Hybrid power energy management method and system for a compound wing unmanned aerial vehicle

Abstract

This invention relates to the field of charging management technology, specifically to a hybrid power energy management method and system for a compound-wing unmanned aerial vehicle (UAV). The method acquires real-time UAV status data to determine the flight mode; when in fixed-wing cruise mode, it predicts the maximum engine load rate based on the current sliding window and calculates a safe charging boundary value that does not consume flight power; it predicts the remaining cruise duration and the remaining battery charge at the end of the mission landing, and combines this with a safety threshold to determine the urgency of battery charging; it inputs real-time battery temperature, internal resistance, and remaining charge into a mapping function and weights this to obtain a charging adaptability coefficient; finally, it combines the above boundary values, urgency level, and adaptability coefficient to obtain the actual charging power for cross-modal charging, performs battery charge verification before mode transition, and disconnects non-critical loads in case of failure. This invention achieves intelligent collaborative allocation of power energy across modes, significantly improving the energy utilization rate and flight safety of UAVs.

Description

Technical Field

[0001] This invention relates to the field of charging management technology, and specifically to a hybrid power energy management method and system for compound-wing unmanned aerial vehicles (UAVs). Background Technology

[0002] With the rapid development of low-altitude economy and industrial drone technology, compound-wing drones, combining the advantages of multi-rotor vertical takeoff and landing (no dedicated runway required, flexible deployment) with the long endurance and high-efficiency cruise characteristics of fixed-wing drones, have demonstrated extremely high application value in industrial scenarios such as power line inspection, geographic mapping, emergency rescue, and logistics transportation. To overcome the endurance bottleneck of pure electric drones, a hybrid power architecture—combining a gasoline-powered system for fixed-wing cruise and an electric system for dedicated rotor vertical takeoff and landing—has become the core development direction for addressing the long-duration operation needs of compound-wing drones.

[0003] However, existing hybrid-electric unmanned aerial vehicles (UAVs) generally suffer from severe fragmentation and limitations in their energy management strategies. Most existing systems employ static and isolated control logic, with the gasoline and electric systems operating independently. They lack a comprehensive, lifecycle-wide collaborative energy management method adapted to the multiple flight modes of hybrid-electric UAVs (such as climb, forward transition, fixed-wing cruise, reverse transition, and landing). Consequently, they cannot achieve cross-modal energy optimization and allocation. Specifically, during the longest fixed-wing cruise phase of the mission, the piston engine typically has a large amount of unused redundant mechanical power during level flight, which is wasted. Simultaneously, the high-voltage battery packs dedicated to driving the rotors not only become useless "dead weight" during long-endurance cruise but also experience unexpected power losses due to system standby, avionics consumption, or attitude fine-tuning in response to severe weather.

[0004] Secondly, the risk of crashing at the end of the mission is extremely high because there is no dynamic cruise recharge mechanism. When the drone returns after completing a large-radius and long-distance line inspection or mapping mission and is preparing to perform a reverse conversion (fixed-wing to rotor) and vertical landing maneuver that consumes a lot of power, it is very likely to face the predicament of insufficient or even exhausted rotor battery power, which will cause the drone to lose lift, fail to land, or even crash and be damaged.

[0005] Finally, the few traditional systems that attempt to introduce in-flight charging often employ a simple fixed-power charging logic. This fails to consider the transient impact of complex high-altitude airflow disturbances on the engine's real-time flight load (blindly charging can easily preempt flight power, leading to stall), nor does it comprehensively assess the high-voltage battery's real-time charging capacity under different ambient temperatures, internal resistance aging, and remaining charge levels (forced charging can easily induce battery overheating, overcurrent lithium deposition, or even thermal runaway). Furthermore, existing systems lack pre-emptive charge safety checks and physical-level emergency power outage protection mechanisms before takeoff and in the face of sudden power failures.

[0006] Therefore, the existing hybrid electric vehicle (HEV) hybrid wing UAV energy management system cannot achieve cross-modal intelligent coordination and safe dynamic allocation of power energy under multiple flight modes. This not only leads to low overall energy utilization efficiency throughout the entire life cycle, but also fails to fundamentally eliminate the risk of landing and crashing due to insufficient rotor batteries at the end of long-endurance operations. Furthermore, it lacks a bottom-level physical safety supply defense line when facing complex high-altitude working conditions and sudden power failures, which seriously restricts the operational flexibility and overall flight reliability of hybrid wing UAVs in large-scale, long-endurance complex industrial scenarios. Summary of the Invention

[0007] To address the technical problem that existing hybrid-electric composite wing UAV energy management systems cannot achieve cross-modal intelligent coordination and safe dynamic allocation of power energy under multiple flight modes, the present invention aims to provide a hybrid power energy management method and system for composite wing UAVs. The specific technical solution adopted is as follows:

[0008] In a first aspect, one embodiment of the present invention provides a hybrid power energy management method for a compound-wing unmanned aerial vehicle (UAV), the method comprising the following steps:

[0009] Real-time acquisition of data on the power unit, energy storage unit, and flight status of the compound wing UAV to determine the current flight mode;

[0010] When in fixed-wing cruise mode, the maximum load rate of the engine is predicted based on the current sliding window data within a preset period in the future. Combined with the rated output power of the engine, the rechargeable safety boundary value that does not occupy flight power is calculated.

[0011] The remaining cruise time is predicted based on the total remaining mileage of the mission and the average cruise ground speed within the current sliding window. The remaining battery power at the end of the mission landing is predicted by combining the historical power consumption rate. The urgency of battery recharging is obtained by combining the preset expected landing remaining power safety threshold.

[0012] The real-time temperature, internal resistance, and remaining capacity of the high-voltage battery pack are input into the corresponding preset piecewise mapping function, and the weighted calculation is used to obtain the charging adaptability coefficient at the current moment.

[0013] Based on the rechargeable safety boundary value, the urgency of battery replenishment, and the charging adaptability coefficient, the actual charging power is obtained to perform cross-mode replenishment; before the flight mode conversion, a pre-emptive power verification based on a preset takeoff safety power threshold is performed, and when a power failure is detected, emergency control is triggered, and the power management module physically cuts off the power supply link of non-critical loads to ensure rotor power supply.

[0014] Furthermore, the method for obtaining the rechargeable safety boundary value is as follows:

[0015] The difference between the constant 1 and the maximum load rate is used as the first reference value;

[0016] The product of the rated output power at the current engine speed and the first reference value is used as the predicted rechargeable safety boundary value for a future preset period; if the rechargeable safety boundary value is less than 0, then the rechargeable safety boundary value is forced to be 0.

[0017] Furthermore, the method for obtaining the remaining cruise duration is as follows:

[0018] Obtain the weighted value of the preset planned average ground speed and the average cruising ground speed within the current sliding window, and use it as the initial reference ground speed;

[0019] The maximum value between the initial reference ground speed and the preset safe cruise ground speed threshold is used as the reference ground speed for the compound wing UAV.

[0020] The ratio of the remaining total distance of the mission to the reference ground speed is used as the predicted remaining cruise time.

[0021] Furthermore, the method for obtaining the remaining battery power is as follows:

[0022] When the current high-voltage battery pack is not charging, the power consumption rate per unit time in the fixed-wing cruise mode is the maximum value between the power consumption rate per unit time obtained by fitting historical data and 0.

[0023] When the current high-voltage battery pack is in a charging state, the power consumption rate per unit time in the fixed-wing cruise mode is the power consumption rate per unit time obtained by fitting the historical data of the most recent non-charging state.

[0024] The product of the power consumption rate per unit time in the fixed-wing cruise mode and the predicted remaining cruise duration is used as the first power consumption analysis value.

[0025] The product of the power consumption rate per unit time of the rotor mode and the fixed duration of the mode transition and vertical landing of the fixed wing to rotor mode is used as the second power consumption analysis value.

[0026] Subtract the first power consumption analysis value from the actual remaining power of the high-voltage battery pack at the current moment, and then subtract the second power consumption analysis value to obtain the predicted remaining battery power at the end of the mission landing.

[0027] Furthermore, the method for obtaining the urgency level of battery replenishment is as follows:

[0028] The difference between the preset expected remaining battery power safety threshold and the predicted remaining battery power is used as the first value;

[0029] Calculate the ratio of the first value to the preset lower limit range of the power buffer to obtain the first characteristic value;

[0030] The maximum value between the first eigenvalue and the constant 0 is used as the battery energy replenishment analysis value;

[0031] The minimum value between the constant 1 and the battery replenishment analysis value is used as the battery replenishment urgency level from the current sliding window to the end of the task.

[0032] Furthermore, the method for obtaining the charging compatibility coefficient is as follows:

[0033] Input the real-time temperature of the high-voltage battery pack into a preset temperature segmentation mapping function to obtain the temperature adaptation score.

[0034] The real-time internal resistance of the high-voltage battery pack is input into a preset internal resistance segmentation mapping function to obtain the internal resistance adaptation score.

[0035] Input the real-time remaining power of the high-voltage battery pack into the preset power segmentation mapping function to obtain the power matching score;

[0036] The weighted sum of the temperature adaptation score, internal resistance adaptation score, and power adaptation score is used as the charging compatibility coefficient at the current moment.

[0037] The rules of the preset temperature segmented mapping function are as follows: when the real-time temperature is within the preset normal temperature range, the output value is 1; when the real-time temperature is below the lower limit of the preset normal temperature range, the output value increases linearly from 0 to 1 as the temperature rises; when the real-time temperature is above the upper limit of the preset normal temperature range, the output value decreases linearly from 1 to 0 as the temperature rises; and when the real-time temperature exceeds the preset safe working range, the output value is 0.

[0038] The preset rules for the internal resistance segmentation mapping function are as follows: based on the battery's nominal internal resistance, when the real-time internal resistance is less than or equal to the nominal internal resistance, the output value is 1; when the real-time internal resistance is greater than the nominal internal resistance and less than or equal to 1.5 times the nominal internal resistance, the output value decreases linearly from 1 to 0 as the internal resistance increases; when the real-time internal resistance is greater than 1.5 times the nominal internal resistance, the output value is 0.

[0039] The rules of the preset power segmentation mapping function are as follows: when the real-time remaining power is within the preset optimal charging and discharging range, the output value decreases linearly from 1.0 to 0.2 as the remaining power increases; when the real-time remaining power is below the lower limit of the preset optimal charging and discharging range, the output value increases linearly from 0.25 to 0.38 as the remaining power increases; when the real-time remaining power is above the upper limit of the preset optimal charging and discharging range, the output value decreases linearly from 0.2 to 0 as the remaining power increases.

[0040] Furthermore, the method for obtaining the actual charging power is as follows:

[0041] The product of the urgency of battery charging and the charging adaptability coefficient is used as the charging control index at the current moment.

[0042] When the charging control index is greater than the preset charging control threshold, the charging process of the high-voltage battery pack is started. The rechargeable safety boundary value is multiplied by the charging control index to obtain the optimal charging power at the current moment.

[0043] The optimal charging power, the rated maximum power generation of the generator unit, and the maximum allowable charging power of the high-voltage battery pack are compared, and the minimum value among them is taken as the actual charging power.

[0044] Furthermore, the method for performing a pre-flight safety power check based on a preset takeoff power threshold before flight mode transition is as follows:

[0045] Before the hybrid-wing UAV takes off, the real-time remaining power of the high-voltage battery pack is obtained for analysis of the remaining power.

[0046] When the remaining battery power is greater than or equal to the preset safe takeoff battery power threshold, the compound wing UAV is allowed to take off.

[0047] When the remaining battery power is less than the preset safe takeoff battery power threshold, a takeoff prohibition warning signal is sent to the flight control computer, and a ground charging prompt is given. The preset safe takeoff battery power threshold is obtained by looking up the takeoff test data of the UAV under typical mission load or the historical flight test database. It covers at least the cumulative power consumption from rotor climb, forward transition to entering stable fixed-wing cruise and leaves a safety margin.

[0048] Furthermore, the method for predicting the engine's maximum load rate within a preset future time period based on current sliding window data is as follows:

[0049] A sliding window dataset containing engine load rate time series data is constructed using a fixed-duration sliding window and a fixed sliding step size.

[0050] For the engine load rate time series data within the current sliding window, a linear regression model is used to fit its trend line, and the maximum positive residual between the actual load rate and the trend line within the sliding window is calculated.

[0051] The predicted value of the linear regression model within a future preset time period is added to the maximum positive residual, and the sum is taken as the maximum load rate of the engine load rate within the future preset time period.

[0052] Secondly, another embodiment of the present invention provides a hybrid power management system for a compound-wing unmanned aerial vehicle, the system comprising: a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the computer program, it implements the steps of any of the above methods.

[0053] The present invention has the following beneficial effects:

[0054] This invention first determines that the current flight mode is fixed-wing cruise mode, ensuring that charging permission is only granted during safe flight phases with stable power redundancy and no drastic power fluctuations, thus avoiding interference from indiscriminate charging to high-risk flight modes from the outset. To cope with the transient impact of complex high-altitude airflow disturbances and sudden flight maneuvers on engine load, it predicts the engine's maximum load rate within a preset future time period based on current sliding window data. Combined with the engine's rated output power, it calculates the safe charging boundary value that does not occupy flight power, accurately identifying the absolutely safe redundant power available for power generation, completely eliminating the stall risk caused by charging preempting flight power. Furthermore, to accurately assess the UAV's terminal maneuvers, including reverse transitions and vertical landing, this invention... The system determines the underlying energy demand and then predicts the remaining cruise time based on the total remaining mission mileage and the average cruise ground speed within the current sliding window. This effectively suppresses prediction jumps caused by high-altitude gusts. Simultaneously, it combines historical power consumption rates to predict the remaining battery power at the end of the mission's descent, completely decoupling the power consumption prediction logic from the charging execution action, thus eliminating high-frequency oscillation deadlocks in the control closed loop at its source. Furthermore, it combines a preset safety threshold for the expected remaining battery power at descent to obtain the urgency of battery recharging, quantifying the absolute power gap from the current point to the mission's end and ensuring a rapid response to the gap. This achieves on-demand recharging, avoiding ineffective blind overcharging or descent crashes due to insufficient recharging. To avoid forced high-power charging under inefficient or extremely dangerous electrochemical conditions... The real-time temperature, internal resistance, and remaining capacity of the high-voltage battery pack are then input into corresponding preset piecewise mapping functions. A weighted calculation yields the charging adaptability coefficient for the current moment, comprehensively considering the coupled effects of thermal safety, health degradation, and capacity margin on the charging process. This effectively prevents severe battery overheating and overcurrent lithium deposition, significantly extending the cycle life of the rotor-specific battery. To achieve a globally optimal balance between flight dynamics safety, the urgent need for end-of-life recharge, and battery physical health, the actual charging power is obtained based on rechargeable safety boundary values, battery recharge urgency, and the charging adaptability coefficient to perform cross-modal recharge. This achieves adaptive and smooth scaling of charging power under multi-dimensional dynamic constraints, effectively utilizing the mechanical energy wasted during the fixed-wing cruise phase. The system effectively converts energy into life-saving power for the rotor, significantly improving the overall energy utilization rate across modes. Considering that the absolute survivability of UAVs is highly dependent on the underlying energy guarantee during high-risk phases of mode transitions and extreme sudden failures, a pre-emptive power check based on a preset safe power threshold for takeoff is performed before flight mode transitions. When a power failure is detected, emergency control is triggered, and the power management module physically cuts off the power supply link of non-critical loads to ensure rotor power supply. This forcibly eliminates the hidden danger of energy shortage before takeoff from the ground and constructs a bottom-level physical power outage protection line in the event of loss of main power, maximizing the power supply for rotor emergency landings and greatly improving the operational reliability and emergency landing survival rate of UAVs in complex industrial scenarios. Attached Figure Description

[0055] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0056] Figure 1 A schematic flowchart illustrating a hybrid power energy management method for a compound-wing unmanned aerial vehicle (UAV) according to an embodiment of the present invention;

[0057] Figure 2 A structural diagram of a hybrid power energy management system for a compound-wing unmanned aerial vehicle provided in one embodiment of the present invention;

[0058] Figure 3 This is a schematic diagram of a computer device provided according to an embodiment of the present invention. Detailed Implementation

[0059] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a hybrid power energy management method and system for a compound-wing unmanned aerial vehicle (UAV) according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0060] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0061] The following description, in conjunction with the accompanying drawings, details a specific solution for a hybrid power energy management method and system for a compound-wing unmanned aerial vehicle (UAV) provided by the present invention.

[0062] Example 1:

[0063] The specific scenario of this embodiment is a hybrid electric vehicle (HEV) with a piston engine driving a fixed-wing cruise and a high-voltage battery pack dedicated to driving a rotor for vertical takeoff and landing. The HEV system comprises: a power unit including a piston engine driving the thrust propeller in fixed-wing flight mode; and at least four rotor motors driving the rotor propellers in multi-rotor vertical takeoff and landing mode. An energy storage unit includes a low-voltage battery pack, such as a 6S lithium battery (nominal voltage 22.2V), to power the engine's electronic control unit, starter motor, and flight control system; and a high-voltage battery pack, such as a 24S battery pack consisting of four 6S lithium batteries connected in series (nominal voltage 88.8V), dedicated to powering the four rotor motors. The low-voltage battery pack is charged by drawing power from the high-voltage battery pack or the power generation unit through a DC-DC converter built into the power management module. The power generation unit is an integrated starter-generator whose rotor is coaxially connected to the output shaft of the piston engine. This generator can be used as a starter motor during startup or as a generator during flight, converting the engine's mechanical energy into electrical energy to charge the high-voltage battery pack via the power management module. The power management module, as the core controller of the system, communicates with the flight control computer via a bus and is connected to the sampling interfaces of the battery management systems of each battery pack, the engine's data interface, and the flight control computer's status data interface. The power management module has a built-in microprocessor and energy management control algorithms, responsible for monitoring, distributing, and managing the entire system's electrical energy. In emergency fault mode, the power management module cuts off the piston engine's ignition signal or fuel pump power supply, stopping the fixed-wing propulsion system's power output and simultaneously cutting off the power supply links to non-critical mission payloads, prioritizing the power supply to the rotor propulsion system and flight control system.

[0064] This invention proposes a hybrid power energy management method for compound-wing unmanned aerial vehicles (UAVs). Please refer to [link / reference]. Figure 1 The diagram illustrates a schematic flowchart of a hybrid power energy management method for a compound-wing unmanned aerial vehicle (UAV) according to an embodiment of the present invention. The method includes the following steps:

[0065] Step S1: Acquire real-time data on the power unit, energy storage unit, and flight status of the compound wing UAV to determine the current flight mode.

[0066] Specifically, it is known that compound-wing UAVs employ a decoupled dual-power architecture, with a gasoline-powered system driving fixed-wing cruise and an electric system driving rotor vertical takeoff and landing. Throughout the entire flight cycle, the operating status of the power source, power demand characteristics, core energy management objectives, and safety constraints differ significantly across different flight phases. If a single energy management strategy is applied to cover the entire flight process, it will be unable to adapt to the dynamic needs of each phase, leading to improper energy allocation or flight safety risks. Therefore, this embodiment acquires real-time data on the power unit, energy storage unit, and flight status of the compound-wing UAV to accurately determine the current flight mode of the UAV. This provides a basis for subsequent tiered energy management decisions, ensuring that power output and energy allocation during flight always adapt to the current flight requirements.

[0067] It should be noted that, based on the UAV's flight status and power operation mode, this embodiment divides its entire flight cycle into the following seven modes: 1. Ready-to-fly mode: The UAV is in a ground-based power-off or ready-to-start state, with no power output requirement, and energy management is in hibernation or self-test mode; 2. Rotor climb / hover operation mode: Powered by the high-voltage battery pack, the rotor motor drives the UAV to climb vertically or hover in the air. This mode has high power requirements and a fast power consumption rate. The core energy management objective is to limit the depth of discharge and avoid over-discharge; 3. Forward transition mode: The UAV transitions from rotor flight to fixed-wing flight. The rotor gradually decelerates, the piston engine starts and accelerates, the power system is in a dynamic switching process, power fluctuations are severe, safety priority is highest, and charging is prohibited; 4. Fixed-wing cruise mode: The piston engine continuously drives the fixed-wing... Cruise mode: The rotor motors are in standby or off state, and the engines have stable redundant power. This is the only mode with safe charging conditions. The core energy management objective is to efficiently recover redundant energy to replenish the high-voltage battery pack. Reverse transition mode: The drone transitions from fixed-wing to rotorcraft flight. The piston engine slows down or stops, and the rotor motors are re-energized and gradually take over lift. Power demand fluctuates greatly, safety is a high priority, and charging is also prohibited. Rotor landing mode: Powered by the high-voltage battery pack, the rotor motors drive the drone to descend vertically. Power demand gradually decreases, and the core energy management objective is to ensure sufficient remaining power for landing. Emergency fault mode: Triggered when an engine, generator, or critical power system failure is detected. The energy management objective is to cut off unnecessary loads and lock power to prioritize rotor landing. The power requirements and energy management objectives differ for each mode.

[0068] In summary, the current mode is determined by real-time data, and dynamic charging management is only implemented when the mode is determined to be fixed-wing cruise mode; if the mode is determined to be other modes, charging of the high-voltage battery pack is prohibited to ensure flight safety.

[0069] Step S2: When in fixed-wing cruise mode, predict the maximum load rate of the engine within a preset time period based on the current sliding window data, and calculate the rechargeable safety boundary value that does not occupy flight power in combination with the rated output power of the engine.

[0070] Specifically, considering that the engine load rate of a compound-wing UAV is not constant during the fixed-wing cruise phase, but is affected by various complex operating conditions such as high-altitude gusts, airflow disturbances, heading, altitude fine-tuning, and atmospheric density changes, exhibiting high-frequency and large-amplitude instantaneous fluctuations, directly using the load rate at the current moment to calculate the rechargeable redundant power is prone to drastic changes in charging power due to instantaneous fluctuations, or even misjudging that there is no charging space, thus failing to stably and safely recover energy. Therefore, in order to accurately assess the engine's true redundancy capability over a future period and avoid interference from instantaneous disturbances on charging decisions, this embodiment predicts the engine's maximum load rate over a preset future period based on the current sliding window data when in fixed-wing cruise mode. Combined with the engine's rated output power, it calculates the rechargeable safety boundary value that does not occupy flight power, accurately reflecting the maximum redundant power that the engine can use for power generation without affecting flight power requirements. The larger the rechargeable safety boundary value, the more abundant the redundant energy available for charging in the current and future periods, and the higher the safety and feasibility of charging.

[0071] Preferably, in one feasible embodiment of this method, the method for predicting the maximum engine load rate within a preset time period based on the current sliding window data is as follows: First, a sliding window dataset containing time-series engine load rate data is constructed using a sliding window of fixed duration and a fixed sliding step size. In this embodiment, the fixed duration of the sliding window is set to 10 seconds to ensure the real-time nature of the prediction results. The implementer can set the fixed duration according to the engine response characteristics and flight scenario, which is not limited here. The sliding step size is set to 50 milliseconds to ensure that data refresh and re-prediction are completed every 50 milliseconds, so that the prediction results can closely follow the changes in engine operating conditions. The implementer can set the sliding step size according to the system computing resources and prediction accuracy requirements, which is not limited here.

[0072] To accurately extract the trend of engine load changes and avoid interference from single-point noise, a linear regression model is used to fit the trend of engine load rate time-series data within the current sliding window, and the maximum positive residual between the actual load rate and the trend line within the sliding window is calculated. It should be noted that linear regression can effectively smooth instantaneous fluctuations and reflect the overall trend of the load rate; this is common knowledge and will not be elaborated further. To assess the load demand under the worst-case scenario and ensure that charging power does not encroach on the flight power safety boundary, the predicted values ​​of the linear regression model within a preset future time period are added to the maximum positive residual. The sum is used as the upper limit of the engine load rate fluctuation range within the preset future time period. Since the load rate will reach the upper limit of the range under the worst-case scenario, using this upper limit as the maximum load rate to calculate redundant power ensures that even if the maximum load fluctuation within the predicted range occurs in the future, the charging operation will not affect flight power safety. In this embodiment, the future preset time period is set to 30 seconds. This duration matches the typical disturbance cycle of UAV attitude adjustment and heading correction, and can provide sufficient foresight for charging decisions while ensuring safety. Implementers can set the future preset time period according to the dynamic characteristics of UAV and mission requirements, and there are no restrictions here.

[0073] Preferably, in one feasible embodiment of this method, the rechargeable safety boundary value is obtained as follows: Given the rated output power of the engine at the current speed (this value can be obtained from the engine's native parameters or bench test data), to quantify the proportion of power available for power generation remaining after meeting the worst-case future operating conditions, the difference between a constant 1 and the maximum load rate is used as a first reference value, accurately reflecting the redundancy proportion of engine power not occupied by the worst-case future operating conditions. To convert this redundancy proportion into a specific power value for subsequent charging power calculation, the product of the rated output power of the engine at the current speed and the first reference value is used as the predicted rechargeable safety boundary value for a future preset time period, accurately reflecting the redundant electrical power that the engine can stably output without compromising flight power safety, provided that the predicted load fluctuation limit is met. Considering that when the predicted maximum load rate is close to or exceeds 100%, the calculated rechargeable safety boundary value may be negative, which physically represents no redundant power and will interfere with the normal execution of subsequent charging logic. Therefore, if the rechargeable safety boundary value is less than 0, the rechargeable safety boundary value is forced to be 0 to ensure that the rechargeable safety boundary value is always non-negative, thereby ensuring the numerical stability and logical correctness of the subsequent charging decision algorithm.

[0074] Step S3: Based on the remaining total mileage of the mission and the average cruising ground speed within the current sliding window, predict the remaining cruise time, combine the historical power consumption rate to predict the remaining battery power at the end of the mission landing, and combine the preset expected landing remaining power safety threshold to obtain the battery recharging urgency.

[0075] Specifically, to predict in advance whether the rotor battery's charge level is sufficient for a safe landing during the fixed-wing cruise phase and to avoid landing failure due to insufficient charge, this embodiment first predicts the remaining cruise time based on the remaining total mission mileage and the average cruise ground speed within the current sliding window. It then combines this prediction with historical power consumption rates to predict the remaining battery charge at the time of landing. To quantify and compare the predicted remaining battery charge with the safety target, a preset expected landing remaining battery charge safety threshold is further used to determine the urgency of battery recharging. This accurately reflects the size of the gap between the rotor battery charge level and the expected safety threshold between the current moment and the end of the mission. A higher battery recharging urgency indicates a more severe charge gap at landing, requiring more urgent charging.

[0076] Preferably, in one feasible embodiment of this invention, the remaining cruise time is obtained as follows: Considering the complexity of aviation meteorological conditions, predicting the remaining long-range flight time solely based on instantaneous or short-term average ground speed results in nonlinear errors. Therefore, a preset planned average ground speed is introduced as a priori benchmark, and it is weighted and fused with the average cruise ground speed measured in real-time within a sliding window to obtain an initial reference ground speed. In this embodiment, the weight of the preset planned average ground speed is set to 0.7, and the weight of the average cruise ground speed is set to 0.3. This weighting mechanism can effectively suppress flight time prediction jumps caused by high-altitude gusts, enabling energy management decisions to maintain the stability of macroscopic flight path planning while possessing real-time response capabilities. Considering that under extreme weather conditions such as strong headwinds, the initial reference ground speed of the UAV may approach zero or even be negative, directly using the initial reference ground speed as the denominator to calculate the remaining cruise time will cause division by zero errors or generate unreasonable maxima, leading to distortion in subsequent battery power prediction. Therefore, to ensure the numerical stability and robustness of the calculations, this embodiment sets a preset safe cruise ground speed threshold based on the minimum controllable ground speed of the UAV or mission safety requirements to avoid division-by-zero errors caused by the ground speed approaching 0 in headwind conditions. That is, the preset safe cruise ground speed threshold is a preset constant that is strictly greater than zero, for example, setting the preset safe cruise ground speed threshold to 5 m / s, without further limitation. Then, the maximum value between the initial reference ground speed and the preset safe cruise ground speed threshold is used as the reference ground speed of the compound-wing UAV, ensuring that the reference ground speed is always positive and not lower than the safety lower limit. To accurately estimate the flight time required for the UAV to complete the remaining route, the ratio of the remaining total mission distance to the reference ground speed is used as the predicted remaining cruise duration.

[0077] Preferably, in one feasible embodiment of this invention, the method for obtaining the remaining battery power is as follows: Considering that in fixed-wing cruise mode, the high-voltage battery pack may be in a non-charging state (standby or low-current discharge) or a charging state, the net power consumption characteristics in the two states are completely different. If the power consumption rate per unit time is obtained by directly fitting historical power data, the fitted value will be negative in the charging state (i.e., the power increases). Directly substituting this into the power prediction formula will lead to an overestimation of the predicted power at the end, masking the actual power shortage. Therefore, it is necessary to perform boundary correction on the power consumption rate per unit time in fixed-wing cruise mode.

[0078] When the high-voltage battery pack is not charging, it means the battery is not receiving energy from the generator. In this case, the net power consumption rate should be non-negative. To avoid negative power consumption rates due to sensor noise or data fluctuations during the fitting process, the unit-time power consumption rate for the fixed-wing cruise mode is set to the maximum value between the unit-time power consumption rate obtained from historical data fitting and zero; that is, the larger of the fitted value and zero. The method for obtaining the unit-time power consumption rate from historical data fitting is as follows: Select the time-series data of the remaining battery charge under the fixed-wing cruise mode from the current sliding window, use the least squares method for linear fitting to obtain the slope of the charge change over time, and take the absolute value of the slope as the power consumption rate. The least squares method is well-known and will not be elaborated further.

[0079] When the high-voltage battery pack is currently charging, it means the generator is injecting energy into the battery, and the battery's net charge is actually increasing. If the data within the current sliding window is used to fit the power consumption rate (usually a negative value), it will incorrectly predict a higher end-point charge than the actual charge, causing the urgency of recharging to instantly drop to zero and incorrectly cutting off charging. Once charging is cut off, the actual flight power consumption rate reappears, and the expected charge drop will immediately trigger charging again, causing the relay and charging circuit to fall into a severe "charge-off-charge" high-frequency oscillation deadlock, making it impossible to complete effective recharging. To ensure that the prediction model can always reflect the actual flight power consumption benchmark of the UAV and maintain the absolute stability of the control system, the unit-time power consumption rate of the fixed-wing cruise mode is taken as the unit-time power consumption rate obtained by fitting the historical data of the most recent non-charging state. This introduces a state memory mechanism, whereby the system "remembers" the actual power consumption level of the aircraft before charging and uses it as a conservative estimate benchmark for future power consumption. This mechanism completely decouples the power consumption prediction logic from the actual charging execution action, eliminating the hidden danger of system oscillation deadlock from the root and ensuring the continuity and safety of the charging process.

[0080] To quantify the total power consumption during the fixed-wing cruise phase, the product of the power consumption rate per unit time in the fixed-wing cruise mode and the predicted remaining cruise duration is used as the first power consumption analysis value, representing the power that the rotor battery may consume from the current moment until the end of the cruise. Considering that the UAV needs to perform reverse conversion (fixed-wing to rotor) and vertical landing after completing the cruise, both of which rely entirely on the high-voltage battery pack for power and have high power requirements, in order to reserve this part of the power, the product of the power consumption rate per unit time in the rotor mode and the preset fixed-wing to rotor mode conversion and vertical landing duration is used as the second power consumption analysis value. The power consumption rate per unit time for rotor mode is obtained by fitting historical rotor flight data. Specifically, the remaining battery power time series data under historical rotor climb, hover, and landing modes are selected, and the power consumption rate per unit time is obtained by linear regression fitting. In addition, the fixed duration of fixed-wing to rotor mode transition and vertical landing is set through typical mission tests or industry experience, for example, 60 seconds. This duration is sufficient to cover the entire process from deceleration in cruise mode, switching power and stable hovering to landing, ensuring sufficient reserved power. Implementers can set the fixed duration of fixed-wing to rotor mode transition and vertical landing according to the UAV model and operating environment, and there is no limitation here.

[0081] To obtain the remaining battery power before landing at the end of the mission, the actual remaining power of the high-voltage battery pack at the current moment is subtracted by a first power consumption analysis value, and then by a second power consumption analysis value, to obtain the predicted remaining battery power at the end of the mission. It should be noted that the predicted remaining battery power comprehensively considers the power consumption during cruise and the power consumption during the transition to landing, and excludes the interference of the charging state, providing a reliable basis for recharging decisions.

[0082] Preferably, in one feasible way of this embodiment, the method for obtaining the battery charging urgency level is as follows: In order to set an engineering-safe power preservation target that is strictly higher than the lower limit of battery hardware discharge, this embodiment sets a preset expected landing remaining power safety threshold of 30% (this value is only an example, and in actual applications it can be adjusted within the range of 20% to 50% according to the task level, battery characteristics and ambient temperature. Implementers can set the preset expected landing remaining power safety threshold according to specific needs, but it must be strictly greater than zero, and is not limited here). The preset expected landing remaining power safety threshold ensures that the UAV still has enough power to cope with unexpected situations such as sudden airflow and go-around when landing, fundamentally avoiding the deadlock problem of "sufficient power but refusing to charge" caused by using the lower limit of hardware discharge as the target. That is, the preset expected landing remaining power safety threshold must be significantly higher than the lower limit of hardware discharge. If the lower limit of hardware is used as the target, the urgency level will always be 0 before the power drops to the critical point, and the system will never start charging.

[0083] To quantify the gap between the current predicted battery level and the expected safety threshold, the difference between the preset expected remaining battery level safety threshold and the predicted remaining battery level is obtained as a first value. When the first value is positive, it indicates that the predicted battery level is lower than the safety threshold, and there is a battery gap; when the first value is negative or zero, it indicates that the battery level is sufficient. To enable the system to respond quickly to the battery gap and avoid invalid trickle charging, the ratio of the first value to the preset lower limit of the battery buffer is calculated to obtain a first characteristic value, reflecting the proportion of the gap within the sensitive buffer. In this embodiment, the preset lower limit of the battery buffer is set to 5% (in practical applications, it can be 5% to 10%) to ensure that when the battery gap reaches or exceeds this buffer, the first characteristic value can quickly approach or exceed 1, thereby triggering high-power charging. At the same time, to avoid division by zero errors, the preset lower limit of the battery buffer is strictly greater than zero when calculating the ratio.

[0084] To ensure that the urgency level does not become negative (negative values ​​represent no shortage and should be uniformly set to zero), the maximum value between the first characteristic value and the constant 0 is used as the battery replenishment analysis value, thus clamping negative values ​​to 0. To limit the battery replenishment urgency level to between 0 and 1, the minimum value between the constant 1 and the battery replenishment analysis value is used as the battery replenishment urgency level from the current sliding window to the end of the task. The battery replenishment urgency level ranges from 0 to 1; a higher urgency level indicates a more urgent need for replenishment.

[0085] Step S4: Input the real-time temperature, internal resistance and remaining power of the high-voltage battery pack into the corresponding preset piecewise mapping function, and calculate the charging adaptability coefficient at the current moment by weighting.

[0086] Specifically, considering that the charging acceptance capacity, energy conversion efficiency, and safety risks of lithium batteries are significantly affected by their own physical state (temperature, internal resistance, and remaining capacity), forced charging under extreme temperature or high internal resistance conditions can easily lead to lithium deposition or thermal runaway. To avoid blind charging in inefficient or unsafe conditions, protect battery life, and maximize energy recovery efficiency, the real-time temperature, internal resistance, and remaining capacity of the high-voltage battery pack are input into corresponding preset piecewise mapping functions, and a weighted calculation is performed to obtain the charging adaptability coefficient at the current moment. This accurately reflects the comprehensive capacity of the current high-voltage battery pack to receive charging energy at both the electrochemical and physical levels. The larger the charging adaptability coefficient, the more suitable the current state of the battery is for charging, and the higher the corresponding energy conversion efficiency and the lower the safety risk.

[0087] Preferably, in one feasible embodiment, the method for obtaining the charging adaptability coefficient is as follows: The real-time temperature of the high-voltage battery pack is input into a preset temperature segmented mapping function to obtain a temperature adaptability score, which characterizes the safety of the current thermal environment for charging; the real-time internal resistance of the high-voltage battery pack is input into a preset internal resistance segmented mapping function to obtain an internal resistance adaptability score, which characterizes the degree of charging heat loss caused by battery health degradation; the real-time remaining capacity of the high-voltage battery pack is input into a preset capacity segmented mapping function to obtain a capacity adaptability score, which characterizes the redundancy acceptance space of the battery's current capacity for charged energy; to comprehensively evaluate the coupled influence of thermal safety, health, and capacity margin on the charging process, the weighted sum of the temperature adaptability score, internal resistance adaptability score, and capacity adaptability score is used as the charging adaptability coefficient at the current moment; wherein, the calculation formula for the charging adaptability coefficient is: In the formula, C is the charging adaptability coefficient at the current moment; T is the temperature adaptability score; R is the internal resistance adaptability score; and D is the power adaptability score. It is the first weight; As the second weight; As the third weight. This embodiment sets... It is 0.5. It is 0.3. The value is set to 0.2 to ensure that temperature, which directly determines the risk of thermal runaway during charging, is given the highest priority. This is followed by internal resistance, which characterizes the internal losses during charging, and finally, the charge level, which constrains the risk of overcharging. Implementers can set this value based on the cell chemistry characteristics of the selected battery and the heat dissipation capacity of the UAV's thermal management system. , and No specific restrictions are imposed here;

[0088] The rules of the preset temperature segmented mapping function are as follows: when the real-time temperature is within the preset room temperature range, the output value is 1; when the real-time temperature is below the lower limit of the preset room temperature range, the output value increases linearly from 0 to 1 as the temperature rises; when the real-time temperature is above the upper limit of the preset room temperature range, the output value decreases linearly from 1 to 0 as the temperature rises; when the real-time temperature exceeds the preset safe operating range, the output value is 0. In this embodiment, the preset room temperature range is set to 15°C to 35°C to ensure that the electrochemical reaction of the lithium battery is most active, the charging acceptance is strongest, and there is almost no risk of lithium plating within this range. Implementers can set the preset room temperature range according to the charge and discharge temperature curve specified in the specific cell specification and the actual working meteorological environment of the drone, which is not limited here.

[0089] The preset rules for the piecewise mapping function of internal resistance are as follows: based on the nominal internal resistance of the battery, when the real-time internal resistance is less than or equal to the nominal internal resistance, the output value is 1; when the real-time internal resistance is greater than the nominal internal resistance and less than or equal to 1.5 times the nominal internal resistance, the output value decreases linearly from 1 to 0 as the internal resistance increases; when the real-time internal resistance is greater than 1.5 times the nominal internal resistance, the output value is 0. It should be noted that the 1.5 times nominal internal resistance is set here because the increase in battery internal resistance will lead to a significant increase in ohmic heat generation (Joule heat) during the charging process. When the internal resistance reaches 1.5 times the nominal internal resistance, its charging energy loss and temperature rise rate are close to the limit of the heat dissipation system of the drone's closed battery compartment. If high-power charging is forcibly continued in this state, it will cause serious heat accumulation and accelerate the electrochemical aging of the battery, and even induce the risk of thermal runaway. Therefore, 1.5 times the nominal internal resistance is used as the physical safety threshold for refusing charging.

[0090] The rules for the preset charge / discharge segmentation mapping function are as follows: when the real-time remaining charge / discharge is within the preset optimal charge / discharge range, the output value linearly decreases from 1.0 to 0.2 as the remaining charge / discharge increases; when the real-time remaining charge / discharge is below the lower limit of the preset optimal charge / discharge range, the output value linearly increases from 0.25 to 0.38 as the remaining charge / discharge increases; when the real-time remaining charge / discharge is above the upper limit of the preset optimal charge / discharge range, the output value linearly decreases from 0.2 to 0 as the remaining charge / discharge increases. This embodiment sets the preset optimal charge / discharge range to 20% to 80% to ensure the highest energy conversion efficiency during charging within this voltage plateau period and effectively avoids the severe polarization regions during deep discharge and near-full charge. Implementers can set the preset optimal charge / discharge range according to the battery BMS management strategy and system cycle life requirements; no limitation is imposed here.

[0091] Step S5: Based on the rechargeable safety boundary value, the urgency of battery replenishment, and the charging adaptability coefficient, obtain the actual charging power to perform cross-mode replenishment; before the flight mode transition, perform a pre-emptive power verification based on a preset takeoff safety power threshold, and trigger emergency control when a power failure is detected, with the power management module physically cutting off the power supply link of non-critical loads to ensure rotor power supply.

[0092] Specifically, it is known that the fixed-wing cruise mode of a hybrid-wing UAV has stable power redundancy, while the rotor vertical landing and hovering operations are highly dependent on independent power supply from high-voltage batteries and consume a large amount of power. Relying solely on the initial power at ground takeoff is prone to power depletion after long-duration or complex weather operations. Therefore, this embodiment obtains the actual charging power based on rechargeable safety boundary values, battery replenishment urgency, and charging adaptability coefficients to perform cross-modal energy replenishment. This achieves efficient and intelligent conversion of redundant mechanical energy in the fixed-wing cruise phase into rotor-specific electrical energy, fundamentally improving the cross-modal energy utilization efficiency of the hybrid power system. Considering that the absolute survivability of the UAV is highly dependent on the energy security mechanism during high-risk phases (such as mode transitions) and extreme sudden failures, if the power... Maintaining non-essential power supply to the entire system when in danger or without main power will quickly deplete the last life-saving energy, leading to a crash. Therefore, before the flight mode transition, a pre-emptive power check based on a preset takeoff safety power threshold is performed to ensure that the UAV has absolutely reliable underlying power support during the entire high-risk, high-energy-consuming phase of takeoff and rotor-to-fixed-wing transition. This completely eliminates the hidden danger of power failure and stall upon takeoff. When a power failure is detected, emergency control is triggered to immediately isolate the damaged main power unit and lock the remaining power of the high-voltage battery pack. The power management module physically cuts off the power supply link of non-critical loads to ensure rotor power supply. This establishes a bottom line for energy safety throughout the entire life cycle, including normal power replenishment and extreme failure conditions, and significantly improves the operational reliability and emergency landing survival rate of the UAV.

[0093] Preferably, in one feasible method of this embodiment, the actual charging power is obtained as follows: Considering that a single dimension of power shortage or battery physical state cannot fully determine the scientificity and safety of charging decisions, in order to achieve a globally optimal balance between ensuring flight power safety, meeting terminal landing requirements, and protecting battery health and lifespan, the product of the battery replenishment urgency and the charging adaptability coefficient is used as the charging control index at the current moment. This accurately reflects the comprehensive priority of performing charging operations at the current moment in terms of both urgency and physical adaptability, which is beneficial for the system to make intelligent decisions on whether to charge and how much to charge in a complex and ever-changing air environment. The larger the charging control index, the more urgent the terminal replenishment demand and the more suitable the current health and thermal state of the battery is for high-power energy absorption. Therefore, this embodiment sets a preset charging control threshold of 0.3 to ensure that the charging circuit is activated only when the comprehensive priority reaches a certain benchmark, avoiding frequent start-stop of relays and DC-DC converters in inefficient or unnecessary states, which would cause device fatigue and wear. The implementer can set the preset charging control threshold according to the energy conversion efficiency curve of the power generation system and the circuit switching loss characteristics, which is not limited here.

[0094] When the charging control index exceeds the preset charging control threshold, it indicates that not only are the mechanical and physical conditions for safe charging met, but the battery also has an urgent need for energy replenishment and its electrochemical state is suitable for receiving electrical energy. In order to recover redundant energy in a timely manner, the charging process of the high-voltage battery pack is initiated at this time. In order to make the actual allocated charging power adaptively adjusted according to the charging urgency and the dynamic improvement of the battery state, and to avoid the risk of thermal runaway caused by a one-size-fits-all full charging, the charging safety boundary value is multiplied by the charging control index to obtain the optimal charging power at the current moment, realizing intelligent and smooth scaling of the charging power within the dynamic safety boundary range. Considering that even if the optimal charging power output by the algorithm is safe in terms of power redundancy logic, it may still exceed the tolerance limit of the relevant power generation and distribution hardware under extreme boundary conditions, in order to build the final hardware-level physical protection network, the optimal charging power, the rated maximum power generation of the generator unit, and the maximum allowable charging power of the high-voltage battery pack are compared, and the minimum value is taken as the actual charging power, which completely eliminates hardware damage accidents such as generator overload overheating and high-voltage battery pack overcurrent lithium deposition.

[0095] Preferably, in one feasible embodiment of this method, the pre-flight safety power threshold-based power verification before flight mode transition is performed as follows: In order to cut off the risk of energy shortage before takeoff and avoid the loss of thrust due to high-voltage battery depletion during the energy-intensive forward transition phase after takeoff, the real-time remaining power of the high-voltage battery pack is obtained before the compound wing UAV takes off, and analyzed as remaining power. When the analyzed remaining power is greater than or equal to the preset takeoff safety power threshold, it indicates that the current energy storage of the high-voltage battery pack is sufficient to support the entire process of full-load takeoff, high-angle climb, and smooth transition to complete piston engine takeoff, at which point it is allowed to proceed. The takeoff of the compound-wing UAV ensures that the mode transition during the initial takeoff phase can be completed smoothly and safely. When the remaining battery power is less than the preset safe takeoff battery power threshold, it indicates that if takeoff is forced, there is a high probability of a catastrophic loss of rotor power in mid-air. At this time, a takeoff prohibition warning signal is sent to the flight control computer, prompting ground charging to forcibly eliminate the high-risk hazard at a stationary state on the ground. The preset safe takeoff battery power threshold is obtained by looking up the takeoff test data of the UAV under typical mission loads or the historical flight test database. It covers at least the cumulative power consumption from rotor climb, forward transition to entering stable fixed-wing cruise and leaves a safety margin.

[0096] In summary, this embodiment acquires UAV status data in real time to determine the flight mode; when in fixed-wing cruise mode, it predicts the maximum engine load rate based on the current sliding window and calculates the rechargeable safety boundary value that does not occupy flight power; it predicts the remaining cruise duration and the remaining battery power at the end of the mission landing, and combines the safety threshold to obtain the battery recharging urgency; it inputs the real-time battery temperature, internal resistance, and remaining power into a mapping function and weights them to obtain the charging adaptability coefficient; finally, it combines the above boundary value, urgency level, and adaptability coefficient to obtain the actual charging power to perform cross-modal recharging, and performs power verification before mode transition and cuts off non-critical loads in case of failure. This invention realizes intelligent collaborative allocation of power energy across modes, significantly improving the energy utilization rate and flight safety of UAVs.

[0097] Example 2:

[0098] This invention also proposes a hybrid power energy management system for compound-wing unmanned aerial vehicles (UAVs). Please refer to [link to relevant documentation]. Figure 2 The diagram illustrates a hybrid power energy management system for a compound-wing unmanned aerial vehicle (UAV) according to an embodiment of the present invention. The system includes: a data acquisition module 10, a rechargeable safety boundary value acquisition module 20, a battery recharge urgency acquisition module 30, a charging adaptability coefficient acquisition module 40, and a management module 50.

[0099] The data acquisition module 10 is used to acquire data on the power unit, energy storage unit and flight status of the compound wing UAV in real time, and to determine the current flight mode.

[0100] The rechargeable safety boundary value acquisition module 20 is used to predict the maximum load rate of the engine within a preset time period based on the current sliding window data when in fixed-wing cruise mode, and calculate the rechargeable safety boundary value that does not occupy flight power in combination with the rated output power of the engine.

[0101] The battery replenishment urgency acquisition module 30 is used to predict the remaining cruise time based on the remaining total mileage of the mission and the average cruise ground speed within the current sliding window, predict the remaining battery power at the end of the mission landing by combining the historical power consumption rate, and obtain the battery replenishment urgency by combining the preset expected landing remaining power safety threshold.

[0102] The charging adaptability coefficient acquisition module 40 is used to input the real-time temperature, internal resistance and remaining power of the high-voltage battery pack into the corresponding preset piecewise mapping function, and calculate the charging adaptability coefficient at the current moment by weighting.

[0103] The management module 50 is used to obtain the actual charging power based on the rechargeable safety boundary value, the urgency of battery replenishment, and the charging adaptability coefficient to perform cross-mode replenishment; before the flight mode conversion, it performs a pre-emptive power verification based on a preset takeoff safety power threshold; and when a power failure is detected, it triggers emergency control, and the power management module physically cuts off the power supply link of non-critical loads to ensure rotor power supply.

[0104] It should be noted that the system provided in the above embodiments is only an example of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above. In addition, the hybrid power energy management system for compound-wing UAVs and the hybrid power energy management method for compound-wing UAVs provided in the above embodiments belong to the same concept, and their specific implementation process can be found in the method embodiment, which will not be repeated here.

[0105] Example 3:

[0106] This invention also proposes a hybrid power energy management device for a compound-wing unmanned aerial vehicle (UAV). The device includes a memory and a processor. The memory stores executable program code, and the processor calls and executes the executable program code to perform a hybrid power energy management method for a compound-wing UAV provided in the embodiments of this application. Specifically, the device may be a chip, component, or module. The chip may include a connected processor and memory; the memory stores instructions, and when the processor calls and executes the instructions, the chip can perform the hybrid power energy management method for a compound-wing UAV provided in the above embodiments.

[0107] In addition, this embodiment also protects a computer device; please refer to [link to relevant documentation]. Figure 3 The computer device includes a memory 401, a processor 402, and a computer program 403 stored in the memory 401 and running on the processor 402, wherein when the processor 402 executes the computer program 403, the computer device is able to execute any of the hybrid power energy management methods for compound-wing unmanned aerial vehicles described above.

[0108] Example 4:

[0109] The present invention also provides a computer-readable storage medium storing computer program code, which, when executed on a computer, causes the computer to perform the aforementioned method steps to implement the hybrid power energy management method for a compound-wing unmanned aerial vehicle provided in the above embodiments.

[0110] Example 5:

[0111] The present invention also provides a computer program product that, when run on a computer, causes the computer to perform the aforementioned related steps to implement the hybrid power energy management method for a compound-wing unmanned aerial vehicle provided in the above embodiments.

[0112] In this embodiment, the device, computer-readable storage medium, computer program product, or chip are all used to execute the corresponding methods provided above. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding methods provided above, and will not be repeated here.

[0113] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0114] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A hybrid power energy management method for a compound-wing unmanned aerial vehicle, characterized in that, The method includes the following steps: Real-time acquisition of data on the power unit, energy storage unit, and flight status of the compound wing UAV to determine the current flight mode; When in fixed-wing cruise mode, the maximum load rate of the engine is predicted based on the current sliding window data within a preset period in the future. Combined with the rated output power of the engine, the rechargeable safety boundary value that does not occupy flight power is calculated. The remaining cruise time is predicted based on the total remaining mileage of the mission and the average cruise ground speed within the current sliding window. The remaining battery power at the end of the mission landing is predicted by combining the historical power consumption rate. The urgency of battery recharging is obtained by combining the preset expected landing remaining power safety threshold. The real-time temperature, internal resistance, and remaining capacity of the high-voltage battery pack are input into the corresponding preset piecewise mapping function, and the weighted calculation is used to obtain the charging adaptability coefficient at the current moment. Based on the rechargeable safety boundary value, the urgency of battery replenishment, and the charging adaptability coefficient, the actual charging power is obtained to perform cross-mode replenishment; before the flight mode conversion, a pre-emptive power verification based on a preset takeoff safety power threshold is performed, and when a power failure is detected, emergency control is triggered, and the power management module physically cuts off the power supply link of non-critical loads to ensure rotor power supply.

2. The hybrid power energy management method for a compound-wing unmanned aerial vehicle as described in claim 1, characterized in that, The method for obtaining the rechargeable safety boundary value is as follows: The difference between the constant 1 and the maximum load rate is used as the first reference value; The product of the rated output power at the current engine speed and the first reference value is used as the predicted rechargeable safety boundary value for a future preset period of time. If the rechargeable safety boundary value is less than 0, then the rechargeable safety boundary value is forced to be 0.

3. The hybrid power energy management method for a compound-wing unmanned aerial vehicle as described in claim 1, characterized in that, The method for obtaining the remaining cruise duration is as follows: Obtain the weighted value of the preset planned average ground speed and the average cruising ground speed within the current sliding window, and use it as the initial reference ground speed; The maximum value between the initial reference ground speed and the preset safe cruise ground speed threshold is used as the reference ground speed for the compound wing UAV. The ratio of the remaining total distance of the mission to the reference ground speed is used as the predicted remaining cruise time.

4. The hybrid power energy management method for a compound-wing unmanned aerial vehicle as described in claim 1, characterized in that, The method for obtaining the remaining battery power is as follows: When the current high-voltage battery pack is not charging, the power consumption rate per unit time in the fixed-wing cruise mode is the maximum value between the power consumption rate per unit time obtained by fitting historical data and 0. When the current high-voltage battery pack is in a charging state, the power consumption rate per unit time in the fixed-wing cruise mode is the power consumption rate per unit time obtained by fitting the historical data of the most recent non-charging state. The product of the power consumption rate per unit time in the fixed-wing cruise mode and the predicted remaining cruise duration is used as the first power consumption analysis value. The product of the power consumption rate per unit time of the rotor mode and the fixed duration of the mode transition and vertical landing of the fixed wing to rotor mode is used as the second power consumption analysis value. Subtract the first power consumption analysis value from the actual remaining power of the high-voltage battery pack at the current moment, and then subtract the second power consumption analysis value to obtain the predicted remaining battery power at the end of the mission landing.

5. A hybrid power energy management method for a compound-wing unmanned aerial vehicle as described in claim 1, characterized in that, The method for obtaining the urgency level of battery charging is as follows: The difference between the preset expected remaining battery power safety threshold and the predicted remaining battery power is used as the first value; Calculate the ratio of the first value to the preset lower limit range of the power buffer to obtain the first characteristic value; The maximum value between the first eigenvalue and the constant 0 is used as the battery energy replenishment analysis value; The minimum value between the constant 1 and the battery replenishment analysis value is used as the battery replenishment urgency level from the current sliding window to the end of the task.

6. A hybrid power energy management method for a compound-wing unmanned aerial vehicle as described in claim 1, characterized in that, The method for obtaining the charging compatibility coefficient is as follows: Input the real-time temperature of the high-voltage battery pack into a preset temperature segmentation mapping function to obtain the temperature adaptation score. The real-time internal resistance of the high-voltage battery pack is input into a preset internal resistance segmentation mapping function to obtain the internal resistance adaptation score. Input the real-time remaining power of the high-voltage battery pack into the preset power segmentation mapping function to obtain the power matching score; The weighted sum of the temperature adaptation score, internal resistance adaptation score, and power adaptation score is used as the charging compatibility coefficient at the current moment. The rules of the preset temperature segmented mapping function are as follows: when the real-time temperature is within the preset normal temperature range, the output value is 1; when the real-time temperature is below the lower limit of the preset normal temperature range, the output value increases linearly from 0 to 1 as the temperature rises; when the real-time temperature is above the upper limit of the preset normal temperature range, the output value decreases linearly from 1 to 0 as the temperature rises; and when the real-time temperature exceeds the preset safe working range, the output value is 0. The preset rules for the internal resistance segmentation mapping function are as follows: based on the battery's nominal internal resistance, when the real-time internal resistance is less than or equal to the nominal internal resistance, the output value is 1; when the real-time internal resistance is greater than the nominal internal resistance and less than or equal to 1.5 times the nominal internal resistance, the output value decreases linearly from 1 to 0 as the internal resistance increases; when the real-time internal resistance is greater than 1.5 times the nominal internal resistance, the output value is 0. The rules of the preset power segmentation mapping function are as follows: when the real-time remaining power is within the preset optimal charging and discharging range, the output value decreases linearly from 1.0 to 0.2 as the remaining power increases; when the real-time remaining power is below the lower limit of the preset optimal charging and discharging range, the output value increases linearly from 0.25 to 0.38 as the remaining power increases; when the real-time remaining power is above the upper limit of the preset optimal charging and discharging range, the output value decreases linearly from 0.2 to 0 as the remaining power increases.

7. A hybrid power energy management method for a compound-wing unmanned aerial vehicle as described in claim 1, characterized in that, The method for obtaining the actual charging power is as follows: The product of the urgency of battery charging and the charging adaptability coefficient is used as the charging control index at the current moment. When the charging control index is greater than the preset charging control threshold, the charging process of the high-voltage battery pack is started. The rechargeable safety boundary value is multiplied by the charging control index to obtain the optimal charging power at the current moment. The optimal charging power, the rated maximum power generation of the generator unit, and the maximum allowable charging power of the high-voltage battery pack are compared, and the minimum value among them is taken as the actual charging power.

8. A hybrid power energy management method for a compound-wing unmanned aerial vehicle as described in claim 1, characterized in that, The method for performing a pre-flight safety power check based on a preset takeoff power threshold before flight mode conversion is as follows: Before the hybrid-wing UAV takes off, the real-time remaining power of the high-voltage battery pack is obtained for analysis of the remaining power. When the remaining battery power is greater than or equal to the preset safe takeoff battery power threshold, the compound wing UAV is allowed to take off. When the remaining battery power is less than the preset safe takeoff battery power threshold, a takeoff prohibition warning signal is sent to the flight control computer, and a ground charging prompt is given. The preset safe takeoff battery power threshold is obtained by looking up the takeoff test data of the UAV under typical mission load or the historical flight test database. It covers at least the cumulative power consumption from rotor climb, forward transition to entering stable fixed-wing cruise and leaves a safety margin.

9. A hybrid power energy management method for a compound-wing unmanned aerial vehicle as described in claim 1, characterized in that, The method for predicting the maximum load rate of the engine within a preset time period based on the current sliding window data is as follows: A sliding window dataset containing engine load rate time series data is constructed using a fixed-duration sliding window and a fixed sliding step size. For the engine load rate time series data within the current sliding window, a linear regression model is used to fit its trend line, and the maximum positive residual between the actual load rate and the trend line within the sliding window is calculated. The predicted value of the linear regression model within a future preset time period is added to the maximum positive residual, and the sum is taken as the maximum load rate of the engine load rate within the future preset time period.

10. A hybrid power energy management system for a compound-wing unmanned aerial vehicle, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the hybrid power energy management method for a compound-wing unmanned aerial vehicle as described in any one of claims 1-9.

Images