An active thermal management method and system based on fuel cell waste heat

Abstract

The application discloses a kind of active thermal management method and system based on fuel cell waste heat, method includes: the closed loop heat transport path consisting of main circuit and multiple distributed branch is constructed, the closed loop heat transport path will fuel cell stack as first heat exchange end, multiple distributed heat user as second heat exchange end;Real-time acquisition the running heat state of the first heat exchange end, the multidimensional operating state characteristic parameter of each second heat exchange end including health state class data, the fluid dynamics parameter of the closed loop heat transport path;Obtain the task reliability prediction value of each second heat exchange end under current operating condition;According to the task reliability prediction value, dynamically correct the safety temperature threshold and heat distribution weight of each second heat exchange end;Waste heat is distributed to each second heat exchange end from the first heat exchange end through each distributed branch.Utilize the scheme of the application, the efficient centralized management and on-demand accurate distribution of fuel cell waste heat are realized.

Description

Technical Field

[0001] This application generally relates to the field of fuel cell auxiliary system technology. More specifically, this application relates to an active thermal management method and system based on fuel cell waste heat. Background Technology

[0002] Proton exchange membrane fuel cells generate a significant amount of waste heat during the electrochemical reaction that produces electricity. Existing fuel cell thermal management systems primarily focus on the heat dissipation safety of the fuel cell stack itself, typically using coolant circulation to dissipate heat from the radiator to the external environment to prevent the stack from overheating.

[0003] However, in practical applications, especially under low-temperature and frigid conditions, existing fuel cell thermal management technologies have the following significant shortcomings: First, energy utilization efficiency is low: existing systems treat the waste heat of the fuel cell stack as "waste heat" and discharge it directly, while distributed heat users in the system (such as servo control mechanisms, sensors, electronic devices, etc.) often need to consume additional electrical energy for electric heating to maintain performance. This "discharging heat on one hand and consuming electricity to supply heat on the other" mode significantly increases the parasitic power consumption of the fuel cell system.

[0004] Secondly, the thermal network topology is fragmented: the cooling circuit of the fuel cell and the thermal protection circuit of the external load are usually independent of each other, lacking a unified fluid scheduling network and directional allocation mechanism from the heat source (stack) to the heated object (distributed load).

[0005] Third, the scheduling strategy is disconnected from reliability assessment: existing fluid control is mostly based on simple temperature feedback, without considering the performance degradation degree and task sensitivity of different heat users. When short-term temperature control requirements (such as rapid cooling of the fuel cell stack) conflict with long-term reliability requirements (such as stable heating of heated components), the system lacks an effective control game and priority arbitration mechanism, making it difficult to achieve efficient and adaptive heat distribution under complex dynamic conditions.

[0006] In view of this, there is an urgent need to provide an active thermal management solution based on fuel cell waste heat to solve the above problems. This solution can coordinate the recovery of fuel cell stack waste heat, fine-tuning of fluids and reliability assessment of components, thereby improving the overall energy efficiency and operational safety of fuel cell systems in complex environments. Summary of the Invention

[0007] In order to at least address one or more of the technical problems mentioned above, this application proposes an active thermal management scheme based on fuel cell waste heat in several aspects.

[0008] In a first aspect, this application provides an active thermal management method based on fuel cell waste heat, comprising: constructing a closed-loop heat transfer path consisting of a main loop and multiple distributed branches, wherein the closed-loop heat transfer path uses the fuel cell stack as a first heat exchange end and multiple distributed heat users as second heat exchange ends; real-time acquisition of the operating thermal state of the first heat exchange end, multi-dimensional operating state feature parameters including health state data of each second heat exchange end, and fluid dynamic parameters of the closed-loop heat transfer path; using the health state data in the multi-dimensional operating state feature parameters as observation input, performing online posterior update of the initial reliability prior model of each second heat exchange end based on a Bayesian inference framework to obtain the task reliability prediction value of each second heat exchange end under the current operating conditions; dynamically correcting the safety temperature threshold and heat allocation weight of each second heat exchange end according to the task reliability prediction value; and, based on the corrected heat allocation weight, distributing waste heat from the first heat exchange end to each second heat exchange end through each distributed branch, provided that the thermal safety boundary of the first heat exchange end is satisfied.

[0009] In a second aspect, this application provides an active thermal management system based on fuel cell waste heat, employing the active thermal management method based on fuel cell waste heat as described in any embodiment of the first aspect. The system includes: a thermal management hardware topology module for constructing a closed-loop heat transfer path consisting of a main loop and multiple distributed branches, wherein the closed-loop heat transfer path uses the fuel cell stack as a first heat exchange end and multiple distributed heat users as second heat exchange ends; and a multi-dimensional status monitoring and acquisition module for real-time acquisition of the operating thermal status of the first heat exchange end, multi-dimensional operating status characteristic parameters including health status data of each second heat exchange end, and the fluid dynamics of the closed-loop heat transfer path. The system includes: a reliability assessment and prediction computing module, which takes the health status data from the multi-dimensional operating state feature parameters as observation input and performs online posterior updates to the initial reliability prior model of each second heat exchanger based on a Bayesian inference framework to obtain the predicted task reliability value of each second heat exchanger under the current operating conditions; a control decision and parameter correction module, which dynamically corrects the safety temperature threshold and heat distribution weight of each second heat exchanger based on the predicted task reliability value; and a waste heat distribution execution module, which, based on the corrected heat distribution weight, distributes waste heat from the first heat exchanger to each second heat exchanger through each distributed branch, provided that the thermal safety boundary of the first heat exchanger is met.

[0010] Through the active thermal management scheme based on fuel cell waste heat provided above, this embodiment of the application achieves efficient centralized management and precise on-demand allocation of fuel cell waste heat by constructing a distributed closed-loop heat transfer path. By deeply integrating multi-dimensional real-time monitoring data with a Bayesian inference framework, the system can iterate the reliability model online based on the actual health status characteristics of heat users, thereby enabling the system to anticipate potential risks. By directly mapping the predicted reliability value of the task to the dynamic correction of the safety temperature threshold and weights, the system breaks through the limitations of rigid boundaries in traditional thermal management. It can proactively identify and prioritize the thermal needs of high-risk or critical task components, significantly improving the long-term operating performance, task assurance capability, and overall energy efficiency of the entire system under complex and variable operating conditions while ensuring the thermal safety boundary of the fuel cell stack itself.

[0011] Furthermore, in some embodiments, a logical hierarchical architecture based on mission reliability thresholds is constructed, achieving deep decoupling and flexible switching between conventional energy efficiency control and targeted reliability assurance. First, differentiated and precise allocation of thermal resources is achieved. Differentiated reliability enhancement strategies are formulated for thermal users with different characteristics, such as servos, hydrogen supply valves, and flight control payloads. By dynamically adjusting weights and boundary parameters, the functional stability of key components under extreme conditions is ensured. Second, a multi-source thermal cascade compensation mechanism is established. By real-time monitoring of the heating rate and thermal demand gap, efficient linkage between fuel cell waste heat and auxiliary heating units is achieved, effectively solving the bottleneck problem of insufficient supply from a single heat source during low-temperature startup or low-power operation. Finally, an optimal balance between system safety and energy efficiency is achieved. While ensuring the reliability of flight missions, the precise locking logic of auxiliary heating units minimizes additional energy consumption. This full-condition, multi-branch closed-loop control system significantly improves the mission robustness, thermal environment adaptability, and global energy management efficiency of the UAV system in complex environments.

[0012] Furthermore, in some embodiments, a dual logical system encompassing transient energy efficiency optimization and long-term health management is established by constructing a multi-dimensional comprehensive energy efficiency index that integrates waste heat recovery rate, temperature control efficiency, and system power consumption. Regarding energy efficiency improvement, through real-time monitoring of the first deviation condition, the system can keenly capture redundant power consumption during operation and, while ensuring thermal safety thresholds, adaptively lower fluid dynamic parameters to achieve dynamic precision control of system power consumption and maximize operational economy. In terms of reliability maintenance, this logic possesses deep trend perception capabilities. Utilizing long-term correlation analysis of historical data based on the second deviation condition, it can proactively identify latent faults such as heat exchange aging, physical blockage, and material degradation from subtle signs of performance degradation. This ensures the system health and continuous combat capability of the UAV under long-term, high-intensity missions.

[0013] Furthermore, by constructing a dynamic thermal resource scheduling algorithm based on mission phase awareness, the allocation of residual heat has been transformed from a passive response to a proactive, scenario-driven, and task-prioritized approach. First, a survivability guarantee mechanism for multi-mission scenarios has been established. During high-risk phases such as takeoff, maneuvering, or attitude adjustment, nodes related to flight control safety are automatically identified as the highest priority, ensuring the thermal reliability of core actuators under extreme dynamic loads. Second, deep optimization of payload performance and operational quality has been achieved. When performing observation or signal processing tasks, precise thermal orientation of core temperature control nodes such as payload sensors ensures signal acquisition accuracy and equipment stability. In addition, global thermal balance control capabilities are provided. During stable operation phases such as cruise or low load, the strategy shifts from directional impact to balanced coverage, aiming to minimize temperature differences between branches, effectively eliminating local thermal stress and improving the overall lifespan of the system. Attached Figure Description

[0014] The above and other objects, features, and advantages of exemplary embodiments of this application will become readily understood by reading the following detailed description with reference to the accompanying drawings. In the drawings, several embodiments of this application are illustrated by way of example and not limitation, and the same or corresponding reference numerals denote the same or corresponding parts, wherein: Figure 1 An exemplary flowchart of an active thermal management method based on fuel cell waste heat according to an embodiment of this application is shown; Figure 2 An exemplary flowchart illustrating an embodiment of this application is shown, showing the predicted reliability values ​​of each second heat exchanger under current operating conditions. Figure 3 An exemplary flowchart illustrating the execution of the full lifecycle rolling update mechanism according to an embodiment of this application is shown; Figure 4 An exemplary flowchart illustrating an embodiment of this application shows how to dynamically adjust the safety temperature threshold and heat allocation weight of each second heat exchanger based on the predicted mission reliability value. Figure 5 An exemplary flowchart illustrating the implementation of short-term power consumption optimization and long-term fault self-diagnosis strategies according to an embodiment of this application is shown; Figure 6 An exemplary flowchart illustrating the implementation of a cascaded heat compensation strategy to address insufficient residual heat at the first heat exchange terminal, according to an embodiment of this application, is shown. Figure 7 An exemplary flowchart illustrating how waste heat is distributed from the first heat exchange terminal to each second heat exchange terminal via distributed branches, according to an embodiment of this application, is shown. Figure 8 An exemplary structural block diagram of an active thermal management system based on fuel cell waste heat according to an embodiment of this application is shown. Detailed Implementation

[0015] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0016] It should be understood that the terms "comprising" and "including" used in the specification and claims of this application indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0017] It should also be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application. As used in this specification and claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this specification and claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes such combinations.

[0018] Figure 1 An exemplary flowchart of an active thermal management method 100 based on fuel cell waste heat according to an embodiment of this application is shown.

[0019] like Figure 1 As shown, in step S110, a closed-loop heat transfer path consisting of a main loop and multiple distributed branches is constructed. The closed-loop heat transfer path uses the fuel cell stack as the first heat exchange end and multiple distributed heat users as the second heat exchange end.

[0020] In the embodiments of this application, during the construction of the main circuit, a main pipeline with a cooling medium circulation pump as the power core is established, connecting the outlet of the first heat exchange end with the inlet of the distributed branch circuit. The main circuit includes an expansion compensation structure (such as an expansion tank) to absorb pressure changes in the cooling medium caused by temperature fluctuations, ensuring the fluid dynamic stability of the system under extreme low temperature (severe cold regions) and high heat load conditions.

[0021] In the embodiments of this application, during the construction of distributed branches, multiple independent regulating branches are drawn out in parallel from the main circuit through distribution manifolds. Each branch is equipped with an independent flow regulation unit (such as a proportional control valve) to achieve decoupled control of the heat flowing through each second heat exchange end, ensuring that heat users with different characteristics can obtain asymmetric heat distribution as needed.

[0022] In the embodiments of this application, the hydrogen fuel cell stack (such as a proton exchange membrane fuel cell, PEMFC) is defined as the first heat exchange end of the system. The cooling medium flows directly through the cooling flow field inside the fuel cell stack, absorbing the waste heat generated by the electrochemical reaction. The fuel cell stack serves as both a power source for distributed heat users (such as drones) and the system's only native heat source. In step S110, it is necessary to ensure that the fuel cell stack outlet is sealed to the main circuit, converting the waste heat that would otherwise need to be dissipated through a radiator into a dispatchable thermal energy resource.

[0023] In the embodiments of this application, distributed components within the UAV that require thermal protection are defined as the second heat exchanger, including servo motor-type heat users, hydrogen supply valve-type heat users, flight control electronics, and mission payload sensors. For servo motor-type heat users, the focus is on thermal coupling of the servo motor housing, bearing housing, and gear pair lubrication areas. For hydrogen supply valve-type heat users, the focus is on addressing frosting and hydrogen supply stability issues in low-temperature environments. For flight control electronics and mission payload sensors (such as infrared / visible light cameras used for power line inspection), the focus is on ensuring that electronic components operate within a suitable temperature range and reducing the impact of temperature drift.

[0024] In the embodiments of this application, lightweight insulation materials (such as aerogel) are wrapped around the main circuit and each second heat exchange terminal to reduce heat loss in complex environments (such as frigid regions). Auxiliary heating units (such as polyimide electric heating films) are pre-installed near key measuring points in each distributed branch. These units, as a supplement to the heat transfer path, together with the waste heat distribution logic, constitute a complete "waste heat + electric heating" cascaded heat compensation architecture.

[0025] In the embodiments of this application, after completing heat exchange, each distributed branch converges through a return confluence pipeline and is ultimately connected to the inlet of the fuel cell stack. This forms a complete closed-loop circulation path of "stack heat absorption - main loop transmission - branch distribution - heat user heat release - return to cold end - stack heat reabsorption", realizing the continuous extraction and recycling of waste heat within the system.

[0026] After completing step S110, in step S120, the operating thermal state of the first heat exchange end, the multidimensional operating state characteristic parameters of each second heat exchange end including health status data, and the fluid dynamic parameters of the closed-loop heat transfer path are collected in real time.

[0027] In the embodiments of this application, the operating thermal state of the first heat exchange end includes core thermodynamic parameters and battery system operating state data. The core thermodynamic parameters include the operating temperature of the fuel cell stack, the inlet and outlet temperatures of the cooling medium, the heat generation intensity of the battery stack under the current power output, and the waste heat recovery potential.

[0028] In the embodiments of this application, the multidimensional operating state characteristic parameters include thermal state data, health state data, and task attribute data.

[0029] Specifically, thermal state data includes the real-time temperature and temperature change rate of key measuring points at each of the second heat exchange ends.

[0030] In the embodiments of this application, for servo motors and other heat-generating devices, the focus is on collecting temperatures from the housing, bearing housing, gear lubrication areas, and control circuit board. For flight control electronics and mission payload sensors, the temperature of the main control chip, power management components, and sensor sensitive areas is monitored. For hydrogen supply valves and other heat-generating devices, the temperature of the valve body and areas prone to frost is monitored.

[0031] In the embodiments of this application, the rate of temperature rise or fall at each measuring point is calculated in real time to identify potential thermal shock risks caused by thermal response delay or drastic environmental changes.

[0032] Specifically, the health status data includes performance degradation observation data of each second heat exchange terminal during operation. The performance degradation observation data is determined based on the response time, drive current, control stability, or signal offset of each second heat exchange terminal.

[0033] In the embodiments of this application, for servo motor-type heat users, the monitoring includes whether the response time is lagging, whether the drive current exhibits abnormal peaks due to increased grease viscosity, positioning errors, and the presence of abnormal oscillations. For flight control electronics and mission payload sensors, the monitoring includes signal offset (temperature drift), control signal stability, and data processing delay. For hydrogen supply valve-type heat users, the monitoring includes valve opening / closing response status and hydrogen supply pressure fluctuation characteristics.

[0034] By using these health status data as observation inputs, it is possible to determine in subsequent steps whether the component has experienced performance degradation due to the low temperature environment.

[0035] Specifically, the mission attribute data includes mission phase information for each second heat exchange terminal, the weight of its impact on flight safety, and the preset mission priority.

[0036] In embodiments of this application, mission phase information is used to identify which phase is currently in: takeoff, maneuvering, attitude adjustment, mission payload operation, cruise, or low load.

[0037] In the embodiments of this application, according to preset logic, servo motors and hydrogen supply valves that are related to flight safety are set to high safety impact weights, and mission loads are set to dynamic priority.

[0038] In the embodiments of this application, the fluid dynamic parameters of the closed-loop heat transfer path include flow rate and distribution status, and path operating pressure. During the acquisition of flow rate and distribution status, the total coolant flow rate in the closed-loop loop, the branch flow rate of each distributed branch, and the opening degree of each branch flow control valve (regulation unit) are monitored. During the acquisition of path operating pressure, the circulating pump drive pressure and pipeline return status are monitored to evaluate the energy transfer efficiency (related parameters of convective heat transfer coefficient) during the heat transfer process.

[0039] In the embodiments of this application, since different sensors (such as temperature sensors and servo controllers) have different sampling frequencies, all data are mapped to a unified time base to ensure that the heat generation of the fuel cell stack, the medium transport, and the heating of the heat user are consistent in time and space. Next, smoothing filtering is applied and combined with physical change trends for judgment. If a value change occurs at a certain measuring point (such as single-point sensor failure), the system will logically eliminate the affected point by combining adjacent measuring points and historical trends to prevent erroneous control commands. Furthermore, the collected data is tagged with "task stage" and "operating condition type" to provide classification reference for subsequent online posterior updates based on Bayesian inference.

[0040] After completing step S120, in step S130, the health status data in the multidimensional operating status feature parameters is used as the observation input, and the initial reliability prior model of each second heat exchanger is updated online based on the Bayesian inference framework to obtain the predicted reliability value of each second heat exchanger under the current operating condition.

[0041] The specific process involved in step S130 in the embodiments of this application can be found in [reference needed]. Figure 2 .

[0042] Figure 2 An exemplary flowchart illustrating an embodiment of this application is provided for obtaining the predicted reliability values ​​of each second heat exchanger under current operating conditions.

[0043] like Figure 2 As shown, in step S210, a low-temperature reliability assessment model is established for each second heat exchanger. The prior information for this model is obtained through historical test data of the same type of component, engineering experience, preliminary test fitting results, or low-temperature performance data of the material. In step S220, failure cycle data, truncated operation data, and performance degradation observation data of each second heat exchanger under current operating conditions are used as observation information to construct a joint likelihood function. In step S230, the low-temperature reliability assessment model is updated posteriorly using the joint likelihood function to obtain the predicted reliability value and its confidence interval under current operating conditions.

[0044] In the embodiments of this application, based on step S210, a basic reliability mathematical model is constructed for each type of second heat exchanger (such as servo motors, flight control electronics, hydrogen supply valves, etc.). Prior information is obtained by integrating historical life test data of components of the same model, cryogenic bench test data (such as torque spectrum tests under simulated flight conditions conducted in high and low temperature environment chambers), engineering experience data, and cryogenic physical property data of materials. The prior information is expressed as a probability distribution function of reliability parameters (such as characteristic lifetime, shape parameters, or failure thresholds in a Weibull distribution). For example, for servo motor-type heat users, a Weibull distribution can be used as their life prior model. For electronic devices, a log-normal distribution can be used.

[0045] In the embodiments of this application, based on step S220, the real-time health status data collected in step S120 is transformed into the input of the likelihood function, and the current real operating status is reflected through multi-source data fusion.

[0046] Specifically, failure cycle data is obtained by recording failure events that clearly occur in bench tests or historical operating conditions (such as servo motor response time exceeding limits, positioning accuracy exceeding limits, continuous low-frequency oscillations in the control loop, mechanical jamming, etc.).

[0047] Specifically, the time or number of cycles during which the drone has been operating stably without failure is recorded. This data is included in the calculation as truncated unfailed operation data, representing the current survival status.

[0048] Specifically, degradation likelihood terms are constructed as performance degradation observation data by monitoring performance degradation trajectories (such as the trend of abnormal increase in servo drive current as temperature decreases, signal offset, and deviation between key measurement points and target temperature).

[0049] In the embodiments of this application, the probability of degradation reaching the failure threshold is transformed into an equivalent likelihood based on a performance degradation trajectory model (such as a linear degradation model, an inverse Gaussian process, or an empirical model based on temperature-time coupling). By multiplying the aforementioned failure cycle data, non-failure truncated operation data, and the probability of degradation reaching the failure threshold, a joint likelihood function reflecting the current specific operating condition (such as a line patrol mission in a frigid region) is formed.

[0050] In the embodiments of this application, Bayes' theorem is used for online inference, combining the prior distribution with the joint likelihood function to calculate the posterior distribution of the model parameters.

[0051] Specifically, based on the currently observed performance degradation rate and thermal environment data, the system automatically corrects the parameters of the initial model. If the servo drive current is observed to continuously increase or the response slows down, the likelihood function shifts the posterior distribution of the drive towards the direction of increasing failure rate.

[0052] Specifically, under the current ambient temperature, flight load, and remaining mission duration, the probability prediction of the component's ability to function normally is used as the mission reliability prediction value, and the uncertainty range of the reliability prediction is given, providing confidence support for subsequent adaptive parameter updates (such as weight allocation in S140).

[0053] In the embodiments of this application, a full lifecycle rolling update mechanism is executed during the process of updating the low-temperature reliability assessment model a posteriori through the joint likelihood function.

[0054] For details on the specific process involved in implementing the full lifecycle rolling update mechanism in the embodiments of this application, please refer to [link / reference]. Figure 3 .

[0055] Figure 3 An exemplary flowchart illustrating the execution of the full lifecycle rolling update mechanism according to an embodiment of this application is shown.

[0056] like Figure 3 As shown, in step S310, the posterior update procedure is triggered according to the set cumulative running time or task cycle. In step S320, the posterior update result of the previous control cycle is extracted and converted into prior information for the low-temperature reliability assessment model of each second heat exchanger in the current control cycle. In step S330, the latest performance degradation observation data and non-failure truncated running data collected in the current control cycle are used as new observation information to update the joint likelihood function. In step S340, the task reliability prediction value is continuously refreshed using the updated joint likelihood function and fed back to the step of dynamically correcting the safe temperature threshold and heat allocation weight of each second heat exchanger.

[0057] In the embodiments of this application, a triggering mechanism is set according to the operational characteristics of the UAV, and updates are automatically executed through the airborne control unit or the ground operation and maintenance platform. Post-updates are triggered by setting cumulative flight time (e.g., every 50 hours of cumulative flight time), cumulative low-temperature exposure time (e.g., 10 hours of cumulative operation in an environment below -30°C), or by mission number / cycle (e.g., after completing a power line inspection mission). At the trigger point, the system extracts all fuel cell thermoelectric states, cooling medium fluid dynamic parameters, temperature control performance, and performance characteristics data of each heat user (e.g., servos, valves) recorded within that cycle.

[0058] In the embodiments of this application, the Bayesian posterior distribution result of the previous control cycle is extracted, which includes the performance degradation characteristics accumulated by the component during the previous operation phase. The aforementioned posterior distribution parameters are used as the initial prior information for the current control cycle. Through this rolling iteration, the reliability assessment model can automatically correct itself as the UAV transitions from a "new" state to an "aged" state, identifying long-term fatigue or wear trends of components without manual intervention.

[0059] In the embodiments of this application, the system uses the latest measured data collected in the current cycle as observational evidence to calibrate the low-temperature reliability assessment model. Specifically, the response time fluctuation of the servo motor, the abnormal increase in drive current, the positioning error trend, and the opening and closing delay of the hydrogen supply valve are extracted as the latest performance degradation data in the current cycle. The survival time of each component successfully completing the task in the frigid environment in the current cycle is recorded as the non-failure truncated data. These data reflecting the latest hardware health status are substituted into the likelihood function. If it is observed that the servo motor takes longer to reach the temperature or the drive current continues to rise under the same waste heat supply, the likelihood function will shift the drive reliability prediction in the direction of increased risk.

[0060] In the embodiments of this application, during the execution of step S340, the safety temperature threshold is dynamically adjusted. If the reliability prediction value of a certain heat user (such as a servo motor) is lower than the preset threshold, the system determines that it is in a sub-healthy state or its lubrication performance has declined. At this time, the safety temperature threshold of the heat user is automatically increased (for example, from -20°C to -10°C) to suppress further degradation of its mechanical properties through higher heat compensation. At the same time, when the total amount of waste heat from the fuel cell is limited, the system automatically increases the flow weight of the sub-healthy component in the distribution manifold to ensure that heat resources are preferentially tilted to high-risk, low-reliability nodes. If the waste heat is still insufficient to maintain the increased temperature threshold, the auxiliary heating unit of the branch is triggered in advance to ensure flight safety. In addition, when the reliability continues to decline or the energy efficiency assessment shows an abnormally low waste heat recovery rate, the system outputs self-diagnostic results including "aging of heat exchange channels, increased contact thermal resistance, degradation of insulation materials, or pipe blockage" to guide maintenance personnel to perform targeted maintenance.

[0061] Through this rolling closed loop of "operational data acquisition - post-hoc update - reliability prediction - parameter reallocation", the risk of hardware aging can be adaptively compensated throughout its entire life cycle, ensuring the thermal reliability of hydrogen fuel cell drones in long-term high-intensity missions.

[0062] After step S130 is completed, in step S140, the safe temperature threshold and heat distribution weight of each second heat exchange end are dynamically adjusted according to the task reliability prediction value.

[0063] The specific process involved in step S140 in the embodiments of this application can be found in [reference needed]. Figure 4 .

[0064] Figure 4 An exemplary flowchart illustrating an embodiment of this application is shown, showing how to dynamically adjust the safety temperature threshold and heat allocation weight of each second heat exchanger based on the predicted mission reliability value.

[0065] like Figure 4As shown, in step S410, it is determined whether the predicted task reliability value is lower than the set task reliability threshold. In response to the predicted task reliability value not being lower than the set task reliability threshold, in step S420, the preset target temperature, safe temperature threshold, and heat distribution weight of each second heat exchanger are maintained, and conventional energy efficiency control is performed. In response to the predicted task reliability value being lower than the set task reliability threshold, in step S430, corresponding reliability enhancement control is performed based on the user type corresponding to the second heat exchanger.

[0066] In the embodiments of this application, short-term power consumption optimization and long-term fault self-diagnosis strategies are implemented during the routine energy efficiency regulation process.

[0067] The specific processes involved in implementing short-term power consumption optimization and long-term fault self-diagnosis strategies in the embodiments of this application can be found in [reference needed]. Figure 5 .

[0068] Figure 5 An exemplary flowchart illustrating the implementation of short-term power consumption optimization and long-term fault self-diagnosis strategies in an embodiment of this application is shown.

[0069] like Figure 5 As shown, in step S510, waste heat recovery utilization rate, temperature attainment time, and total system power consumption are extracted based on the operating data of the closed-loop heat transfer path to construct a comprehensive energy efficiency index. In step S520, it is determined whether the comprehensive energy efficiency index meets the first deviation condition or the second deviation condition. In response to not meeting the first or second deviation condition, in step S530, the current fluid dynamics parameters of the closed-loop heat transfer path are maintained. In response to meeting the first or second deviation condition, in step S540, it is determined whether the first or second deviation condition is met. In response to meeting the first deviation condition, in step S550, while maintaining the real-time temperature of the key measuring points at each second heat exchange end not lower than the safe temperature threshold, the preset parameter values ​​of the corresponding closed-loop heat transfer path fluid dynamics parameters are lowered. In response to meeting the second deviation condition, in step S560, a system maintenance warning is generated, and self-diagnostic results including heat exchange channel aging, increased contact thermal resistance, insulation material degradation, or pipe blockage are output.

[0070] In the embodiments of this application, the first deviation condition is that the current value of the comprehensive energy efficiency index deviates from the set energy efficiency range.

[0071] In the embodiments of this application, the second deviation condition is that the time for temperature to meet the standard in the comprehensive energy efficiency index continues to extend within a preset historical period or the waste heat recovery utilization rate continues to decline within a preset historical period.

[0072] In the embodiments of this application, during routine control, a comprehensive index is first constructed based on the waste heat recovery utilization rate (the ratio of recovered heat to total heat production), temperature attainment time (the time required for the measuring point to reach the target temperature), and the total power consumption of the circulation system. If the real-time energy efficiency index deviates from the set range (for example, the fuel cell stack generates a large amount of heat, and high-flow circulation is not required to maintain the temperature), the system, while ensuring that the temperature at the measuring point is higher than the safety threshold, reduces the speed of the circulation pump or the opening of the branch valve to minimize parasitic power consumption. If the system detects that the temperature attainment time continues to increase or the waste heat recovery efficiency tends to decrease over multiple consecutive task cycles, it will automatically trigger a "self-diagnosis report" to identify physical bottlenecks caused by non-control parameters, such as: aging of heat exchange channels due to fouling on the heat exchanger surface, increased contact thermal resistance due to cracking of the thermal interface material, damage or degradation of the insulation layer, or partial blockage of the pipeline. At the same time, early warnings are output through the ground station or dashboard to prevent small deviations from developing into major failures.

[0073] In the embodiments of this application, user types include servo motor-type hot users, hydrogen supply valve-type hot users, flight control electronic equipment, and mission payload sensors.

[0074] In the embodiments of this application, for servo motor type heat users, during the process of performing corresponding reliability enhancement control based on the user type corresponding to the second heat exchange terminal, the heat allocation weight of the corresponding distributed branch is increased, and its safe temperature threshold is raised within the preset upper and lower limit constraints.

[0075] Specifically, for servo motor users, considering the response lag caused by increased grease viscosity and increased drive current at low temperatures, the safe temperature threshold will be increased (e.g., from -20℃ to -10℃) and the branch flow distribution weight will be increased within the preset physical constraints (e.g., the tuning range is limited to 10%-20% of the initial value) to compensate for mechanical degradation with heat.

[0076] In the embodiments of this application, for heat users such as hydrogen supply valves, during the process of performing corresponding reliability enhancement control based on the user type corresponding to the second heat exchange terminal, when the operating condition of low temperature start-up or hydrogen supply status fluctuation is identified, the heat allocation weight of the corresponding distributed branch is increased by a preset weight.

[0077] Specifically, for heat users such as hydrogen supply valves, when a decrease in reliability is detected and the system is in a sensitive operating condition such as takeoff, drastic changes in altitude, or fluctuations in hydrogen supply, the system directly increases the weight of that branch by a preset fixed offset (such as increasing the weight by 15%) to prevent the valve body from frosting and causing the action to become stuck.

[0078] Specifically, for heat users such as hydrogen supply valves, when a decrease in reliability is detected but not during sensitive operating conditions such as takeoff, drastic altitude changes, or fluctuations in hydrogen supply, a preventative robust control strategy is implemented: the control unit no longer adopts a step-like, forceful heating approach, but instead moderately increases the safe temperature threshold of the heat user within preset upper and lower limits, and fine-tunes its heat allocation weight incrementally. This aims to utilize the stable waste heat output during the cruise phase to bring the performance indicators of sub-optimal components back to the high-reliability temperature range ahead of schedule. At the same time, the system will use this relatively stable operating window to monitor the temperature compliance trend through the parameter adaptive update unit. If the temperature rise performance still falls short of expectations after a slight increase in heat, the system will formally trigger the long-term fault self-diagnosis logic for aging heat exchange channels, increased contact thermal resistance, or degradation of the insulation layer, and output targeted maintenance warning suggestions to the ground platform.

[0079] In the embodiments of this application, for flight control electronics or mission payload sensors, during the process of performing corresponding reliability enhancement control based on the user type corresponding to the second heat exchange terminal, the heat distribution weight of the corresponding distributed branch is increased, so that the waste heat of the first heat exchange terminal is directionally distributed to the second heat exchange terminal. By increasing its branch weight, the main control chip and sensors are ensured to be in the optimal temperature drift control range, and the decrease in control signal stability is suppressed.

[0080] In the embodiments of this application, during the execution of reliability enhancement control, a cascaded heat compensation strategy is implemented to address insufficient residual heat at the first heat exchange end.

[0081] In the embodiments of this application, the specific process involved in implementing the cascade heat compensation strategy to address insufficient waste heat at the first heat exchange terminal can be found in [reference needed]. Figure 6 .

[0082] Figure 6 An exemplary flowchart illustrating an embodiment of this application demonstrates a cascaded heat compensation strategy for addressing insufficient residual heat at the first heat exchange terminal.

[0083] like Figure 6 As shown, in step S610, the heating rate of the distributed branch corresponding to the second heat exchange end based on pure waste heat supply is calculated in real time. In step S620, it is determined whether the heating rate is lower than a preset slope or the allocated waste heat is insufficient to make the temperature of the key measuring point reach the safe temperature threshold. In response to the condition that the heating rate is lower than the preset slope or the allocated waste heat is insufficient to make the temperature of the key measuring point reach the safe temperature threshold, in step S630, the auxiliary heating unit set on the corresponding distributed branch is triggered to achieve heat compensation. In response to the condition that the heating rate is not lower than the preset slope or the allocated waste heat is insufficient to make the temperature of the key measuring point reach the safe temperature threshold, in step S640, the directional heat transfer based on the waste heat of the fuel cell stack is maintained, and the auxiliary heating unit on the corresponding distributed branch is locked in the off state.

[0084] In the embodiments of this application, when the fuel cell operates at low power or the environment is extremely cold, resulting in insufficient heat source, the system activates a tiered compensation mechanism. When the measured temperature rise rate is lower than the preset rate (e.g., the temperature rise rate cannot keep up with the ambient temperature drop rate) or calculations show that even with 100% activation of the allocated waste heat flow, the temperature at key measuring points cannot rise above the safe threshold, the auxiliary heating unit located on the corresponding distributed branch is triggered to achieve heat compensation. After the auxiliary heating is activated, a dual heat source mode of "waste heat + electric heating" is formed. Once reliability recovers and the temperature enters the safe range, the auxiliary heating is immediately locked to the off state, and the system is maintained by waste heat.

[0085] After step S140 is completed, in step S150, based on the corrected heat distribution weight, and on the premise of satisfying the thermal safety boundary of the first heat exchange terminal, the waste heat is distributed from the first heat exchange terminal to each second heat exchange terminal through each distributed branch.

[0086] In the embodiments of this application, it is essential to ensure that the temperature of the cooling medium at the fuel cell stack outlet is always maintained within the permissible physical limits of the fuel cell stack (e.g., 60°C to 85°C), preventing both overheating and undercooling. When the fuel cell stack is operating at high power output and generating a large amount of waste heat, even if the distributed branches (second heat exchange end) are already fully loaded with heat absorption, the system must prioritize activating the main radiator to forcibly dissipate heat to the environment, preventing the proton exchange membrane from dehydrating due to high temperatures and shortening the stack's lifespan. During the initial takeoff phase of the UAV or during low-power cruise, if each branch excessively extracts heat, the stack temperature may drop below the reaction activation temperature (e.g., below 55°C), leading to a decrease in reaction kinetics or a "flooding" phenomenon. In this case, the system will limit the heat distribution to heat users and may even shut down some non-critical branches, prioritizing the stack's "self-heating."

[0087] In the embodiments of this application, the specific process of distributing waste heat from the first heat exchange terminal to each of the second heat exchange terminals through each distributed branch can be found in [reference needed]. Figure 7 .

[0088] Figure 7 An exemplary flowchart illustrating how waste heat is distributed from a first heat exchanger to a second heat exchanger via distributed branches, according to an embodiment of this application, is shown.

[0089] like Figure 7As shown, in step S710, it is determined whether multiple second heat exchange terminals simultaneously have heat demand. If no multiple second heat exchange terminals simultaneously have heat demand, in step S720, based on the deviation between the key measuring point temperature and the target temperature of the second heat exchange terminal with heat demand, the heat allocation weight of the corresponding distributed branch is independently controlled to perform directional heat allocation. If multiple second heat exchange terminals simultaneously have heat demand, in step S730, the task stage and preset task priority corresponding to each second heat exchange terminal are obtained, and the heat allocation priority of each second heat exchange terminal is dynamically set based on the task stage to execute the corresponding heat allocation strategy.

[0090] In the embodiments of this application, under non-competitive conditions (no multiple heat demands), the opening of the independent PID control branch valve is adjusted based on the deviation between the key measuring point of a single heat user (such as the servo bearing housing or the back of the circuit board) and the target temperature.

[0091] In the embodiments of this application, the mission phases include the takeoff phase, the maneuver phase, the attitude adjustment phase, the mission payload operation phase, the cruise phase, and the low-load phase.

[0092] In the embodiments of this application, for the takeoff phase, maneuver phase or attitude adjustment phase, in the process of dynamically setting the heat allocation priority of each second heat exchange terminal based on the mission phase to execute the corresponding heat allocation strategy, the second heat exchange terminal that is involved in flight safety and has the highest preset mission priority is set as the first priority for heat allocation, and the available residual heat is preferentially allocated to the second heat exchange terminal with the first priority for heat allocation.

[0093] Specifically, servos, flight control electronics, and hydrogen supply pressure relief valves are prioritized for thermal allocation. The cryogenic risks to these components are mitigated first, even if this may cause non-core payloads (such as high-definition cameras) to temporarily fall below the target temperature range.

[0094] In the embodiments of this application, during the task load operation phase, in the process of dynamically setting the heat allocation priority of each second heat exchange terminal based on the task phase to execute the corresponding heat allocation strategy, the second heat exchange terminal responsible for signal processing is set to the highest heat allocation priority.

[0095] Specifically, the optoelectronic pod, signal processing unit, and key sensors are set as the highest priority. Control targets are adjusted to suppress temperature drift in electronic components, ensuring the accuracy of mission data and positioning precision.

[0096] In the embodiments of this application, during the cruise phase or low load phase, in the process of dynamically setting the heat allocation priority of each second heat exchange terminal based on the task phase to execute the corresponding heat allocation strategy, each second heat exchange terminal is set to the same priority, and the balanced heat allocation is executed with the goal of minimizing the real-time temperature deviation of each second heat exchange terminal.

[0097] Specifically, by treating all second heat exchange ends with equal priority and aiming to minimize the real-time temperature deviation between components, fluid circulation is used to make the overall thermal environment of the machine more uniform, thus preventing structural fatigue caused by uneven distribution of thermal stress.

[0098] This application also provides an active thermal management system based on fuel cell waste heat. It can use the aforementioned active thermal management method 100 based on fuel cell waste heat to perform active thermal management based on fuel cell waste heat, or it can use other methods to perform active thermal management based on fuel cell waste heat. This application does not limit this.

[0099] Figure 8 An exemplary structural block diagram of an active thermal management system based on fuel cell waste heat according to an embodiment of this application is shown.

[0100] like Figure 8 As shown, the system 800 includes a thermal management hardware topology module 810, a multi-dimensional status monitoring and acquisition module 820, a reliability assessment and prediction computing module 830, a control decision and parameter correction module 840, and a waste heat distribution execution module 850.

[0101] Specifically, the thermal management hardware topology module 810 is used to construct a closed-loop heat transfer path consisting of a main loop and multiple distributed branches. The closed-loop heat transfer path uses the fuel cell stack as the first heat exchange end and multiple distributed heat users as the second heat exchange end.

[0102] Specifically, the multidimensional condition monitoring and acquisition module 820 is used to collect in real time the operating thermal state of the first heat exchange end, the multidimensional operating condition characteristic parameters of each second heat exchange end including health status data, and the fluid dynamic parameters of the closed-loop heat transfer path.

[0103] Specifically, the reliability assessment and prediction computing module 830 is used to take the health status data in the multi-dimensional operating status feature parameters as the observation input, and perform online posterior update of the initial reliability prior model of each second heat exchange end based on the Bayesian inference framework to obtain the predicted reliability value of each second heat exchange end under the current operating conditions.

[0104] Specifically, the control decision and parameter correction module 840 is used to dynamically correct the safe temperature threshold and heat distribution weight of each second heat exchange end based on the predicted mission reliability value.

[0105] Specifically, the waste heat distribution execution module 850 is used to distribute waste heat from the first heat exchange terminal to each of the second heat exchange terminals through each distributed branch, based on the modified heat distribution weight and on the premise of satisfying the thermal safety boundary of the first heat exchange terminal.

[0106] When system 800 employs the aforementioned active thermal management method 100 based on fuel cell waste heat, it executes the aforementioned step S110 through the thermal management hardware topology module 810, step S120 through the multi-dimensional state monitoring and acquisition module 820, step S130 through the reliability assessment and prediction computing module 830, step S140 through the control decision and parameter correction module 840, and step S150 through the waste heat allocation execution module 850. The specific execution process can be found above and will not be repeated here.

[0107] While numerous embodiments of this application have been shown and described herein, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Many modifications, alterations, and alternatives will arise for those skilled in the art without departing from the spirit and intent of this application. It should be understood that various alternatives to the embodiments of this application described herein may be employed in the practice of this application. The appended claims are intended to define the scope of protection of this application and therefore cover equivalents or alternatives within the scope of these claims.

Claims

1. An active thermal management method based on fuel cell waste heat, characterized in that, include: A closed-loop heat transfer path consisting of a main loop and multiple distributed branches is constructed. The closed-loop heat transfer path uses the fuel cell stack as the first heat exchange end and multiple distributed heat users as the second heat exchange end. The system collects in real time the operating thermal status of the first heat exchange end, the multidimensional operating status characteristic parameters of each second heat exchange end including health status data, and the fluid dynamic parameters of the closed-loop heat transfer path. Using the health status data in the multidimensional operating status feature parameters as observation input, the initial reliability prior model of each second heat exchanger is updated online based on the Bayesian inference framework to obtain the predicted reliability value of each second heat exchanger under the current operating condition. The safe temperature threshold and heat distribution weight of each second heat exchanger are dynamically adjusted based on the predicted mission reliability value. Based on the corrected heat distribution weight, and provided that the thermal safety boundary of the first heat exchange terminal is met, the waste heat is distributed from the first heat exchange terminal to each second heat exchange terminal through each distributed branch. The multidimensional operating state feature parameters include thermal state data, health state data, and task attribute data. The thermal state data includes the real-time temperature and temperature change rate of key measuring points at each second heat exchange end. The health status data includes performance degradation observation data of each second heat exchange terminal during operation. The performance degradation observation data is determined based on the response time, drive current, control stability, or signal offset of each second heat exchange terminal. The task attribute data includes information on the task stage to which each second heat exchanger belongs, the weight of its impact on flight safety, and the preset task priority. In obtaining the predicted reliability values ​​of each second heat exchanger under the current operating conditions, the following steps are performed: A low-temperature reliability assessment model is established for each second heat exchange end. The prior information of the low-temperature reliability assessment model is obtained through historical test data of the same type of component, engineering experience, preliminary test fitting results or low-temperature performance data of materials. The failure cycle data, non-failure truncated operation data, and performance degradation observation data of each second heat exchange end under the current operating conditions are used as observation information to construct a joint likelihood function. By performing a posteriori update on the low-temperature reliability assessment model using the joint likelihood function, the predicted reliability value of the task under the current operating condition and its confidence interval are obtained. In the process of dynamically adjusting the safe temperature threshold and heat distribution weight of each second heat exchanger based on the predicted mission reliability value, the following steps are performed: Determine whether the predicted task reliability value is lower than a set task reliability threshold; In response to the fact that the predicted mission reliability value is not lower than the set mission reliability threshold, the preset target temperature, safe temperature threshold and heat distribution weight of each second heat exchange end are maintained, and routine energy efficiency control is performed. In response to a task reliability prediction value being lower than a set task reliability threshold, corresponding reliability enhancement controls are executed based on the user type corresponding to the second hot-swap terminal.

2. The active thermal management method based on fuel cell waste heat according to claim 1, characterized in that, The user types include servo motor-type thermal users, hydrogen supply valve-type thermal users, flight control electronic equipment, and mission payload sensors; For servo motor-type heat users, during the process of performing corresponding reliability enhancement control based on the user type corresponding to the second heat exchange terminal, the heat allocation weight of the corresponding distributed branch is increased, and its safe temperature threshold is raised within the preset upper and lower limit constraints. For heat users such as hydrogen supply valves, during the process of performing corresponding reliability enhancement control based on the user type corresponding to the second heat exchange end, when the working condition of low temperature start-up or hydrogen supply status fluctuation is identified, the heat allocation weight of the corresponding distributed branch is increased by the preset weight. For flight control electronics or mission payload sensors, during the process of performing corresponding reliability enhancement control based on the user type corresponding to the second heat exchange terminal, the heat distribution weight of the corresponding distributed branch is increased so that the waste heat of the first heat exchange terminal is directionally distributed to the second heat exchange terminal.

3. The active thermal management method based on fuel cell waste heat according to claim 2, characterized in that, During the implementation of reliability enhancement control, a cascade heat compensation strategy is implemented to address the insufficient residual heat at the first heat exchange end. In implementing the cascade heat compensation strategy to address insufficient waste heat at the first heat exchange terminal, the following steps are performed: The heating rate of the distributed branch corresponding to the second heat exchange end based on pure waste heat supply is calculated in real time. Determine whether the heating rate is lower than the preset slope or the allocated residual heat is insufficient to make the temperature of the key measuring point reach the safe temperature threshold. In response to the situation where the heating rate is lower than the preset slope or the allocated residual heat is insufficient to make the temperature of the key measuring point reach the safe temperature threshold, the auxiliary heating unit set on the corresponding distributed branch is triggered to achieve heat compensation. In response to situations where the heating rate is lower than the preset slope or the allocated waste heat is insufficient to bring the temperature of key measuring points to a safe temperature threshold, directional heat transfer based on the waste heat of the fuel cell stack is maintained, and the auxiliary heating units on the corresponding distributed branches are locked in a shut-off state.

4. The active thermal management method based on fuel cell waste heat according to claim 1, characterized in that, During the implementation of routine energy efficiency regulation, short-term power consumption optimization and long-term fault self-diagnosis strategies are implemented. The following steps are performed during the execution of short-term power consumption optimization and long-term fault self-diagnosis strategies: Based on the operational data of the closed-loop heat transfer path, waste heat recovery utilization rate, temperature attainment time and total system power consumption are extracted to construct a comprehensive energy efficiency index; Determine whether the comprehensive energy efficiency index meets the first deviation condition or the second deviation condition; In response to the failure to meet the first or second deviation condition, the current fluid dynamic parameters of the closed-loop heat transfer path are maintained. In response to whether the first deviation condition or the second deviation condition is met, determine whether the first deviation condition or the second deviation condition is met; In response to meeting the first deviation condition, while maintaining the real-time temperature of the key measuring points at each second heat exchange end at no lower than the safe temperature threshold, the fluid dynamics parameters of the corresponding closed-loop heat transfer path are adjusted to the preset parameter value. In response to the fulfillment of the second deviation condition, a system maintenance warning is generated, and self-diagnostic results including aging of heat exchange channels, increased contact thermal resistance, degradation of insulation materials, or pipe blockage are output. The first deviation condition is that the current value of the comprehensive energy efficiency index deviates from the set energy efficiency range, and the second deviation condition is that the temperature compliance time in the comprehensive energy efficiency index continues to extend within a preset historical period or the waste heat recovery utilization rate continues to decrease within a preset historical period.

5. The active thermal management method based on fuel cell waste heat according to claim 1, characterized in that, In the process of distributing waste heat from the first heat exchange terminal to each of the second heat exchange terminals through each distributed branch, the following steps are performed: Determine if multiple second heat exchange terminals simultaneously have heat demand; In response to the absence of multiple second heat exchange terminals having heat demand simultaneously, the heat allocation weight of the corresponding distributed branch is independently controlled based on the deviation between the temperature of the key measuring point of the second heat exchange terminal with heat demand and the target temperature, so as to perform directional heat allocation. In response to the simultaneous heat demand of multiple second heat exchange terminals, the task stage and preset task priority corresponding to each second heat exchange terminal are obtained, and the heat allocation priority of each second heat exchange terminal is dynamically set based on the task stage to execute the corresponding heat allocation strategy. The mission phases include takeoff, maneuvering, attitude adjustment, mission payload operation, cruise, and low-load phases. During the takeoff phase, maneuver phase, or attitude adjustment phase, in the process of dynamically setting the heat allocation priority of each second heat exchange terminal based on the mission phase to execute the corresponding heat allocation strategy, the second heat exchange terminal with the highest preset mission priority related to flight safety is set as the first priority for heat allocation, and the available residual heat is preferentially allocated to the second heat exchange terminal with the first priority for heat allocation. During the task load operation phase, in the process of dynamically setting the heat allocation priority of each second heat exchange terminal based on the task phase to execute the corresponding heat allocation strategy, the second heat exchange terminal responsible for signal processing is set to the highest heat allocation priority. During the cruise phase or low-load phase, in the process of dynamically setting the heat allocation priority of each second heat exchanger based on the task phase to execute the corresponding heat allocation strategy, each second heat exchanger is set to the same priority, and the heat allocation is balanced with the goal of minimizing the real-time temperature deviation of each second heat exchanger.

6. The active thermal management method based on fuel cell waste heat according to claim 1, characterized in that, During the process of updating the low-temperature reliability assessment model a posteriori using the joint likelihood function, a full lifecycle rolling update mechanism is executed. During the implementation of the full lifecycle rolling update mechanism, the following steps are performed: The retrospective update procedure is triggered based on the set cumulative running time or task cycle. Extract the posterior update results of the previous control cycle and convert them into prior information of the low-temperature reliability assessment model of each of the second heat exchange ends under the current control cycle; The joint likelihood function is updated by using the latest performance degradation observation data and non-failure truncated operation data collected in the current control cycle as new observation information. The reliability prediction value of the task is refreshed by rolling the updated joint likelihood function and fed back to the step of dynamically correcting the safe temperature threshold and heat allocation weight of each second heat exchange end.

7. An active thermal management system based on fuel cell waste heat, characterized in that, The system employs the active thermal management method based on fuel cell waste heat as described in any one of claims 1-6, wherein the system comprises: The thermal management hardware topology module is used to construct a closed-loop heat transfer path consisting of a main loop and multiple distributed branches. The closed-loop heat transfer path uses the fuel cell stack as the first heat exchange end and multiple distributed heat users as the second heat exchange end. The multidimensional status monitoring and acquisition module is used to collect in real time the operating thermal status of the first heat exchange end, the multidimensional operating status characteristic parameters of each second heat exchange end including health status data, and the fluid dynamic parameters of the closed-loop heat transfer path. The reliability assessment and prediction computing power module is used to take the health status data in the multi-dimensional operating status feature parameters as observation input, and perform online posterior update of the initial reliability prior model of each second heat exchange end based on the Bayesian inference framework to obtain the task reliability prediction value of each second heat exchange end under the current operating conditions. The control decision and parameter correction module is used to dynamically correct the safe temperature threshold and heat distribution weight of each second heat exchange end based on the predicted task reliability value. The waste heat distribution execution module is used to distribute waste heat from the first heat exchange terminal to each second heat exchange terminal through each distributed branch, based on the modified heat distribution weight and provided that the thermal safety boundary of the first heat exchange terminal is met.

Images