Unmanned aerial vehicle hydrogen fuel cell hybrid power system and control method thereof

Abstract

The invention belongs to the technical field of new energy power, and particularly relates to an unmanned aerial vehicle hydrogen fuel cell hybrid power system and a control method thereof. The system comprises a fuel cell stack which is used for providing a reference power supply and has adjustable power output; the super capacitor bank is used for outputting peak current when transient high power is needed, and has the characteristics of high power density and quick response; the lithium iron phosphate battery pack is used for providing an emergency power supply when the fuel cell fails; the dynamic bus controller integrates the voltage hysteresis comparator and the CAN bus communication module and is used for controlling energy output of the fuel cell stack, the super capacitor bank and the lithium iron phosphate battery pack; the LSTM neural network prediction module is used for predicting a future power demand according to the flight attitude, the meteorological data and a historical load curve and triggering power climbing of the fuel cell in advance; and the Buck-Boost circuit is used for the fuel cell to directly charge the super capacitor bank, so that the dynamic response speed and the system stability of the unmanned aerial vehicle in a violent maneuvering state are remarkably improved.

Description

Technical Field

[0001] The present invention belongs to the field of new energy power technology, and specifically relates to a hydrogen fuel cell hybrid power system for unmanned aerial vehicles based on the triple energy synergy of hydrogen fuel cell, supercapacitor and lithium battery, and a control method thereof. Background Art

[0002] With the continuous development of drone technology, its application scope has gradually expanded from entertainment and photography to logistics and transportation, agricultural monitoring, disaster relief and other fields. Consequently, the requirements for drone endurance, power response speed, and system reliability are becoming increasingly stringent. Existing hydrogen fuel cell power systems, due to their high energy density and environmentally friendly characteristics, are ideal for long-duration drone flight. However, hydrogen fuel cells have limitations in power response speed, typically requiring more than one second to complete power response. This is insufficient for the demands of drones in demanding maneuvers such as wind resistance, climbing, and sharp turns.

[0003] The traditional solution is to use lithium batteries as an auxiliary power source to handle instantaneous high-power demands. However, frequent high-current discharges can cause lithium batteries to rapidly degrade in lifespan, with a lifespan degradation rate exceeding 40% severely impacting the long-term use of the system. Furthermore, the lithium battery power module accounts for approximately 30% of the total power system weight, increasing the drone's weight and, in turn, impacting flight time. Furthermore, leaving lithium batteries fully charged for extended periods accelerates aging, reducing their lifespan and increasing system safety risks. Summary of the Invention

[0004] In view of this, and targeting the pain points of existing UAV hydrogen fuel cell power systems in terms of response delay, output fluctuation and emergency landing risks, the present invention provides a UAV hydrogen fuel cell hybrid power system based on the triple energy synergy of hydrogen fuel cell-supercapacitor-lithium battery and its control method. By utilizing the high transient response capability of supercapacitors and the high energy density characteristics of lithium batteries, through a dynamic bus control algorithm, efficient collaboration and reliable operation of the system are achieved, which significantly improves the power response speed and system stability of the UAV under intense maneuvering conditions, while extending the service life of the lithium battery, reducing the system weight, increasing the flight time, and improving the overall reliability and safety of the system.

[0005] One embodiment of the present invention provides a hydrogen fuel cell hybrid system for an unmanned aerial vehicle, the system comprising: Fuel cell stack, used to provide reference power, with adjustable power output; Supercapacitor bank, used to output peak current during transient high power demand, with high power density and fast response characteristics; Lithium iron phosphate battery pack, used to provide emergency power in the event of fuel cell failure; A dynamic bus controller with integrated voltage hysteresis comparator and CAN bus communication module to control the energy output of the fuel cell stack, supercapacitor group and lithium iron phosphate battery pack; An LSTM neural network prediction module is used to predict future power demand based on flight attitude, meteorological data, and historical load curves, and trigger fuel cell power ramp-up in advance; Buck-Boost circuit is used for fuel cells to directly charge supercapacitor banks.

[0006] As a further solution of the present invention, the operating voltage range of the fuel cell stack is 48-65V, the rated power is 4kW, and the power ramp rate is ≥50% / s.

[0007] As a further solution of the present invention, the supercapacitor group is composed of 24 strings of supercapacitors or 2 groups of 12 strings of supercapacitors in parallel, with a total capacity ≥40F, an internal resistance <1mΩ, directly connected to the high-voltage bus, and a full voltage of 64.8V to a minimum operating voltage of 44V.

[0008] As a further solution of the present invention, the supercapacitor group adopts 2 groups of 12 series-parallel topology, each group is equipped with an independent voltage balancing circuit, and 50% of the peak output capacity is retained when a single group fails.

[0009] As a further solution of the present invention, the lithium iron phosphate battery pack includes a 14-string structure, a nominal voltage of 44.8V, and a storage energy of ≥100Wh.

[0010] As a further solution of the present invention, a voltage hysteresis comparator and a CAN bus communication module are integrated on the dynamic bus controller, and the response time of the voltage hysteresis comparator is less than 10 μs.

[0011] As a further solution of the present invention, the voltage range of the supercapacitor group and the lithium battery group naturally covers the 44-65V system requirement, and no DC / DC conversion module is required.

[0012] As a further solution of the present invention, the dynamic bus controller is configured as follows: a) When a step load greater than 100A is detected, the supercapacitor bank is activated to output peak current within 0.1ms; b) Synchronously control the fuel cell stack to linearly increase power from 30% to 100% within 2 seconds; c) When the fuel cell fails, it switches to the lithium battery pack within 2 seconds and outputs a constant power of 1400W for ≥180 seconds.

[0013] As a further solution of the present invention, the Buck-Boost circuit enables the fuel cell stack to bypass the lithium battery pack and directly charge the supercapacitor pack. The lithium battery pack is locked at 80% SOC (44.8V) during flight and automatically discharges to 50% SOC (43.2V) after shutdown.

[0014] As a further solution of the present invention, the dynamic bus controller includes an LSTM neural network prediction module, which inputs flight attitude, meteorological data and historical load curves, predicts power demand in the next 10 seconds and triggers fuel cell power ramp-up in advance.

[0015] One solution of the present invention also provides a control method based on a hydrogen fuel cell hybrid power system of a UAV, comprising the following steps: Steady-state load stage: directly powered by the fuel cell stack (≤50A); Step load response stage: After detecting a current request greater than 100A, the supercapacitor bank starts discharging within 0.1ms; Synchronously instruct the fuel cell stack to linearly ramp up from 30% power to 100% power within 2 seconds; Fault emergency stage: T=0-2 seconds: supercapacitor bank outputs full power to prevent stall; T=2-180 seconds: The lithium battery pack takes over the load at a constant power of 1400W.

[0016] As a further solution of the present invention, an LSTM neural network is used to predict the power demand in the next 10 seconds, triggering the fuel cell power ramp-up in advance and reducing the frequency of supercapacitor intervention.

[0017] As a further solution of the present invention, a Buck-Boost circuit is used to directly charge the supercapacitor group from the fuel cell, thereby avoiding the high current impact of the lithium battery.

[0018] Compared with the prior art, the technical solution provided by the present invention has the following beneficial effects: The present invention directly connects the supercapacitor group to the high-voltage bus to achieve microsecond transient response. The dynamic bus controller controls the smooth transition of the fuel cell power, effectively avoiding the overshoot risk caused by chemical response lag. The supercapacitor and lithium battery naturally cover the 44-65V system voltage, eliminating the DC / DC conversion module. By setting up a Buck-Boost direct charging circuit, the fuel cell directly charges the supercapacitor, avoiding the high current impact of the lithium battery, increasing the capacitor recharging speed, and reducing the risk of thermal runaway. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 This is a topological diagram of a hydrogen fuel cell hybrid power system for a UAV according to one embodiment of the present invention; Figure 2 This is a flow chart of a normal flight mode in a hydrogen fuel cell hybrid system for a UAV according to one embodiment of the present invention; Figure 3 This is a flow chart of peak load response in a hydrogen fuel cell hybrid system for a drone according to one embodiment of the present invention; Figure 4 This is a flow chart of an emergency response to a fuel cell failure in a hydrogen fuel cell hybrid system for a drone according to one embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0021] In order to solve the problems of morphological adaptability defects, motion displacement errors and lack of multi-source coordination in the existing technology, the present invention provides a UAV hydrogen fuel cell hybrid power system and its control method, which significantly improves the conformality, efficiency and accuracy of radiotherapy through a triangular configuration multi-source array, a bionic two-stage collimator and a four-modal fusion tracking system.

[0022] In order to facilitate understanding of the following embodiments of the present application, the following embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0023] See Figure 1As shown, this embodiment of the present application provides a hydrogen fuel cell hybrid system for a drone, which includes a fuel cell stack, a supercapacitor group, a lithium iron phosphate battery group, a dynamic bus controller, an LSTM neural network prediction module, and a Buck-Boost circuit. The fuel cell stack is used to provide a reference power supply with adjustable power output; the supercapacitor group is used to output peak current in transient high power demand, with high power density and fast response characteristics; the lithium iron phosphate battery group is used to provide emergency power supply in the event of a fuel cell failure; the dynamic bus controller integrates a voltage hysteresis comparator and a CAN bus communication module to control the energy output of the fuel cell stack, the supercapacitor group, and the lithium iron phosphate battery group; the LSTM neural network prediction module is used to predict future power demand based on flight attitude, meteorological data, and historical load curves, and trigger the fuel cell power ramp in advance; the Buck-Boost circuit is used for the fuel cell to directly charge the supercapacitor group.

[0024] The drone hydrogen fuel cell hybrid power system of the present invention addresses the core pain points of response delay, output fluctuation, and emergency landing risks in drone hydrogen fuel cell power systems. When the drone is maneuvering at high speed or suddenly loaded, a supercapacitor is directly connected to the high-voltage busbar in milliseconds, instantly providing a 120A peak current, completely eliminating the 0.5-second power gap caused by chemical reaction lag in the fuel cell. A dynamic bus control algorithm coordinates the lithium battery and fuel cell to reduce the voltage fluctuation rate of the entire power system to less than 0.5% (far exceeding the industry standard of 2%), ensuring stable operation of the multi-motor system. Supercapacitors replace lithium batteries for instantaneous energy replenishment during high maneuverability, further improving the instantaneous dynamic response speed while significantly reducing the weight of the lithium battery and increasing flight time. A unique three-level fault-tolerant mechanism transforms power interruptions into a controllable emergency mode. Even if the fuel cell fails completely and the supercapacitor is depleted, the lithium battery can take over the load within 1 second and maintain a power output sufficient to ensure a safe landing of the drone for up to 3 minutes. Combined with a flight control path reconstruction algorithm, this system achieves 100% safe landing. The introduction of a lithium-ion battery health management system, through an 80% SOC operating point lock and automatic discharge upon shutdown, extends the battery's calendar life from the industry average of two years to over five years, significantly reducing lifecycle operation and maintenance costs. This solution systematically overcomes the reliability bottleneck in the hydrogen fuel cell hybrid power sector for drones, demonstrating measured performance in performance metrics such as power response speed under high maneuverability and minimal voltage drop under a 120A step load.

[0025] In this embodiment, the fuel cell stack has an operating voltage range of 48-65V, a rated power of 4kW, and a power ramp rate of ≥50% / s. The supercapacitor bank consists of 24 supercapacitors in series or two groups of 12 supercapacitors in parallel, with a total capacity of ≥40F and an internal resistance of <1mΩ. It is directly connected to the high-voltage bus, with a full voltage of 64.8V and a minimum operating voltage of 44V, and can withstand a peak current of 120A for 0-2 seconds.

[0026] In this embodiment, the supercapacitor bank utilizes two groups of 12 supercapacitors in a parallel topology, each equipped with an independent voltage-sharing circuit. This ensures that 50% of peak output capacity is retained in the event of a single group failure. The lithium iron phosphate battery pack comprises 14 supercapacitors in series, has a nominal voltage of 44.8V, stores ≥100Wh of energy, and supports a continuous discharge of ≥35A, enabling power delivery for a three-minute emergency landing.

[0027] In this embodiment, the dynamic bus controller integrates a voltage hysteresis comparator and a CAN bus communication module. The voltage hysteresis comparator has a response time of less than 10μs, enabling seamless capacitor-to-lithium battery switching. The voltage range of the supercapacitor and lithium battery pack naturally covers the 44-65V system requirement, eliminating the need for a DC / DC converter.

[0028] In this embodiment, the dynamic bus controller is configured as follows: a) When the supercapacitor is detected to respond to a step load greater than 100A within 100ms, the supercapacitor group is activated to output peak current within 0.1ms to maintain the system bus voltage fluctuation less than 5%; b) Synchronously control the fuel cell stack to linearly increase power from 30% to 100% within 2 seconds; c) When the fuel cell fails, it switches to the lithium battery pack within 2 seconds and outputs a constant power of 1400W for ≥180 seconds. The lithium battery takes over the load within 2 seconds and outputs a constant power of ≥180 seconds to ensure the safe landing of the drone.

[0029] The Buck-Boost circuit enables the fuel cell stack to bypass the lithium-ion battery pack and directly charge the supercapacitor pack. The lithium-ion battery pack is locked at 80% SOC (44.8V) during flight and automatically discharges to 50% SOC (43.2V) after shutdown. The dynamic bus controller includes an LSTM neural network prediction module, which inputs flight attitude, meteorological data, and historical load curves to predict power demand in the next 10 seconds and trigger the fuel cell power ramp in advance.

[0030] See Figures 1 to 4 As shown, this embodiment of the present application provides a control method based on a hydrogen fuel cell hybrid power system of a drone, comprising the following steps: 1. Steady-state load stage: directly powered by the fuel cell stack (≤50A); 2. Step load response stage: After detecting a current request greater than 100A, the supercapacitor bank starts discharging within 0.1ms; Synchronously instruct the fuel cell stack to linearly ramp up from 30% power to 100% power within 2 seconds; 3. Fault emergency stage: T=0-2 seconds: supercapacitor bank outputs full power to prevent stall; T=2-180 seconds: The lithium battery pack takes over the load at a constant power of 1400W.

[0031] In this embodiment, an LSTM neural network is used to predict the power demand in the next 10 seconds, triggering the fuel cell power ramp-up in advance and reducing the frequency of supercapacitor intervention. Specifically, the fuel cell is directly charged into the supercapacitor bank through a Buck-Boost circuit, avoiding the high current impact of the lithium battery.

[0032] In this embodiment, see Figure 2 As shown, the fuel cell is used to provide basic cruising power (3000-4000W); the supercapacitor is used to maintain a full charge state (64.8V) to respond to sudden load changes at any time; the lithium battery is used to maintain 80% SOC (44.8V) and is only used as an emergency backup. In peak load response (such as climbing, sharp turns, and wind resistance), see Figure 3 As shown, when encountering a sudden load change (0ms) such as climbing, strong wind, or sharp turn, the motor system sends a 120A instantaneous current request to the main controller, triggering the peak load response mechanism and the capacitor instantaneous response (0-0.1ms). The main controller activates the supercapacitor bank within 0.1ms, achieves zero-delay discharge through direct connection to the high-voltage bus, and outputs 120A current to fill the power gap.

[0033] The supercapacitor's internal resistance is less than 1mΩ, resulting in a voltage drop of only 0.84V, compared to >2V in traditional solutions. During the fuel cell power transition (1-2 seconds), the main controller synchronously instructs the fuel cell to slowly ramp up power (linearly from 30% to 100%), avoiding the risk of overshoot caused by chemical response lag. Energy transition is seamless (2 seconds), with the fuel cell completely taking over the load, the supercapacitor shutting down, and recharging initiated, achieving dynamic bus voltage coordination of 44-65V (without DC / DC conversion).

[0034] See also Figure 4 As shown in the figure, during the fuel cell failure emergency phase, the following are included: Phase 1 (0-2 seconds): The supercapacitor independently supports 120A peak current; Flight control lock attitude angle <5° to prevent instability; Phase 2 (2-180 seconds): The lithium battery outputs 1400W constant power; The descent rate is controlled in segments according to the altitude (100m↑: 3m / s→<30m: 1m / s).

[0035] In energy maintenance mode, the fuel cell is slowly charged (20-30A) for high-altitude cruising, and the lithium battery is fast charged (50A, required to simultaneously meet the requirements of altitude <50m and attitude angle <5°) for low-altitude safety. During lithium battery maintenance, the battery will automatically discharge to 50% SOC after landing and allowing it to sit for >2 hours. During this discharge mode, the avionics power consumption exceeds the heat dissipation of the ceramic resistor.

[0036] The present invention uses 2 groups of 12 supercapacitors in series and parallel, which is matched with the power composition of the drone in accordance with the present invention. The power system in the present invention adopts an operating voltage of 48-65V, and the output power of the fuel cell stack is 4kw. These parameters match the corresponding motor and blades. If the configuration of the power system changes, such as the power of the stack increases and the horsepower of the motor increases, the combination of supercapacitors can be changed. For example, 32 strings, 48 ​​strings, or 2 or even multiple groups of capacitors in series at the front end, and then multiple groups in parallel; or multiple groups are connected in parallel first, and then in series. Various permutations and combinations of series and parallel can be formed to match different power systems or reflect different power responses. This requires users to make flexible combinations according to actual needs.

[0037] Through magnetic resonance wireless power transmission (Qi 2.0 industrial grade), lithium batteries and capacitors are charged contactlessly, completely eliminating the risk of high-voltage short circuits and isolating the spread of thermal runaway. Charging efficiency is reduced to 85%, charging time is longer, and additional heat dissipation space is required. A fuel cell direct charging channel is added with a dedicated Buck-Boost circuit, allowing the fuel cell to bypass the lithium battery on the bus and directly charge the capacitor, avoiding high current impact on the lithium battery (especially in low temperature conditions) and extending the service life of the lithium battery.

[0038] The present invention directly connects the supercapacitor group to the high-voltage bus to achieve microsecond transient response. The dynamic bus controller controls the smooth transition of the fuel cell power, effectively avoiding the overshoot risk caused by chemical response lag. The supercapacitor and lithium battery naturally cover the 44-65V system voltage, eliminating the DC / DC conversion module. By setting up a Buck-Boost direct charging circuit, the fuel cell directly charges the supercapacitor, avoiding the high current impact of the lithium battery, increasing the capacitor recharging speed, and reducing the risk of thermal runaway.

[0039] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A hydrogen fuel cell hybrid power system for drones, characterized in that: The system includes: Fuel cell stack, used to provide reference power, with adjustable power output; Supercapacitor bank, used to output peak current during transient high power demand, with high power density and fast response characteristics; Lithium iron phosphate battery pack, used to provide emergency power in the event of fuel cell failure; A dynamic bus controller with integrated voltage hysteresis comparator and CAN bus communication module to control the energy output of the fuel cell stack, supercapacitor group and lithium iron phosphate battery pack; An LSTM neural network prediction module is used to predict future power demand based on flight attitude, meteorological data, and historical load curves, and trigger fuel cell power ramp-up in advance; Buck-Boost circuit is used for fuel cells to directly charge supercapacitor banks.

2. The UAV hydrogen fuel cell hybrid power system according to claim 1, characterized in that: The operating voltage range of the fuel cell stack is 48-65V, the rated power is 4kW, and the power ramp rate is ≥50% / s.

3. The UAV hydrogen fuel cell hybrid power system according to claim 2, characterized in that: The supercapacitor group consists of 24 strings of supercapacitors or 2 groups of 12 strings of supercapacitors in parallel, with a total capacity of ≥40F, an internal resistance of <1mΩ, directly connected to the high-voltage bus, and a full voltage of 64.8V to a minimum operating voltage of 44V.

4. The UAV hydrogen fuel cell hybrid power system according to claim 3, characterized in that: The supercapacitor bank adopts a 2-group 12-series parallel topology, each group is equipped with an independent voltage balancing circuit, and retains 50% of the peak output capacity when a single group fails.

5. The UAV hydrogen fuel cell hybrid power system according to claim 1 or 4, characterized in that: The lithium iron phosphate battery pack includes a 14-string structure, a nominal voltage of 44.8V, and a storage energy of ≥100Wh.

6. The UAV hydrogen fuel cell hybrid power system according to claim 5, characterized in that: The dynamic bus controller is integrated with a voltage hysteresis comparator and a CAN bus communication module, and the response time of the voltage hysteresis comparator is less than 10 μs.

7. The UAV hydrogen fuel cell hybrid power system according to claim 6, characterized in that: The voltage range of the supercapacitor pack and lithium battery pack naturally covers the 44-65V system requirement, without the need for a DC / DC conversion module.

8. The UAV hydrogen fuel cell hybrid power system according to claim 7, characterized in that: The dynamic bus controller is configured to: a) When a step load greater than 100A is detected, the supercapacitor bank is activated to output peak current within 0.1ms; b) Synchronously control the fuel cell stack to linearly increase power from 30% to 100% within 2 seconds; c) When the fuel cell fails, it switches to the lithium battery pack within 2 seconds and outputs a constant power of 1400W for ≥180 seconds.

9. The UAV hydrogen fuel cell hybrid power system according to claim 8, characterized in that: The Buck-Boost circuit enables the fuel cell stack to bypass the lithium battery pack and directly charge the supercapacitor pack. The lithium battery pack is locked at 80% SOC during flight and automatically discharges to 50% SOC after shutdown.

10. A control method for a hydrogen fuel cell hybrid power system for a UAV according to any one of claims 1 to 9, characterized in that: The following steps are involved: Steady-state load stage: powered directly by the fuel cell stack; Step load response stage: After detecting a current request greater than 100A, the supercapacitor bank starts discharging within 0.1ms; Synchronously instruct the fuel cell stack to linearly ramp up from 30% power to 100% power within 2 seconds; Fault emergency stage: T=0-2 seconds: supercapacitor bank outputs full power to prevent stall; T=2-180 seconds: The lithium battery pack takes over the load at a constant power of 1400W.

Images