modules to be started and modules to be stopped, avoiding the determination of the start-up and shutdown sequence of modules based solely on module number, rated power or a single operating parameter.

[0027] Specifically, the ranking of modules to be started is used to prioritize candidate modules that can meet the current power requirements, and the minimum power or the lower limit of the comfort zone is used as the power required for startup, depending on whether the system already has working modules. The selection of modules to be shut down is based on the power range that the remaining modules need to handle after the candidate modules are shut down, and a preset hysteresis is used to create an interval between the startup and shutdown conditions.

[0028] In practical applications, this selection method can reduce the likelihood of new modules starting up below the minimum power or lower limit of the comfort zone, and also prevent the remaining working modules from being unable to stably handle the system load after a module stops. Setting the hysteresis can reduce the possibility of the same module repeatedly starting and stopping when the power demand fluctuates near the critical value, making the module quantity adjustment process smoother.

[0029] Preferably, step S104 includes: S401, set the target power of each working module to the corresponding minimum power p5, and calculate the remaining power between the power requirement and the sum of the target power of each working module; S402, allocate the remaining power according to the health status p7 of each working module, so that the target power of each working module is increased from the minimum power p5 to the lower limit of the comfort zone p2; S403, when there is still residual power, power is allocated according to the health status p7 of each working module, so that the target power of each working module is successively increased to the highest efficiency point power p8, the upper limit of the comfort zone p3, the upper limit of the efficiency range p10 and the maximum power p4. S404, in any allocation phase, if the target power of the fuel cell module reaches the current maximum allowable power p6, then the target power of the fuel cell module is fixed at p6, and the remaining power is redistributed to other working modules with power regulation margin.

[0030] The above technical solution can allocate system power demand step by step according to different power operation stages, so that each working module first obtains the basic power required to maintain normal operation, and then increases the target power in sequence to the comfort zone, the highest efficiency point, the efficiency range and the maximum output range, avoiding direct average allocation without considering the module operation range.

[0031] Specifically, the system first sets the target power of each working module to the minimum power and calculates the remaining power based on the power demand. Then, within each allocation phase, it determines the allocation amount for each module based on its health status, gradually bringing the target power closer to the corresponding phase's upper limit. When a module is limited by its current maximum allowable power, its target power is fixed, and the unallocated power is distributed to other modules with adjustment capabilities.

[0032] In practical applications, this phased power allocation method ensures that the formation of the module's target power is aligned with its power boundaries, comfortable operating range, and efficiency characteristics, reducing situations where some modules are in unsuitable power ranges while others still have available adjustment space. Simultaneously, health status participates in the power allocation at each stage, ensuring that the current operating results match the actual state of each module.

[0033] Preferably, step S105 includes: S501, determine the exit transition module where the target power p1 is zero and the current power p0 is greater than zero, and calculate the sum of the current powers of the exit transition modules; S502, when the sum of the current power reaches the preset fine-tuning threshold, a candidate module is determined from the working modules other than the exit transition module and which have downward adjustment margin, and the target power of the candidate module is reduced. S503, determine the start-up transition module where the target power p1 is greater than zero and the current power p0 is lower than the minimum power p5, and calculate the power deficit between the target power and the current power of the start-up transition module; S504, when the power deficit reaches a preset fine-tuning threshold, a candidate module is determined from the working modules other than the start-up transition module that have an upward adjustment margin, and the target power of the candidate module is increased. S505 limits the target power that is higher than the current maximum allowable power p6 to p6, and adjusts the target power that is greater than zero and lower than the minimum power p5 to p5.

[0034] The above technical solution can compensate for the transition process where the actual power of the fuel cell module cannot immediately reach the target power after receiving a start or stop command, so that the system target power allocation result is coordinated with the actual output of each module during the start and stop phases.

[0035] Specifically, for exit transition modules where the target power has been set to zero but are currently still maintaining power output, the output of the exit module that has not yet disappeared is offset by reducing the target power of other working modules. For startup transition modules that have obtained the target power but whose current output has not yet reached the minimum power, the output of the startup module that has not yet been established is temporarily supplemented by increasing the target power of other working modules. Adjustments are only performed when the corresponding power difference reaches the fine-tuning threshold to reduce frequent control caused by small deviations.

[0036] In practical applications, this fine-tuning process can reduce the impact of fuel cell module start-up hysteresis or shutdown delay on the actual system output, enabling temporary power compensation relationships between working modules during the transition phase. After fine-tuning, boundary corrections are performed using the current maximum and minimum allowable power to avoid generating target power commands that exceed the module's actual output limit or fall below the stable operating limit.

[0037] Preferably, compensating for the power deficit caused by the faulty module includes: S511, when the target power p1 of the fault-limited module exceeds its current maximum allowable power p6, the target power p1 of the fault-limited module is limited to the current maximum allowable power p6, its fixed flag isFixed is set to true, and the difference between the target power p1 before limitation and the current maximum allowable power p6 is determined as the power deficit. S512, determine candidate modules from working modules other than the fault-limited module and whose fixed flag isFixed is false, and determine the difference between the current maximum allowable power p6 and the target power p1 of the candidate module as the increaseable power capacity of the corresponding candidate module; S513, set the power to be compensated to the power deficit, and traverse each candidate module in a preset order; S514, the smaller value between the power to be compensated and the increaseable power capacity of the current candidate module is determined as the power increment, the power increment is increased to the target power p1 of the current candidate module, and the power increment is deducted from the power to be compensated; when the target power p1 of the current candidate module reaches its current maximum allowable power p6, the fixed flag isFixed of the candidate module is set to true. S515, Repeat step S514 until the power to be compensated is not greater than zero or all candidate modules have been traversed.

[0038] The above technical solution enables the limiting of a fault-limited module's target power to within the current maximum allowable power p6 when the module's target power exceeds its current actual output capacity. The power deficit resulting from the limiting operation is then transferred to other working modules that still have upward adjustment margin. The fixed flag isFixed indicates whether the stack module's target power has reached the current maximum allowable power, rather than solely based on whether the current maximum allowable power p6 is lower than the maximum power p4. This differentiates the criteria for determining fault-limited and power-fixed states. Existing specifications also consider reaching p6 and being unable to further increase power as the condition for isFixed to be true.

[0039] Specifically, when the original target power p1 of a fault-limited module is higher than the current maximum allowable power p6, the system controller uses the difference between the two as the power to be compensated. For other working modules with the fixed flag is false, the system controller determines the increaseable power capacity based on the difference between its current maximum allowable power p6 and the target power p1, and performs power compensation one by one in a preset order. The power increment allocated each time does not exceed the increaseable power capacity of the current candidate module to avoid the compensated target power exceeding the actual allowable output limit of the candidate module.

[0040] In practical applications, power deficit compensation ends when all the power to be compensated has been allocated. When all candidate modules have reached their corresponding current maximum allowable power, but the power to be compensated is still greater than zero, the system retains the uncompensated power deficit and no longer issues target power commands exceeding the current maximum allowable power to any fuel cell module. This allows fuel cell modules with limited performance but not completely failed to continue derated operation, and uses the remaining power capacity of other working modules to compensate for their output deficit.

[0041] Preferably, step S106 includes: S601, set the target power p1 of the health shutdown module and the failure module to zero; S602, based on the operating status of each fuel cell stack module in the current control cycle, the current power p0, the target power p1, and the control cycle time increment, update the health status p7 of each fuel cell stack module; S603, based on the updated health status p7, failure flag isFailed, operating status, current maximum allowable power p6 and maximum power p4, update the healthy working module, healthy shutdown module, fault-limited module and failure module; S604, add the healthy working module and the fault-limited module to the working module list a2, add the healthy shutdown module and the failed module to the shutdown module list a3, and update the fixed flag isFixed according to whether the target power p1 of each stack module has reached the corresponding current allowable maximum power p6; when the target power p1 reaches the current allowable maximum power p6, set the fixed flag isFixed to true, otherwise set the fixed flag isFixed to false; S605, use the updated health status p7, the working module list a2, the shutdown module list a3 and the fixed flag isFixed as the input parameters for the next control cycle.

[0042] Through the above technical solution, after the module quantity adjustment, phased power allocation, start-stop transition power fine-tuning, and fault-limited module power deficit compensation are completed in the current control cycle, the target power, health status, module status, belonging list, and fixed flags of each fuel cell module can be uniformly updated, and the update results can be fed back to the next control cycle, so that a continuous closed-loop control relationship can be formed between each control cycle.

[0043] Specifically, the system controller first sets the target power p1 of the healthy shutdown module and the failed module to zero to prevent fuel cell modules that have already exited operation or no longer have the conditions to continue outputting power from bearing the system power. Subsequently, the health status p7 is updated according to the running time, power output status and control cycle time increment of each fuel cell module in the current control cycle, so that the health status can reflect the state changes of the fuel cell modules after the current control cycle.

[0044] In practical applications, fuel cell modules that do not have a failure flag, are in operation, and whose current maximum allowable power p6 is equal to the maximum power p4 are updated to healthy operating modules; fuel cell modules that do not have a failure flag and are in a shutdown state are updated to healthy shutdown modules; fuel cell modules that do not have a failure flag, are in operation, whose current maximum allowable power p6 is lower than the maximum power p4, and still have output capability are updated to fault-limited modules; and fuel cell modules that have a failure flag are updated to failure modules.

[0045] After the module status update is completed, healthy operating modules and fault-limited modules together constitute the operating module list a2 for the next control cycle, and healthy shutdown modules and failed modules together constitute the shutdown module list a3. The fixed flag isFixed is determined only based on whether the target power p1 reaches the current maximum allowable power p6, and is no longer directly determined based on whether p6 is lower than p4. The updated health status p7, operating module list a2, shutdown module list a3, and fixed flag isFixed continue to be used for module start / stop selection, phased power allocation, and power deficit compensation in the next control cycle, thus forming a cyclical process of mutual feedback between health status update and power control.

[0046] Compared with the prior art, the present invention has the following advantages: 1. This invention determines the target power in stages based on the health status, comfort zone, peak efficiency point, and current maximum allowable power of each fuel cell stack module. After some modules reach their allowable output limits, the remaining power is transferred to modules that still have adjustment margins. This method reduces the likelihood of fuel cell stacks operating in inefficient or unsuitable power ranges for extended periods, ensuring that fuel cell stacks in different operating states bear loads commensurate with their actual capabilities, and allowing the healthy state to continuously participate in power allocation during subsequent control cycles.

[0047] 2. This invention performs power fine-tuning on modules that have just started but have not yet established a target output, as well as modules that have received a shutdown command but whose current power has not completely disappeared. Other working modules with adjustment margins temporarily compensate for the corresponding power deficit or power margin. At the same time, for modules whose performance has degraded but still have output capability, they can be derated according to the current maximum allowable power, and the resulting power deficit can be redistributed, thereby reducing the impact on the total system output during start-up, shutdown, and restricted operation. Attached Figure Description

[0048] Figure 1 This is a flowchart illustrating the overall process of the power distribution control method for the hydrogen fuel cell power generation system of the present invention. Figure 2 This is the main control flowchart of the power distribution control method for the hydrogen fuel cell power generation system of the present invention; Figure 3 This is a flowchart of the method for adjusting the number of working fuel cell modules of the present invention; Figure 4 This is a flowchart of the phased power allocation method of the present invention; Figure 5 This is a flowchart of the start-stop transition power fine-tuning method of the present invention; Figure 6 This is a flowchart of the method for selecting the fuel cell stack module to be started according to the present invention; Figure 7 This is a flowchart of the method for selecting the fuel cell stack module to be shut down according to the present invention; Figure 8 This is a flowchart of the power allocation method for the candidate fuel cell stack module of the present invention. Detailed Implementation

[0049] The present invention will be further described below with reference to the accompanying drawings and specific embodiments. It should be understood that the following embodiments are for illustrative purposes only and are not intended to limit the scope of protection of the present invention. All equivalent substitutions, improvements, and variations made within the spirit and principles of the present invention should fall within the scope of protection of the present invention.

[0050] Working principle This invention applies to stationary hydrogen fuel cell power generation systems consisting of multiple fuel cell stack modules connected in parallel. Each fuel cell stack module has a corresponding power output capability, efficiency characteristics, and health status. The system controller determines the number of fuel cell stack modules participating in operation based on the system power demand generated by external loads and allocates the target power among the participating fuel cell stack modules.

[0051] like Figure 1 As shown, in each control cycle, the present invention first receives the power demand of the power generation system and determines whether it is necessary to increase or decrease the number of fuel cell stack modules participating in operation; after the number of working modules is adjusted, power allocation calculation is performed; then, further power fine-tuning is performed on the fuel cell stack modules that have just started or just exited to solve the problem of power instability during the start-up and exit process; finally, the target power of each fuel cell stack module is output to complete the control cycle.

[0052] like Figure 2 As shown, the main control flow of this invention further includes power demand anomaly identification, module status classification, handling of minimal power demand, working module adjustment, phased power allocation, start-stop transition power fine-tuning, power deficit compensation for fault-limited modules, and health and module status updates. Each control process is executed in a predetermined order, forming a closed loop of module start-stop scheduling, stable power allocation, transient power compensation, and health status feedback.

[0053] (1) Control parameters like Figure 2 As shown, at the beginning of each control cycle, the system controller acquires system-level operating parameters and operating parameters of each fuel cell stack module.

[0054] The operating parameters of each fuel cell stack module include current power p0, target power p1, lower limit of comfort zone p2, upper limit of comfort zone p3, maximum power p4, minimum power p5, current maximum allowable power p6, health status p7, power at the highest efficiency point p8, lower limit of efficiency range p9, and upper limit of efficiency range p10.

[0055] Among them, the lower limit of the comfort zone p2 and the upper limit of the comfort zone p3 constitute the comfort zone [p2, p3], and the lower limit of the efficiency zone p9 and the upper limit of the efficiency zone p10 constitute the efficiency zone [p9, p10].

[0056] System-level operating parameters include system power requirement a0, actual system power a1, working module list a2, and shutdown module list a3.

[0057] The current maximum allowable power p6 and the maximum power p4 have different meanings. The maximum power p4 represents the maximum output capability of the fuel cell module under normal conditions; the current maximum allowable power p6 represents the actual allowable output limit determined based on the current performance, degradation status, or fault-limited status of the fuel cell module.

[0058] When the fuel cell module is in normal condition, the current maximum allowable power p6 can be equal to the maximum power p4; when the fuel cell module experiences performance degradation but has not yet completely failed, the current maximum allowable power p6 is lower than the maximum power p4, and at this time, derating control is applied to the fuel cell module.

[0059] (2) Module status classification and main process initialization like Figure 2 As shown, the system controller classifies the status of each fuel cell module according to the working module list a2 and the shutdown module list a3 formed in the previous control cycle, forming a set of healthy working modules, a set of healthy shutdown modules, and a set of failed modules.

[0060] For a fuel cell module in the working module list a2, if its failure flag isFailed is false, it is added to the healthy working module set; if its failure flag isFailed is true, its target power p1 is set to zero and it is added to the failed module set.

[0061] For the fuel cell stack module in the shutdown module list a3, if its failure flag isFailed is false, it is added to the healthy shutdown module set; if its failure flag isFailed is true, its target power p1 is set to zero and it is added to the failed module set.

[0062] For fuel cell modules whose current maximum permissible power p6 is lower than the maximum power p4 but which still have power output capability, they are identified as fault-limited modules. Fault-limited modules are not directly disconnected from the power generation system, but continue to operate at reduced capacity within the current maximum permissible power p6 range.

[0063] The system controller also sets a fixed flag, isFixed, for each fuel cell stack module. When the target power of a fuel cell stack module cannot be increased further due to reaching the current maximum allowable power p6, its fixed flag isFixed is set to true; otherwise, its fixed flag is set to false.

[0064] (3) Identification of abnormal power demand like Figure 2 As shown, the system controller first determines whether the system power demand a0 is within the system's allowable power range.

[0065] In this embodiment, when the system power demand a0 is less than zero or greater than 1000, the power demand is determined to be abnormal. The 1000 is the upper limit of the power demand for the power generation system corresponding to this process. In hydrogen fuel cell power generation systems of different capacities, the corresponding upper limit value can be set according to the actual maximum allowable output power of the system.

[0066] When the system power demand is abnormal, the system controller performs a protective return and does not generate a target power command that exceeds the system output capacity or the power boundary of each fuel cell module.

[0067] The system controller also determines whether the sum of the current power p0 of each fuel cell module is within the allowable deviation range corresponding to the system power demand a0. In this embodiment, the determination condition is: |a1-a0|≤ΔP.

[0068] Wherein, ΔP is the preset allowable power deviation. When the absolute value of the difference between the actual power a1 of the system and the power demand a0 is not greater than the preset allowable power deviation ΔP, it indicates that the current actual output of the system is within the allowable deviation range corresponding to the power demand. The system controller maintains the current power allocation result to reduce repeated adjustments caused by small power fluctuations.

[0069] (4) Handling extremely low power requirements like Figure 2 As shown, after completing the anomaly identification, the system controller determines whether the system power demand a0 is lower than the minimum start threshold MIN POWER FOR START.

[0070] When the system power demand a0 is lower than the minimum start-up threshold, it indicates that the current power demand is insufficient to support the stable operation of at least one fuel cell stack module within its minimum power or suitable efficiency range.

[0071] In this case, the system controller sets the target power p1 of all healthy working modules to zero, transfers all healthy working modules to the healthy shutdown module set, and clears the healthy working module set.

[0072] Subsequently, the target power p1 of the healthy shutdown module is set to zero. The health status p7 is updated according to the operating status of each fuel cell module and the control cycle time increment within this control cycle. The healthy shutdown module and the failed module are merged to form a shutdown module list, and the updated working module list and shutdown module list are returned.

[0073] This process can prevent multiple fuel cell stack modules from operating in an inefficient, uncomfortable state under extremely low power requirements, thereby reducing ineffective hydrogen consumption and auxiliary system idling losses.

[0074] (5) Adjustment of the number of working modules When the system power demand a0 is not lower than the minimum startup threshold, the system controller follows... Figure 3 The process shown adjusts the number of fuel cell stack modules involved in the operation.

[0075] like Figure 3 As shown, the process of adjusting the number of working modules first sets the maximum number of iterations (max iter) to 10. In each iteration, the system controller records the number of working modules (len before) before the adjustment and calls the module selection process (select module to start).

[0076] If the module selection process returns a required fuel cell stack module, then that fuel cell stack module is removed from the healthy shutdown module set and added to the working module set.

[0077] If no fuel cell stack module needs to be added, the "select module tostop" procedure is invoked. If the "select module tostop" procedure returns a fuel cell stack module that needs to be removed, that fuel cell stack module is removed from the working module set and added to the healthy shutdown module set.

[0078] After completing an increase or decrease operation, the system controller determines whether the number of working modules has changed. If the number of working modules has changed and the number of iterations has not reached 10, the next round of adjustment continues; if the number of working modules no longer changes or the number of iterations reaches 10, the adjusted set of working modules and the set of healthy shutdown modules are returned.

[0079] Figure 3 The process shown addresses the question of "how many fuel cell modules should bear the current power demand," and is distinct from the subsequent phased power allocation process of "how much power each working module should bear."

[0080] (6) Selecting the module to be started exist Figure 3 During the adjustment process of the shown work modules, when it is necessary to determine whether to add a work module, the function is called. Figure 6 The process shown is to select the module to start.

[0081] like Figure 6As shown, healthy shutdown modules are first filtered based on the failure flag isFailed, and fuel cell stack modules that can start normally are added to the candidate module set candidates.

[0082] If the candidate module set is empty, return None, indicating that there is no bootable fuel cell module.

[0083] If the candidate module set is not empty, then the candidate modules are sorted in ascending order according to the following sort key: The first sort key is 1-p6 / p4; The second sort key is p2 / p4; The third sorting key is health level p7; The fourth sorting key is the module identifier id.

[0084] After sorting, initialize the candidate index i, and read the candidate module cand in sequence according to the sorting result.

[0085] When the current set of working modules is empty, the minimum power p5 of the candidate module is used as the required power for startup; when the current set of working modules is not empty, the lower limit of the comfort zone p2 of the candidate module is used as the required power for startup.

[0086] Subsequently, by combining the system power requirement a0, the sum of the lower limit of the comfort zone of the current working module, and the required power for the candidate module to start, it is determined whether the candidate module meets the startup conditions.

[0087] If a candidate module meets the startup conditions, the candidate module is returned; if it does not meet the conditions, the next candidate module is evaluated; if none of the candidate modules meet the startup conditions, None is returned.

[0088] pass Figure 6 The process shown can avoid adding new working modules when the system power demand is insufficient, thereby preventing the newly started modules from operating in an inefficient state below the minimum power or the lower limit of the comfort zone.

[0089] (7) Selecting modules to be stopped exist Figure 3 During the adjustment process of the shown work modules, if there are no work modules that need to be added, call... Figure 7 The process shown is for selecting modules to stop.

[0090] like Figure 7 As shown, the working modules are first filtered according to the failure flag isFailed, and the modules that can be processed according to the normal shutdown logic are added to the candidate module set candidates.

[0091] If the candidate module set is empty, return None, indicating that there are no selectable modules to be shut down.

[0092] If the candidate module set is not empty, then the candidate modules are sorted in ascending order according to the following sort key: The first sort key is p6 / p4; The second sort key is -p2 / p4; The third sort key is -p7; The fourth sorting key is the module identifier id.

[0093] After sorting, read the candidate modules (cand) sequentially according to the sorting results.

[0094] When the current number of working modules is 1, the power conditions that need to be met after shutdown are determined based on the minimum power p5 of that working module.

[0095] When the number of current working modules is greater than 1, the required power is determined by the sum of the lower comfort zone limits p2 of the remaining working modules after the candidate module exits.

[0096] Subsequently, the system controller, in conjunction with the system power requirement a0, the required power, and the preset hysteresis quantity HYSTERESIS, determines whether the candidate module meets the shutdown conditions.

[0097] If the shutdown condition is met, return the candidate module; if not, continue to the next candidate module; if none of the candidate modules meet the shutdown condition, return None.

[0098] The preset hysteresis is used to create an interval between module start-up and shutdown conditions, preventing the fuel cell module from frequently starting and stopping when the system power demand fluctuates near the critical value.

[0099] (8) Phased power allocation After adjusting the number of working modules, the system controller calls... Figure 4 The diagram shows the phased power allocation process.

[0100] like Figure 4 As shown, the minimum power p5 is first allocated to each working module in a predetermined order, and the target power p1 of each working module is set to the corresponding minimum power p5.

[0101] Then calculate the remaining power: Remaining power = a0 - Σp1.

[0102] If the remaining power is not greater than zero, then power allocation ends.

[0103] If there is still residual power, the allocation ratio of each working module is determined according to the health level p7, and the residual power is allocated to each working module so that the target power of each working module is increased to the lower limit of the comfort zone p2.

[0104] After each allocation, it is determined whether there is a fuel cell module whose target power exceeds the current maximum allowed power p6. If so, the target power of that module is limited to p6, and the remaining power is reallocated.

[0105] After completing the allocation of the lower limit of the comfort zone, the remaining power is recalculated. If there is still remaining power, it is further allocated to each working module according to the health level p7, so that the target power of each working module is increased to the highest efficiency point power p8.

[0106] Subsequently, the remaining power is allocated sequentially to the upper limit of the comfort zone p3, the upper limit of the efficiency zone p10, and the maximum power p4 in the same manner.

[0107] therefore, Figure 4 The power allocation process shown is executed step by step in the following order: Minimum power p5; Comfort zone lower limit p2; The highest efficiency point power is p8; Comfort zone upper limit p3; The upper limit of the efficiency range is p10; Maximum power p4.

[0108] In any allocation phase, as long as a stack module reaches the current maximum allowable power p6, the target power of that module is fixed, and the remaining power is redistributed to other working modules that still have power regulation margin.

[0109] (9) Power allocation for candidate modules Figure 4 The phased power allocation process shown Figure 5 The start-stop transition power fine-tuning process and the fault-limited module power deficit compensation process shown can both be invoked. Figure 8 The candidate module power allocation process is shown in the diagram.

[0110] like Figure 8 As shown, the candidate module power allocation process receives the candidate module list (candidates) and the power to be allocated (amount), and initializes the remaining power to be allocated (remaining) to amount.

[0111] Then, the candidate module list (candidates) is traversed sequentially.

[0112] If the candidate module list has been traversed, exit the loop and return the remaining power to be allocated.

[0113] When traversing the current candidate modules, first check if remaining is less than or equal to zero. If remaining is less than or equal to zero, it means that all the power to be allocated has been allocated, and the loop exits.

[0114] If remaining is greater than zero, then determine the potential increase in power capacity cap for the current candidate module and calculate: inc=min(remaining,cap).

[0115] Wherein, inc represents the power added to the current candidate module.

[0116] Then, update the target power of the current candidate module according to the following formula: p1 = p1 + inc.

[0117] Simultaneously update the remaining power to be allocated: remaining = remaining-inc.

[0118] After the update is complete, continue traversing the next candidate module until remaining is not greater than zero or all candidate modules have been traversed.

[0119] Figure 8 The process shown guarantees that the power added to any candidate module does not exceed the module's power capacity and can return the power not absorbed by the system's remaining capacity.

[0120] (10) Fine-tuning of start-stop transition power Finish Figure 4 After the phased power allocation shown, the system controller calls... Figure 5 The diagram illustrates the fine-tuning process of start-stop transition power allocation.

[0121] like Figure 5 As shown, the first step is to check if the set of working modules is empty. If the set of working modules is empty, the set of working modules is returned directly and the fine-tuning ends.

[0122] If the set of working modules is not empty, then exit transition power fine-tuning is performed.

[0123] The system controller filters out exit transition modules stopping where the target power p1 is zero but the current power p0 is greater than zero, and calculates the sum of the current power of all exit transition modules overshoot.

[0124] If there is no exit transition module, or if the overshoot does not reach the preset fine-tuning threshold, then no exit transition compensation will be performed.

[0125] If the overshoot reaches the preset fine-tuning threshold, then candidate modules with corresponding adjustment margins are selected from the working modules other than the exit transition module, and sorted according to 1-p6 / p4, p2 / p4, p7 and module identifier.

[0126] Then call Figure 8 The power allocation process for the candidate modules shown compensates for the actual output power that has not yet disappeared from the exiting transition module.

[0127] After completing the exit transition fine-tuning, the system controller performs the startup transition power fine-tuning.

[0128] The system controller selects startup transition modules whose target power p1 is greater than zero and whose current power p0 is lower than the minimum power p5, and calculates the power miss between the target power p1 and the current power p0 of each startup transition module.

[0129] If there is no startup transition module, or if missing has not reached the preset fine-tuning threshold, then startup transition compensation will not be performed.

[0130] If the missing threshold is reached, candidate modules are selected from working modules that have upward adjustment margin, excluding the start-up transition module. The allocation ratio is determined according to the health of the candidate modules, and they are sorted according to 1-p6 / p4, p2 / p4, p7 and module identifier.

[0131] Then call Figure 8 The power allocation process shown in the candidate module temporarily allocates the power shortage that the start-up transition module has not yet actually output to other working modules.

[0132] After fine-tuning, the system controller performs boundary protection: if the target power p1 of the working module is higher than the current maximum allowable power p6, the target power is limited to p6; if the target power p1 is greater than zero but lower than the minimum power p5, the target power is adjusted to p5.

[0133] pass Figure 5 The fine-tuning process shown can compensate for the actual output that has not yet disappeared from the exit module and the actual output that has not yet been established from the start module, thereby reducing power overshoot, power deficit and system load oscillation during the start-up and shutdown process.

[0134] (11) Power deficit compensation for fault-limited modules like Figure 2 and Figure 4As shown, when the target power of the fuel cell stack module exceeds its current maximum allowable power p6, the system controller limits the target power of the module to p6 and sets its fixed flag isFixed to true.

[0135] The difference between the target power before the limitation and p6 constitutes the power deficit.

[0136] The system controller selects candidate modules with increased power capacity from other working modules whose fixed flag is false, and calls them. Figure 8 The power allocation process shown in the candidate module allocates the power deficit to other working modules.

[0137] like Figure 8 If the remaining time returned by the process is zero, it means that the power deficit has been fully compensated; if the remaining time is greater than zero, it means that all available working modules have reached their corresponding power boundaries, and the current available capacity of the system cannot fully meet the system power requirements.

[0138] In this case, the system controller retains the uncompensated power deficit and does not output any command to any fuel cell module that exceeds the current maximum allowable power p6 or maximum power p4.

[0139] This mechanism allows fuel cell modules that have degraded in performance but have not yet completely failed to continue operating at a safe derated rate, while other healthy operating modules compensate for the power they cannot handle.

[0140] (12) Target power output and health update like Figure 2 As shown, after completing power distribution, start-stop transition power fine-tuning, and fault-limited module power deficit compensation, the system controller sets the target power p1 of the shutdown module to zero.

[0141] Subsequently, the health status p7 of each fuel cell stack module is updated based on the set of working modules, the set of shutdown modules, and the control cycle time increment.

[0142] Health status p7 is used to characterize the total workload, degradation rate, or overall health status of the fuel cell stack module. The health status evaluation result is updated at the end of the current control cycle and serves as the input for the ordering of modules to be started, the ordering of modules to be shut down, and the phased power allocation in the next control cycle.

[0143] After completing the health update, the system controller merges the healthy shutdown modules and failed modules to form a complete shutdown module set. It updates the fixed flag isFixed based on whether the current maximum allowable power p6 of each stack module is lower than the maximum power p4, and returns the updated list of working modules and the list of shutdown modules.

[0144] thus, Figure 1The overall control process shown forms a closed loop: the current health status of the fuel cell module affects the module start-up, shutdown, and power distribution in this control cycle; the running time and power output of this control cycle are used to update the health status; and the updated health status continues to participate in the control decision-making of the next control cycle.

[0145] Example 1: Phased Power Distribution under Increased Load Conditions This embodiment illustrates how the system controller adds working modules when the system power demand increases, and how it allocates power in stages based on health status, comfort zone, and peak efficiency point.

[0146] In this embodiment, the hydrogen fuel cell power generation system includes multiple stack modules connected in parallel. At the start of the control cycle, the first and second stack modules are in a healthy operating state, while the third and fourth stack modules are in a healthy shutdown state, and the failure flag isFailed for each stack module is false.

[0147] The first to fourth fuel cell stack modules have different health levels p7, comfort zones [p2, p3], peak efficiency point power p8, and current maximum allowable power p6. The current maximum allowable power p6 for each fuel cell stack module is not lower than the target power expected to be allocated to the corresponding module within this control cycle; therefore, there are no fault-limited modules that require immediate derating.

[0148] like Figure 1 and Figure 2 As shown, the system controller obtains the system power demand a0 and the operating parameters of each fuel cell module for the current control cycle, and first determines whether the system power demand a0 is within the allowable range.

[0149] When the system power demand a0 does not exceed the maximum allowable power demand of the system, and the deviation between the actual system power a1 and the system power demand a0 is not within the allowable deviation range, the system controller enters the process of adjusting the number of working modules.

[0150] Because the system power demand in this control cycle increases compared to the previous control cycle, the output power of the first and second fuel cell stack modules is insufficient to meet the system power demand within their respective power boundaries. Therefore, if Figure 3 As shown, the system controller determines that an additional working module needs to be added and calls... Figure 6 The process of selecting modules to be started is shown.

[0151] The system controller first excludes modules with the failure flag "isFailed" true from the healthy shutdown modules. Since neither the third nor the fourth fuel cell stack module has failed, they are selected as candidate modules to be started.

[0152] Subsequently, the system controller sorts the third and fourth fuel cell stack modules according to 1-p6 / p4, p2 / p4, health status p7, and module identifier.

[0153] Among them, the ratio of the current maximum allowable power p6 to the maximum power p4 is used to reflect the available output capability of the fuel cell stack module, the ratio of the comfort zone lower limit p2 to the maximum power p4 is used to reflect the minimum load level required for the fuel cell stack module to enter the comfort zone, and the health status p7 is used to reflect the current health status of the fuel cell stack module.

[0154] With the first and second fuel cell stacks already in operation, the system controller uses the lower limit of the comfort zone p2 of the candidate module as the starting power that the new module needs to achieve.

[0155] When the system power demand a0 is sufficient to support the existing working modules and the third fuel cell stack module to meet the corresponding power requirements, the system controller selects the third fuel cell stack module as the module to be started and transfers the third fuel cell stack module from the healthy shutdown module set to the healthy working module set.

[0156] After a module addition is completed, the system controller reassesses whether further additions of working modules are needed. If the first, second, and third fuel cell stack modules are already able to meet the system power requirements within their respective allowable power ranges, then the fourth fuel cell stack module will not be started, and the adjustment of the number of working modules will end.

[0157] Therefore, the healthy operating modules of this control cycle include the first fuel cell stack module, the second fuel cell stack module, and the third fuel cell stack module, while the fourth fuel cell stack module remains in a healthy shutdown state.

[0158] Subsequently, the system controller followed Figure 4 The process shown involves phased power allocation.

[0159] First, set the target power p1 of the first, second, and third fuel cell stack modules to their respective minimum power p5, and then calculate: Remaining power = System power demand a0 - Sum of target power p1 from the first stack module to the third stack module.

[0160] If there is still residual power, the system controller allocates the residual power according to the health status p7 of each healthy working module, so that the target power of the first to third fuel cell stack modules is increased from the minimum power p5 to the lower limit of the comfort zone p2.

[0161] When all three working modules have reached their respective comfort zone lower limit p2 and there is still residual power, the system controller continues to allocate the residual power according to the health status p7, so that the target power of each working module is increased towards the highest efficiency point power p8.

[0162] In this embodiment, assuming that the health status of the first fuel cell stack module is better than that of the second fuel cell stack module, and the health status of the second fuel cell stack module is better than that of the newly started third fuel cell stack module, the system controller will assign a higher power allocation weight to the module with the better health status according to a unified health evaluation direction.

[0163] Therefore, provided that the power limit of the corresponding stage is not exceeded, the first fuel cell module can handle a relatively large amount of output power, the second fuel cell module can handle a smaller amount, and the third fuel cell module can handle a relatively small amount of output power in the initial stage of startup.

[0164] When each healthy operating module reaches its maximum efficiency point power p8 and still has residual power, the system controller continues to increase the target power step by step in the order of comfort zone upper limit p3, efficiency zone upper limit p10, and maximum power p4.

[0165] At any stage, if a fuel cell module reaches its current maximum allowable power p6, the system controller limits the module's target power to p6 and sets its fixed flag isFixed to true; the remaining power is then released. Figure 8 The power allocation process for the candidate modules shown is used to redistribute them to other unfixed working modules.

[0166] After the phased power allocation under steady state is completed, since the third stack module is a newly started module in this control cycle, its target power p1 is already greater than zero, but the current power p0 may not have reached the minimum power p5.

[0167] Therefore, as Figure 5 As shown, the system controller identifies the third fuel cell stack module as a startup transition module and calculates the power deficit between the target power p1 and the current power p0 of the third fuel cell stack module.

[0168] When the power deficit reaches the preset fine-tuning condition, the system controller selects a compensation module with upward adjustment margin from the first and second fuel cell stack modules, and temporarily increases the target power of the compensation module to compensate for the output power that the third fuel cell stack module has not yet actually established.

[0169] As the current power p0 of the third fuel cell module gradually increases, the temporary compensation power undertaken by the first and second fuel cell modules decreases accordingly until each module reaches the stable target power determined by the phased power allocation.

[0170] This embodiment first adjusts the number of working modules, then allocates them sequentially according to minimum power, lower limit of the comfort zone, highest efficiency point, upper limit of the comfort zone, upper limit of the efficiency range, and maximum power. This ensures that the fuel cell stack modules put into operation are preferentially positioned within a suitable efficiency range. Simultaneously, incorporating health status into the power allocation at each stage helps avoid uneven degradation caused by indiscriminately averaging output across all fuel cell stack modules.

[0171] Example 2: Reduced load and minimal power demand conditions This embodiment illustrates the process of reducing the number of working modules when the system power demand decreases, and the overall shutdown process when the system power demand falls below the minimum startup threshold.

[0172] In this embodiment, at the start of the control cycle, the first fuel cell stack module, the second fuel cell stack module, and the third fuel cell stack module are all in a healthy working state, while the fourth fuel cell stack module is in a healthy shutdown state.

[0173] As the external load decreases, the system power demand a0 decreases. The system controller follows... Figure 2 The process shown reads the system power demand a0 and determines whether it is within the system's allowed power range.

[0174] When the system power demand a0 is still higher than the minimum start-up threshold, but the three healthy working modules are sharing the power demand, they need to operate in a state close to or below their respective comfort zone lower limit p2. The system controller determines that there are too many working modules and needs to reduce the number of working modules.

[0175] like Figure 3 As shown, during the adjustment of working modules, the system controller first determines that there are no working modules that need to be added, and then calls... Figure 7 The process of selecting modules to be shut down is shown.

[0176] The system controller excludes modules from the first to the third fuel cell stacks whose failure flag isFailed is true, and sorts the remaining candidate modules according to p6 / p4, -p2 / p4, -p7 and module identifier.

[0177] The system controller determines, according to the sorting results, whether the remaining working modules can still meet the system power requirement a0 after the candidate modules are removed.

[0178] When a candidate module exits, if the remaining working module can meet the system power requirements without exceeding the current maximum allowable power p6, and can maintain the corresponding minimum power p5 or the lower limit of the comfort zone p2, the system controller will identify the candidate module as a module to be shut down.

[0179] When determining whether a candidate module should be shut down, the system controller also introduces a preset hysteresis variable, HYSTERESIS. A candidate module is only allowed to exit when the system power demand and the power capability of the remaining working modules meet the shutdown conditions including the hysteresis variable.

[0180] Assuming the third fuel cell module is in accordance with Figure 7 If the sorting result shown is determined to be a module to be shut down, the system controller will transfer the third fuel cell stack module from the set of healthy working modules to the set of healthy shutdown modules and set its target power p1 to zero.

[0181] Subsequently, the system controller determines again whether the number of working modules needs to be adjusted. If the first and second fuel cell stack modules can already meet the system power requirements within a suitable power range, the adjustment of the number of working modules is terminated.

[0182] When the third fuel cell module receives the shutdown command, its target power p1 is already zero, but the current power p0 may still be greater than zero.

[0183] like Figure 5 As shown, the system controller identifies the third fuel cell stack module as an exit transition module and calculates the current power p0 of the third fuel cell stack module before it disappears.

[0184] When the current power of the exit transition module reaches the preset fine-tuning condition, the system controller selects a compensation module with downward adjustment margin from the first and second fuel cell stack modules, and reduces its target power accordingly to offset the actual output of the third fuel cell stack module that still exists during the shutdown transition.

[0185] As the current power of the third fuel cell module gradually drops to zero, the target power of the first and second fuel cell modules gradually recovers to a stable distribution.

[0186] When the external load continues to decrease and the system power demand a0 falls below the minimum start-up threshold MIN_POWER_FOR_START, the system controller will no longer maintain any fuel cell stack module in operation.

[0187] like Figure 2 As shown, the system controller sets the target power p1 of the first and second fuel cell stack modules to zero, transfers it to the healthy shutdown module set, and keeps the third and fourth fuel cell stack modules in a healthy shutdown state.

[0188] Before the end of the control cycle, the system controller updates the health status p7 of the first and second fuel cell stack modules based on their actual operating time and power output before shutdown, and forms an empty working module list a2 and a shutdown module list a3 including each healthy shutdown module.

[0189] In the next control cycle, the system controller will only follow the start-up conditions when the system power demand meets the restart requirements again. Figure 6 The process shown in the diagram allows you to reselect the fuel cell stack module that needs to be put into operation.

[0190] This embodiment sets a minimum start-up threshold, causing the system to stop operating the fuel cell stack modules under extremely low power demand conditions. This avoids multiple fuel cell stack modules operating in an inefficient, uncomfortable state for extended periods. By setting a hysteresis condition between the start-up and shutdown conditions, frequent start-ups and shutdowns occurring when the load fluctuates near the critical value can be reduced.

[0191] Example 3: Fault-limited and start-stop transient compensation conditions This embodiment illustrates how the system controller derating a fuel cell module when it experiences performance degradation but has not yet completely failed, and how other healthy operating modules compensate for the power deficit.

[0192] In this embodiment, the first fuel cell stack module, the second fuel cell stack module, and the third fuel cell stack module are all in working condition, while the fourth fuel cell stack module is in a healthy shutdown state.

[0193] Among them, the first and second fuel cell stack modules are in normal condition, and their current maximum allowable power p6 is equal to or close to their respective maximum power p4.

[0194] The third fuel cell module has experienced performance degradation but still retains the ability to continue outputting power. The current maximum allowable power p6 of the third fuel cell module is lower than its maximum power p4, and its failure flag isFailed is false.

[0195] like Figure 2 As shown, the system controller identifies the third fuel cell module as a fault-limited module based on the current maximum allowable power p6 and maximum power p4 of the third fuel cell module, rather than cutting it off from the system as a completely failed module.

[0196] During the phased power allocation process, the system controller still allows the third fuel cell stack module to participate in power allocation. When the target power p1 of the third fuel cell stack module calculated according to its health status and phase power limit does not exceed its current maximum allowable power p6, the third fuel cell stack module operates according to the calculated target power.

[0197] When the calculated target power p1 of the third fuel cell stack module exceeds its current maximum allowable power p6, the system controller limits the target power of the third fuel cell stack module to p6 and sets the fixed flag isFixed of the third fuel cell stack module to true.

[0198] Subsequently, the system controller calculates the difference between the target power before the limitation of the third stack module and the current maximum allowable power p6, and uses this difference as the power deficit to be compensated.

[0199] The system controller determines the compensation candidate modules from the first and second fuel cell modules where the fixed flag isFixed is false, and calculates the increaseable power capacity of the first and second fuel cell modules respectively.

[0200] like Figure 8 As shown, the power deficit to be compensated is taken as the power to be allocated, and the remaining power to be allocated, remaining, is initialized to amount.

[0201] The system controller first reads a compensation candidate module and determines its potential power capacity cap based on the difference between the module's stage power limit and the current target power.

[0202] Then calculate: inc=min(remaining,cap).

[0203] The system controller increases the power increment inc to the target power p1 of the compensation candidate module, and makes: remaining=remaining-inc.

[0204] If remaining is still greater than zero, the remaining power will continue to be allocated to another compensation candidate module; if remaining is not greater than zero, the power deficit compensation will end.

[0205] Therefore, the third fuel cell module can continue to output power within the safe range of the current maximum allowable power p6, and the portion it cannot bear is compensated by the first and second fuel cell modules.

[0206] If the first and second fuel cell stack modules reach their respective maximum allowable power and the remaining power is still greater than zero, the system controller retains the uncompensated power and does not output control commands to any fuel cell stack module that exceed its current maximum allowable power p6 or maximum power p4.

[0207] Furthermore, if the fourth fuel cell module needs to be started due to an increase in external load during this control cycle, the system controller will proceed according to... Figure 6 The process shown adds the fourth fuel cell stack module to the set of working modules.

[0208] Because the current power p0 of the fourth fuel cell module is lower than its target power p1 at the initial startup stage, the system controller follows... Figure 5 The fine-tuning process shown calculates the startup power deficit of the fourth fuel cell module, and the first and second fuel cell modules temporarily assume this power deficit.

[0209] As the current power of the fourth fuel cell module gradually increases, the amount of temporary compensation undertaken by the first and second fuel cell modules gradually decreases; the third fuel cell module is always limited to its current maximum allowable power p6.

[0210] This embodiment differentiates between fault-limited modules and completely failed modules, allowing fuel cell modules with degraded performance but not yet completely failed to continue operating at a reduced derating rate, while other healthy modules compensate for their power deficit. Compared to directly disconnecting the fuel cell module completely, this control process helps preserve the system's available output capacity and power supply continuity.

[0211] Meanwhile, by temporarily compensating for the actual power deficit of the newly started module, the system power drop when fault derating and module startup occur simultaneously can be reduced. This invention uses derating operation, deficit compensation, and transient fine-tuning during start-up and shutdown as the main measures to enhance the system's online fault tolerance and operational stability.

[0212] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0213] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0214] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A power distribution control method for a hydrogen fuel cell power generation system, the system comprising multiple parallel fuel cell stack modules, characterized in that, include: S101, obtain power requirements and operating parameters of each fuel cell stack module; S102, perform demand anomaly judgment, and classify each fuel cell stack module into healthy working module, healthy shutdown module, fault-limited module and failed module according to the failure flag, operating status and current maximum allowable power of each fuel cell stack module; S103, add or remove working modules according to power requirements; S104: Starting from the minimum power, the target power of the working module is allocated sequentially according to the lower limit of the comfort zone, the highest efficiency point, the upper limit of the comfort zone, the upper limit of the efficiency zone, and the maximum power, and the allocation amount for each stage is determined based on the health status; the target power of the module that has reached the current maximum allowable power is fixed, and the remaining power is redistributed. S105, compensate for power fluctuations based on the difference between the current power and the target power of the start-stop transition module; S106 outputs the target power and updates the health and module status.

2. The power distribution control method according to claim 1, characterized in that, The operating parameters include: current power p0, target power p1, comfort zone consisting of lower comfort zone limit p2 and upper comfort zone limit p3, maximum power p4, minimum power p5, current maximum allowable power p6, health status p7, power at the highest efficiency point p8, and efficiency range consisting of lower efficiency range limit p9 and upper efficiency range limit p10; the parameters of the hydrogen fuel cell power generation system include power demand a0, actual power a1, working module list a2, and shutdown module list a3, wherein the actual power a1 is the sum of the current power p0 of each stack module.

3. The power distribution control method according to claim 1, characterized in that, Step S102 includes: S201, determine whether the power demand is within the system's allowed power range. If the power demand exceeds the power range, perform a protective return and do not output the target power that exceeds the system's capacity. S202, If the deviation between the actual power of the system and the power demand is within the preset allowable deviation range, then the current power allocation result is maintained; S203, if the power demand is lower than the minimum start-up threshold, the target power of the healthy working module is set to zero, and the healthy working module is transferred to the healthy shutdown state; S204 sets the target power of the fuel cell stack module with the failure flag to zero.

4. The power distribution control method according to claim 1, characterized in that, Step S102 also includes: S211, classify fuel cell stack modules that do not have a failure flag, are in operation, and whose current maximum allowed power p6 is equal to the maximum power p4 as healthy working modules; S212, classify fuel cell stack modules that do not have a failure flag and are in a shutdown state as healthy shutdown modules; S213, classify fuel cell stack modules that do not have a failure indicator, are in operation, have a current maximum allowable power p6 lower than the maximum power p4, and still have power output capability as fault-limited modules; S214, the fuel cell stack module with failure flag is classified as a failure module, and the target power p1 of the failure module is set to zero; S215, when the target power p1 of the fault-limited module reaches or exceeds the current maximum allowable power p6, its target power p1 is limited to p6, and its fixed flag isFixed is set to true.

5. The power distribution control method according to claim 1, characterized in that, Step S103 includes: S301 records the number of currently working modules; S302, determine whether an additional working module is needed based on power requirements. If so, select a module to be started from the health shutdown module and transfer the module to be started to the working module list. S303, when there is no need to add working modules, determine whether it is necessary to reduce working modules. If so, select the module to be stopped from the working module list and transfer the module to be stopped to the healthy shutdown module. S304. Compare the number of working modules before and after the adjustment. If the number of working modules changes, repeat steps S301 to S303 until the number of working modules no longer changes or the number of iterations reaches ten.

6. The power distribution control method according to claim 5, characterized in that: When selecting a module to be started, healthy shutdown modules with failure flags are excluded. The remaining healthy shutdown modules are sorted in ascending order using 1-p6 / p4, p2 / p4, p7 and module identifier as the first to fourth sorting keys. When the working module list is empty, the minimum power p5 of the candidate module is used as the power required for startup. When the working module list is not empty, the lower limit of the comfort zone p2 of the candidate module is used as the power required for startup. Candidate modules that meet the power requirements are selected according to the sorting results. When selecting modules to be shut down, modules with failure flags are excluded. The remaining modules are sorted in ascending order using p6 / p4, -p2 / p4, -p7 and module identifier as the first to fourth sorting keys. Based on the minimum power or the sum of the lower limit of the comfort zone of the remaining modules after the candidate modules are removed, the power requirements, and the preset hysteresis, the candidate modules that meet the shutdown conditions are selected according to the sorting results.

7. The power distribution control method according to claim 1, characterized in that, Step S104 includes: S401, set the target power of each working module to the corresponding minimum power p5, and calculate the remaining power between the power requirement and the sum of the target power of each working module; S402, allocate the remaining power according to the health status p7 of each working module, so that the target power of each working module is increased from the minimum power p5 to the lower limit of the comfort zone p2; S403, when there is still residual power, power is allocated according to the health status p7 of each working module, so that the target power of each working module is successively increased to the highest efficiency point power p8, the upper limit of the comfort zone p3, the upper limit of the efficiency range p10 and the maximum power p4. S404, in any allocation phase, if the target power of the fuel cell module reaches the current maximum allowable power p6, then the target power of the fuel cell module is fixed at p6, and the remaining power is redistributed to other working modules with power regulation margin.

8. The power distribution control method according to claim 1, characterized in that, Step S105 includes: S501, determine the exit transition module where the target power p1 is zero and the current power p0 is greater than zero, and calculate the sum of the current powers of the exit transition modules; S502, when the sum of the current power reaches the preset fine-tuning threshold, a candidate module is determined from the working modules other than the exit transition module and which have downward adjustment margin, and the target power of the candidate module is reduced. S503, determine the start-up transition module where the target power p1 is greater than zero and the current power p0 is lower than the minimum power p5, and calculate the power deficit between the target power and the current power of the start-up transition module; S504, when the power deficit reaches a preset fine-tuning threshold, a candidate module is determined from the working modules other than the start-up transition module that have an upward adjustment margin, and the target power of the candidate module is increased. S505 limits the target power that is higher than the current maximum allowable power p6 to p6, and adjusts the target power that is greater than zero and lower than the minimum power p5 to p5.

9. The power distribution control method according to claim 1, characterized in that, Compensation for power deficit caused by faulty modules includes: S511, limit the target power of the fault-limited module to its current maximum allowable power p6, and calculate the power deficit caused by the limitation; S512, determine candidate modules from the working modules where the fixed flag isFixed is false, and determine the increaseable power capacity of each candidate module; S513, Set the power to be compensated to the power deficit, and iterate through each candidate module in turn; S514, take the smaller value between the power to be compensated and the increaseable power capacity of the current candidate module as the power increment, increase the power increment to the target power of the current candidate module, and deduct the power increment from the power to be compensated; S515, Repeat step S514 until the power to be compensated is not greater than zero or all candidate modules have been traversed.

10. The power distribution control method according to claim 1, characterized in that, Step S106 includes: S601 sets the target power of the health shutdown module and the fuel cell stack module with failure flag to zero; S602, based on the operating status, power output, and control cycle time increment of each fuel cell module in the current control cycle, update the health status p7 of each fuel cell module; S603, based on the updated health status p7, failure flag and current maximum allowable power p6, update the healthy working module, healthy shutdown module and fault-limited module; S604, generate a working module list a2 and a shutdown module list a3 based on the updated module status, and update the fixed flag isFixed according to whether the target power p1 of each stack module has reached the current maximum allowable power p6; when the target power p1 reaches the current maximum allowable power p6, set the fixed flag isFixed to true, otherwise set the fixed flag isFixed to false. S605 uses the updated health status p7, working module list a2, shutdown module list a3, and fixed flag isFixed for power allocation in the next control cycle.

Images