Power supply method and system for multi-functional emergency lighting drone systems

Abstract

This invention belongs to the field of power diversion, and particularly relates to a diversion power supply method and system for multi-functional emergency lighting UAV systems. The method collects the state of charge of the onboard energy storage device, the real-time power consumption of the flight controller, and the target brightness parameters of the LED lighting. A mode arbitration signal is generated through threshold comparison, corresponding to three power supply modes. A multi-channel power distribution control unit achieves coordinated power management of the three channels, and output feedback enables smooth mode switching and overload protection. This invention significantly improves the stability and response speed of the onboard power supply system, ensures the safety of flight controller power supply, balances charging and lighting needs, and enhances the reliability of UAV emergency operations.

Description

Technical Field

[0001] This invention belongs to the field of power diversion, and particularly relates to a power diversion method and system for use in multifunctional emergency lighting drone systems. Background Technology

[0002] In the scenario of powering tethered multi-functional emergency lighting drones, the ground power supply provides relatively stable power through cables. However, there is intense dynamic competition among the three factors: onboard lithium battery charging, flight power consumption, and high-power LED array lighting. The power flow and priority requirements are very different during takeoff and climb, hovering lighting, and return-to-base charging. Furthermore, rapid protection and power redistribution are required under abnormal operating conditions. In existing technologies, one type of solution adopts a passive shunt architecture with AC-DC conversion and a fixed voltage bus. When the flight load changes abruptly or the lighting power increases suddenly, it cannot balance response speed and distribution accuracy. In order to be compatible with constant current and constant voltage charging of lithium batteries and constant current driving of LEDs, an additional independent DC-DC module is often required, which significantly increases the airborne size and weight, sacrifices payload and hovering time, and the high margin design results in low efficiency under normal conditions. Moreover, it is impossible to implement fine current limiting and power stripping for multiple loads on the same bus. Another type of solution uses a ground programmable power supply and simple airborne communication to achieve remote voltage regulation. Although it improves the overall power regulation capability, global voltage regulation cannot simultaneously meet the contradictory requirements of high voltage stability of flight control and low current ripple of LEDs. Loose coupling of the ground-to-air control loop causes sluggish response to high power demand. A sudden drop in bus voltage can easily cause flight control reset or LED extinguishing. At the same time, it lacks real-time analysis and mode arbitration of flight power consumption and lighting demand, and cannot smoothly switch between charging priority, lighting priority and balanced power supply modes. The aforementioned defects lead to frequent occurrences of lighting flicker, battery over-discharge, or flight control undervoltage protection landings during actual system operation. These problems are dynamically exacerbated by changes in cable voltage drop with flight altitude and temperature-induced LED characteristic drift. Therefore, the specific technical problems that urgently need to be solved can be summarized as follows: how to construct an integrated power supply topology that combines single-stage high-voltage boost transmission with multi-stage efficient airborne conversion, breaking the limitations of passive current shunting and global voltage regulation on a single bus, under the constraints of adaptive compensation for tethered cable voltage drop and strict airborne weight limitations; how to design a multi-channel independent current limiting control mechanism on a single DC bus that can simultaneously ensure the amplitude and recovery time of flight control voltage drops, LED peak current ripple rate, and the constant current and voltage accuracy of charging management, considering the differentiated transient characteristics of flight control, battery, and lighting loads; and how to integrate flight power consumption prediction and real-time lighting power analysis to establish a mode arbitration strategy, enabling the system to perform fine-grained, priority-aware dynamic power allocation and seamless switching between three conflicting objectives—charging priority, lighting priority, and balanced power supply—under conditions without inrush current or bus jitter. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention proposes a power distribution method and system for multi-functional emergency lighting UAV systems. This method acquires high-voltage DC bus power, collects real-time state-of-charge (SOC) parameters of the onboard energy storage device, real-time flight control power consumption parameters, and LED target brightness commands. It compares the SOC with a preset charging threshold and the brightness command with a lighting priority threshold to generate a mode arbitration signal indicating charging priority, lighting priority, or balanced power supply mode. In balanced mode, it prioritizes maintaining stable flight control power supply voltage and allocates available power margin to the charging and lighting channels according to a weighted ratio determined by the SOC and brightness command. The onboard multi-channel power distribution control unit performs current limiting control on the first and second output channels and voltage regulation control on the third output channel based on the mode signal. It collects the output voltage and current of each channel and feeds them back to the power distribution algorithm to update the mode signal, achieving smooth switching between the three modes. Furthermore, it performs shutdown or power callback when any channel experiences overcurrent. This invention enables dynamic priority power allocation for flight, charging, and lighting, ensuring flight control stability while balancing charging speed and lighting effect.

[0004] To achieve the above objectives, the present invention provides the following technical solution:

[0005] Power diversion methods applied to multi-functional emergency lighting drone systems include:

[0006] Obtain a high-voltage DC bus power supply and collect in real time the state of charge parameters of the airborne energy storage device, the real-time power consumption parameters of the UAV flight control unit, and the target brightness command parameters of the LED lighting load;

[0007] The state of charge parameter is compared with a preset charging threshold, and the target brightness command parameter is compared with a preset lighting priority threshold. A mode arbitration signal is generated based on the comparison result. When the comparison result meets the charging priority trigger condition, the charging priority mode is indicated. When the comparison result meets the lighting priority trigger condition, the lighting priority mode is indicated. The remaining comparison results indicate the balanced power supply mode. The balanced power supply mode distributes the available power margin between the first power output channel and the second power output channel according to the weighted ratio determined by the state of charge parameter and the target brightness command parameter, and maintains the output voltage of the third power output channel within a preset voltage regulation range.

[0008] The high-voltage DC bus power supply is input to the airborne multi-channel power distribution control unit. The multi-channel power distribution control unit performs current limiting control on the first power output channel and the second power output channel and voltage regulation control on the third power output channel according to the mode indicated by the mode arbitration signal, so as to distribute the power of the high-voltage DC bus power supply.

[0009] The real-time output voltage and real-time output current of the first power output channel, the second power output channel and the third power output channel are collected and fed back to the power allocation algorithm to update the total power demand, available power margin and mode arbitration signal, and continuously control the switching operation between the charging priority mode, the lighting priority mode and the balanced power supply mode.

[0010] Specifically, when the mode arbitration signal indicates the charging priority mode, the multi-channel power distribution control unit sets the current limiting threshold of the first power output channel to a first preset current limiting value, and adjusts the output voltage clamping value or output current limiting value of the second power output channel to a limit value associated with the available power margin.

[0011] When the mode arbitration signal indicates the lighting priority mode, the multi-channel power distribution control unit adjusts the output current of the second power output channel to the target current value that matches the target brightness command parameter, sets the current limiting threshold of the first power output channel to the second preset current limiting value or shuts down the first power output channel, and adjusts the loop response of the multi-channel power distribution control unit based on the real-time output voltage of the third power output channel so that the output voltage of the third power output channel is maintained within the preset voltage regulation range.

[0012] Specifically, when the mode arbitration signal indicates the balanced power supply mode, the multi-channel power distribution control unit determines the minimum dynamic power consumption value based on the real-time power consumption parameters, controls the output power of the third power output channel to be no less than the minimum dynamic power consumption value, calculates the remaining power obtained by deducting the allocated power of the third power output channel from the available power margin, and allocates the remaining power to the first power output channel and the second power output channel according to the weighted ratio.

[0013] The multi-channel power distribution control unit is also used to shut down the corresponding channel or perform power callback when the real-time output current of any power output channel exceeds the overcurrent protection threshold of the corresponding channel.

[0014] Specifically, a pattern arbitration signal is generated based on the comparison results, including:

[0015] The State of Charge (SOC) parameter is compared with a preset charging threshold SOC_th. When SOC is less than SOC_th, a mode arbitration signal indicating a charging priority mode is generated. When SOC is greater than or equal to SOC_th, the target brightness command parameter is compared with a preset lighting priority threshold. If the target brightness command parameter reaches or exceeds the preset lighting priority threshold, a mode arbitration signal indicating a lighting priority mode is generated. Otherwise, a mode arbitration signal indicating a balanced power supply mode is generated.

[0016] Specifically, the process of determining the weighting ratio includes:

[0017] Read the remaining power value obtained after deducting the allocated power of the third power output channel from the high-voltage DC bus power supply;

[0018] Based on the state of charge parameters, determine the maximum allowable charging power and the minimum effective charging power corresponding to the current state of charge;

[0019] Based on the target brightness command parameters and the forward voltage drift compensation value under the current junction temperature conditions, the target driving power required for the LED lighting load to reach the target illuminance is determined, and the minimum effective lighting power is determined based on the human eye flicker perception threshold.

[0020] Specifically, the process of determining the weighting ratio also includes:

[0021] The maximum allowable charging power and the rated safe power of the LED lighting load are respectively determined as the upper limit boundary of the allocated power of the first power output channel and the upper limit boundary of the allocated power of the second power output channel. The minimum effective charging power and the minimum effective lighting power are respectively determined as the lower limit boundary of the allocated power of the first power output channel and the lower limit boundary of the allocated power of the second power output channel.

[0022] The remaining power value is compared with the sum of the lower limit boundary of the first power output channel and the lower limit boundary of the second power output channel, and the sum of the upper limit boundary of the first power output channel and the upper limit boundary of the second power output channel.

[0023] Specifically, the process of determining the weighting ratio also includes:

[0024] When the remaining power value is within the range of the sum of the lower limit boundary to the sum of the upper limit boundary, the initial weighting ratio is calculated by the ratio of the inverse proportional function value of the state of charge parameter to the direct proportional function value of the target brightness command parameter.

[0025] When the remaining power value is lower than the sum of the lower limit boundaries, the initial weighting ratio is corrected according to the preset priority bias coefficient so that the allocation result first satisfies the allocation power lower limit boundary corresponding to the channel with higher priority.

[0026] When the remaining power value is higher than the sum of the upper limit boundaries, the allocated power of the first power output channel is limited to the upper limit boundary of the allocated power of the first power output channel, the allocated power of the second power output channel is limited to the upper limit boundary of the allocated power of the second power output channel, and the power exceeding the sum of the upper limit boundaries is marked as a power margin to be allocated pool.

[0027] Specifically, the power margin allocation pool is configured to gradually transfer accumulated power to the first power output channel, the process including:

[0028] When the accumulated value of the power margin to be allocated pool continuously exceeds the preset accumulated upper limit threshold and the duration reaches the preset duration, an auxiliary charging enable signal is generated;

[0029] Obtain the state of charge (SOC) parameters of the airborne energy storage device for the current control cycle, and determine the auxiliary charging pulse duty cycle for the current control cycle based on the difference between the SOC parameters and the fully charged SOC.

[0030] Specifically, the process of gradually transferring the power accumulated in the power margin allocation pool to the first power output channel further includes:

[0031] Based on the auxiliary charging pulse duty cycle and the preset auxiliary charging pulse frequency, a set of intermittent auxiliary charging pulse sequences is generated. The intermittent auxiliary charging pulse sequences are applied to the current limiting control loop of the first power output channel to superimpose an intermittent power margin consumption current component on the basis of the preset current limiting value of the first power output channel.

[0032] During the execution of the intermittent auxiliary charging pulse sequence, the cumulative value of the power margin allocation pool is continuously monitored. When the cumulative value of the power margin allocation pool drops below a preset cumulative lower limit threshold, the auxiliary charging enable signal is revoked, the output of the intermittent auxiliary charging pulse sequence is stopped, and the current limiting control of the first power output channel is restored to the normal state where the power margin consumption current component is not superimposed.

[0033] A power distribution system for use in multi-functional emergency lighting drone systems includes:

[0034] The parameter acquisition module is configured to: obtain a high-voltage DC bus power supply and acquire in real time the state of charge parameters of the airborne energy storage device, the real-time power consumption parameters of the UAV flight control unit, and the target brightness command parameters of the LED lighting load;

[0035] The power budget module is configured to: compare the state of charge parameter with a preset charging threshold, compare the target brightness command parameter with a preset lighting priority threshold, and generate a mode arbitration signal based on the comparison result. When the comparison result meets the charging priority trigger condition, it indicates the charging priority mode; when it meets the lighting priority trigger condition, it indicates the lighting priority mode; and other comparison results indicate the balanced power supply mode. The balanced power supply mode distributes the available power margin between the first power output channel and the second power output channel according to the weighted ratio determined by the state of charge parameter and the target brightness command parameter, and maintains the output voltage of the third power output channel within a preset voltage regulation range.

[0036] The multi-channel power distribution control module is configured to: input the high-voltage DC bus power supply to the airborne multi-channel power distribution control unit, wherein the multi-channel power distribution control unit performs current limiting control on the first power output channel and the second power output channel and voltage regulation control on the third power output channel according to the mode indicated by the mode arbitration signal, so as to distribute the power of the high-voltage DC bus power supply;

[0037] The real-time output voltage and real-time output current of the first power output channel, the second power output channel and the third power output channel are collected and fed back to the power allocation algorithm to update the total power demand, available power margin and mode arbitration signal, and continuously control the switching operation between the charging priority mode, the lighting priority mode and the balanced power supply mode.

[0038] Compared with the prior art, the beneficial effects of the present invention are:

[0039] This invention collects the state-of-charge (POC) parameters of an airborne energy storage device and the target brightness command parameters of an LED lighting load, and generates a mode arbitration signal with mode memory by performing hysteresis comparisons with preset charging thresholds and preset lighting priority thresholds. The charging priority trigger condition introduces a dual determination: the falling edge is below the preset charging threshold minus the first hysteresis width, and the current mode is not charging priority. The lighting priority trigger condition introduces a dual determination: the rising edge is above the preset lighting priority threshold plus the second hysteresis width, and the current mode is not lighting priority. This effectively eliminates mode oscillations near the thresholds, making the mode arbitration decision more stable and reliable. In balanced power supply mode, available power margin is allocated according to a weighted ratio determined by the POC parameters and the target brightness command parameters. Upper and lower limits for the allocated power are constructed using the maximum allowable charging power, the minimum effective charging power, the target driving power after junction temperature compensation, and the minimum effective lighting power, respectively. The system employs a constraint mechanism. When the remaining power falls within the sum of the upper and lower limits, the initial weighting ratio is calculated using the ratio of the inverse proportionality function of the state of charge to the direct proportionality function of the target brightness. If the remaining power falls below the sum of the lower limits, a priority bias coefficient is used to correct and ensure the lower limit of the channel with higher priority. If the remaining power exceeds the sum of the upper limits, the excess power is marked as a power margin pool to be allocated and subsequently transferred to the first power output channel using an intermittent pulse duty cycle negatively correlated with the state of charge. This achieves fine-grained, priority-aware dynamic power scheduling for three loads on a single DC bus, ensuring stable power supply voltage for the flight control system while balancing charging speed and lighting output quality. During mode switching, the current limiting threshold is gradually adjusted within a preset transition time using a preset gradient function, and the output voltage of the third power output channel is monitored in real time to trigger the flight control priority protection interruption. This achieves smooth, shock-free switching between the three power supply modes and overload safety protection. Attached Figure Description

[0040] Figure 1 This is a flowchart of the power diversion method for a multi-functional emergency lighting drone system according to Embodiment 1 of the present invention;

[0041] Figure 2 This is a flowchart illustrating the determination of the weighting ratio based on the state of charge parameters and the target brightness command parameters in Embodiment 2 of the present invention.

[0042] Figure 3 This is a flowchart illustrating how the power accumulated in the power margin allocation pool is gradually transferred to the first power output channel in Embodiment 3 of the present invention.

[0043] Figure 4 This is a block diagram of the power supply method for a multi-functional emergency lighting drone system according to Embodiment 5 of the present invention. Detailed Implementation

[0044] Example 1

[0045] Please see Figure 1 One embodiment of the present invention provides a power diversion method for a multi-functional emergency lighting drone system, comprising:

[0046] S1. Obtain ground AC mains power input, boost it to the preset transmission voltage level through a step-up transformer module, and then process it through an inverter and rectifier filter module to obtain a high-voltage DC bus power supply; at the same time, collect the state of charge parameters of the airborne energy storage device, the real-time power consumption parameters of the UAV flight control unit, and the target brightness command parameters of the LED lighting load in real time.

[0047] In one optional implementation, the step-up transformer module is positioned between the ground AC mains input terminal and the inverter and rectifier filter module. It is used to boost the voltage level of the ground single-phase or three-phase AC mains from the conventional mains level to a preset transmission voltage level, thereby reducing line loss current and improving power transmission efficiency during long-distance transmission of the tethered cable. Specifically, the step-up transformer module is implemented using a high-frequency step-up transformer with a ferrite core or an amorphous alloy core. The primary winding receives the high-frequency square wave AC power output from the full-bridge inverter after the DC bus voltage has undergone active power factor correction in the preceding stage. The secondary winding boosts the voltage to a preset transmission voltage level of 800V to 1000V by adjusting the turns ratio. A triple-insulated wire and an electrostatic shielding layer are used between the primary and secondary windings to meet high-voltage isolation and safe creepage distance requirements. The transformer's operating frequency is set within the range of 50kHz to 100kHz, significantly reducing the transformer's size and weight compared to a power frequency transformer, facilitating integration into the ground power supply box.

[0048] In one specific embodiment, the preset transmission voltage level is set to 800V to 1000V, which is based on a comprehensive consideration of the matching relationship between the diameter, weight, and transmission distance of the tethered cable. In this embodiment, the tethered cable uses special alloy wire of AWG18 to AWG20 specifications, with a length of 100 meters to 200 meters, and the resistance per unit length of the cable is approximately 0.02 ohms per meter to 0.03 ohms per meter. If the transmission voltage level is selected too low, the current in the cable will increase, and the line loss power is proportional to the square of the current, which will cause excessive energy to be dissipated in the form of heat in the cable, reducing system efficiency. If the transmission voltage level is selected too high, higher requirements will be placed on the insulation design of the transformer, the withstand voltage level of the power switching devices, and the input withstand voltage of the airborne step-down converter, increasing system cost and size. The compromise of 800V to 1000V achieves an optimal balance between cable weight and insulation safety requirements that is engineering-acceptable, while ensuring a transmission efficiency of over 90%. The transformer's operating frequency is set within the range of 50kHz to 100kHz. This setting is based on the following: if the frequency is too low, the cross-sectional area of ​​the transformer core needs to be increased, leading to an increase in size and weight; if the frequency is too high, the high-frequency loss of the core and the switching loss of the power switching devices will increase significantly, reducing conversion efficiency and exacerbating electromagnetic interference. The 50kHz to 100kHz range can fully utilize the low-loss characteristics of the ferrite core in this frequency band, allowing the transformer's power density to reach its optimal level.

[0049] The inverter and rectifier filter module is located after the step-up transformer module and consists of a cascaded front-stage inverter unit and a rear-stage rectifier filter unit. It converts the high-voltage, high-frequency AC power output from the step-up transformer module into a stable high-voltage DC bus power supply, which is then transmitted to the airborne terminal via a tethered cable. The front-stage inverter unit is implemented using a full-bridge topology composed of four wide-bandgap semiconductor power switching devices, which are silicon carbide metal-oxide-semiconductor field-effect transistors (MOSFETs) or gallium nitride high electron mobility transistors (HNTPs). The drive signal for the full-bridge inverter is generated by the phase-shift pulse width modulation module of the ground-side controller. Soft-switching regulation of the inverter output voltage is achieved by adjusting the phase shift angle between the bridge arms, with zero-voltage turn-on of the switching transistors. The system achieves this through the resonance of the transformer leakage inductance and the parasitic capacitance of the switching transistor. The subsequent rectifier and filter unit is located on the secondary side of the step-up transformer module. It consists of a full-bridge rectifier composed of four high-voltage silicon carbide Schottky diodes and a thin-film capacitor filter network with low equivalent series resistance. The full-bridge rectifier rectifies the high-voltage high-frequency AC output from the secondary side into pulsating DC. After the switching frequency ripple is filtered out by the thin-film capacitor filter network, a stable high-voltage DC bus power supply is output, with the output voltage ripple coefficient controlled within 1%. The inverter and rectifier filter module works in conjunction with the step-up transformer module to form a high-frequency step-up isolated DC-DC converter link at the ground end, realizing the complete power conversion function from AC mains input to high-voltage DC bus output.

[0050] In one specific embodiment, the output voltage ripple factor is controlled to be within 1% based on the following: the high-voltage buck converter in the front-end of the onboard multi-channel power distribution control unit has a certain tolerance range for input voltage ripple, but excessive input voltage ripple will cause the input filter capacitor of the buck converter to generate a large ripple current, accelerating capacitor aging and increasing electromagnetic interference. A 1% ripple factor corresponds to a peak-to-peak ripple voltage of no more than 8V at 800V output and no more than 10V at 1000V output. To achieve this ripple target, the cutoff frequency of the low equivalent series resistance thin-film capacitor filter network is designed to be one-tenth to one-twentieth of the switching frequency. Specifically, the capacitance and equivalent series resistance of the thin-film capacitor are selected so that the filter pole is located at approximately 5kHz. The capacitance of the thin-film capacitor is selected to be between 10µF and 22µF, and the equivalent series resistance is less than 5mΩ, ensuring at least 40dB of attenuation for switching frequency ripple from 50kHz to 100kHz.

[0051] In an optional specific implementation, the method for obtaining the state of charge (SOC) parameter further includes: estimating the SOC of the airborne energy storage device in real time using an extended Kalman filter algorithm through an airborne battery management system; wherein the battery terminal voltage, charging and discharging current, and cell temperature are used as input observations of the extended Kalman filter algorithm, state recursion and covariance update are performed based on the second-order RC equivalent circuit model of the airborne energy storage device, and the estimated SOC value at the current moment is output as the SOC parameter and passed to the power allocation algorithm. In one specific implementation, the specific process of performing state recursion and covariance update based on the second-order RC equivalent circuit model of the airborne energy storage device is as follows: The battery terminal voltage, charging / discharging current, and cell temperature of the airborne energy storage device are used as input observations for the extended Kalman filter algorithm. The state vector of the second-order RC equivalent circuit model consists of the battery open-circuit voltage, the polarization voltage of the first RC stage, and the polarization voltage of the second RC stage. The state recursion equation is established based on the discretized state-space expression of the second-order RC equivalent circuit model. The discretization sampling period is consistent with the recursion period of the extended Kalman filter algorithm, both being 10 milliseconds. Within each recursion period, the posterior state estimate of the previous period and the charging / discharging current at the current moment are input. Based on the state transition matrix and the input matrix, the prior state estimate of the current moment is predicted one step ahead. Simultaneously, the posterior error covariance matrix of the previous period, the Jacobian matrix of the state transition matrix, and the process noise covariance matrix are used to perform matrix multiplication. The prior error covariance matrix at the current moment is calculated using addition and addition operations. Subsequently, the prior state estimate is refreshed using the battery terminal voltage measurement value collected at the current moment. The observed predicted value of the terminal voltage is obtained by subtracting the polarization voltage of the two RC links and the voltage drop across the ohmic internal resistance from the open-circuit voltage. The difference between the observed and predicted terminal voltage values ​​constitutes the refresh sequence. The Kalman gain matrix is ​​jointly calculated from the prior error covariance matrix, the Jacobian matrix of the observation matrix, and the measurement noise covariance matrix. The prior state estimate is corrected by multiplying the Kalman gain matrix by the refresh sequence to obtain the posterior state estimate at the current moment. At the same time, the posterior error covariance matrix is ​​updated for use in the next recursive cycle. The open-circuit voltage component in the posterior state estimate is converted into the current state of charge estimate based on the preset nonlinear mapping relationship between the state of charge of the airborne energy storage device and the open-circuit voltage. This state of charge estimate is output as the state of charge parameter to the power allocation algorithm.In a specific numerical example, a lithium polymer battery pack with a nominal voltage of 22.2 volts and a capacity of 10,000 mAh, operating at a SOC of 35% and a discharge current of 5 amps, has the following parameters in its second-order RC equivalent circuit model: ohmic internal resistance R0 equals 30 milliohms, polarization internal resistance R1 of the first RC stage equals 15 milliohms and polarization capacitance C1 equals 1500 farads, and polarization internal resistance R2 of the second RC stage equals 10 milliohms and polarization capacitance C2 equals 3000 farads. After approximately 50 recursive cycles (0.5 seconds) of convergence, the extended Kalman filter algorithm outputs a steady-state error better than 3% between the estimated state of charge and the actual state of charge, providing an accurate energy reference for determining the weighting ratio.

[0052] In one specific embodiment, the parameters of the second-order RC equivalent circuit model, namely the ohmic internal resistance R0, the polarization internal resistance R1 and polarization capacitance C1 of the first RC stage, and the polarization internal resistance R2 and polarization capacitance C2 of the second RC stage, are obtained as follows: After the airborne energy storage device leaves the factory but before installation, offline parameter identification is performed using a hybrid pulse power characteristic test method. During the test, a discharge pulse and a charging pulse of a certain pulse width and amplitude are applied to a fully charged battery. The ohmic voltage drop of the battery terminal voltage at the instant of pulse application and the voltage relaxation recovery curve after the pulse ends are recorded. The transient voltage drop component of R0, as well as the time constant and polarization voltage components of the two RC stages, are extracted using a least-squares fitting algorithm, thereby obtaining the model parameters at the corresponding state of charge and temperature. This parameter identification process is repeated at multiple state of charge points and multiple temperature points, ultimately forming a two-dimensional lookup table of model parameters changing with state of charge and temperature, which is stored in non-volatile memory for online retrieval by the extended Kalman filter algorithm. The process noise covariance matrix and measurement noise covariance matrix are set by the standard deviation determined by the offline system identification experiment. The setting is based on the following: the measurement noise level of battery terminal voltage and current is recorded in the parameter identification experiment, as well as the statistical characteristics of the model residual. The diagonal element values ​​of the measurement noise covariance matrix and the process noise covariance matrix are determined by calculating the variance of the measurement noise and the variance of the model prediction error, respectively, to ensure the convergence speed and steady-state accuracy of the extended Kalman filter algorithm in actual operation.

[0053] In an optional implementation, the method for obtaining the real-time power consumption parameter of the UAV flight control unit further includes: obtaining the real-time power consumption parameter through a power consumption summation function; specifically, the power consumption summation function collects the real-time current sampling values ​​of the flight control main control chip, inertial measurement unit, barometer, satellite positioning module, data link radio and each servo motor channel, multiplies each real-time current sampling value by the rated voltage of the corresponding power supply branch and sums them to obtain the real-time total power consumption of the UAV flight control unit, which is then passed as the real-time power consumption parameter to the power allocation algorithm.

[0054] In one specific embodiment, the rated voltage values ​​of each power supply branch are as follows: 5.0V for the flight control main control chip power supply branch, 3.3V for the inertial measurement unit power supply branch, 3.3V for the barometer power supply branch, 3.3V for the satellite positioning module power supply branch, 5.0V for the data link radio power supply branch, and 5.0V for each servo channel power supply branch. The real-time current sampling value is detected by a high-precision sampling resistor connected in series in each branch. The resistance of the sampling resistor is 10 milliohms to 50 milliohms. The voltage signal across the sampling resistor is amplified by an instrumentation amplifier and then acquired by the analog-to-digital converter of the airborne microcontroller at a 12-bit resolution and a 1 kHz sampling frequency. The obtained digital value is converted into the actual current value by a calibration coefficient. The power consumption summation function accumulates the current of each branch at a refresh rate of 200 Hz. In each calculation cycle, the current of each branch is multiplied by the corresponding rated voltage and summed, and the unit is converted to watts as the real-time power consumption parameter output. The refresh rate of the power consumption summation function is set to twice the control cycle of 100 Hz to ensure that the flight control power consumption value obtained each time the power allocation algorithm is called reflects the latest current changes.

[0055] In an optional implementation, the method for obtaining the target brightness command parameter of the LED lighting load further includes: obtaining the target brightness command parameter through a dimming mapping function; specifically, the dimming mapping function receives a brightness percentage command issued from a ground station or flight control unit, maps the brightness percentage command to a corresponding target current value or pulse width modulation duty cycle value, and passes it as the target brightness command parameter to the power allocation algorithm.

[0056] In one specific embodiment, the brightness percentage command is sent from the ground station to the flight control unit via a custom message frame using the MAVLink protocol. The flight control unit then forwards the command to the lighting driver microcontroller via a serial peripheral interface or internal integrated circuit bus. The message frame format includes a one-byte brightness percentage field with a value ranging from 0 to 100, corresponding to a brightness command of 0% to 100%. The dimming mapping function is a linear mapping relationship. When the target current value is used as the output, the mapping formula is: target current value equals brightness percentage divided by 100 multiplied by the maximum driving current of the LED lighting module, 8.0 amps. When the pulse width modulation duty cycle value is used as the output, the mapping formula is: duty cycle value equals brightness percentage divided by 100 multiplied by 4095, where 4095 is the maximum duty cycle count value corresponding to a 12-bit pulse width modulation resolution. The dimming mapping function can also embed a temperature compensation correction step. The brightness percentage command is corrected based on the current junction temperature value of the LED array collected by the airborne temperature sensor before the mapping is executed, so as to compensate for the luminous flux drift caused by the change of LED junction temperature. The correction coefficient is determined by the luminous flux and junction temperature relationship curve provided in the LED lamp chip specification.

[0057] S2. Based on the state of charge parameters, real-time power consumption parameters, and target brightness command parameters, calculate the total power demand and available power margin of the current system through a power allocation algorithm. At the same time, compare the state of charge parameters with a preset charging threshold and the target brightness command parameters with a preset lighting priority threshold to generate a mode arbitration signal.

[0058] In one specific embodiment, the total power demand is the sum of the power required by the three types of loads in the system during the current control cycle. Specifically:

[0059] Based on the real-time power consumption parameters of the UAV flight control unit, multiplied by a preset safety margin coefficient, the minimum dynamic power consumption value of the third power output channel is calculated.

[0060] Based on the state of charge parameters of the airborne energy storage device, the maximum allowable charging power corresponding to the current state of charge is obtained by querying the upper limit curve table of acceptable charging power.

[0061] Based on the target brightness command parameters of the LED lighting load and the forward voltage drift compensation value under the current junction temperature conditions, determine the target driving power after junction temperature compensation required to achieve the target illuminance.

[0062] The total power demand for the current control cycle is obtained by summing the minimum dynamic power consumption value, the maximum allowable charging power, and the target drive power after junction temperature compensation.

[0063] For example, when the flight controller's real-time power consumption is 120 watts (minimum dynamic power consumption 144 watts), the maximum allowable charging power corresponding to the SOC is 180 watts, and the target drive power after junction temperature compensation is 114 watts, the total power demand is 438 watts. This value is used to assess the current power supply and demand balance of the system and serves as a benchmark for subsequent calculations of available power margin.

[0064] In one specific embodiment, the preset charging threshold SOC_th is set to 80%. This threshold is set based on the following: the airborne energy storage device uses a lithium polymer battery pack with a nominal voltage of 22.2 volts. Its charging characteristic curve is in a constant current charging stage before the state of charge reaches 80%, allowing for a relatively high charging current. When the state of charge exceeds 80%, it enters a constant voltage charging stage, where the charging current needs to be gradually reduced to prevent overvoltage. Setting SOC_th to 80% ensures that when the battery level is below this threshold, it enters a charging priority mode to quickly replenish the battery with the maximum charging current. When the battery level is above this threshold, charging is no longer the highest priority, preventing rapid charging under high charge conditions from accelerating battery aging or triggering overvoltage protection. The preset lighting priority threshold is set at 70%. This threshold is set based on the following: In emergency lighting missions, when the target brightness command parameter is below 70%, it indicates that the ground operators' lighting brightness requirement is at a low to medium level. In this case, the remaining power can be evenly distributed between charging and lighting, provided that flight control power supply is guaranteed. When the target brightness command parameter reaches or exceeds 70%, it indicates that the mission has a higher demand for lighting brightness, and the system should prioritize ensuring lighting output to meet operational requirements. This threshold is set by those skilled in the art during the system configuration phase based on the flight mission profile and emergency lighting requirements, and is written into non-volatile memory. It can be modified via ground station software before mission execution.

[0065] Specifically, the mode arbitration signal in this embodiment is as follows: The state of charge (SOC) parameter is compared with a preset charging threshold SOC_th. When SOC is less than SOC_th, a mode arbitration signal indicating a charging priority mode is generated. When SOC is greater than or equal to SOC_th, the target brightness command parameter is compared with a preset lighting priority threshold. If the target brightness command parameter reaches or exceeds the preset lighting priority threshold, a mode arbitration signal indicating a lighting priority mode is generated; otherwise, a mode arbitration signal indicating a balanced power supply mode is generated. The balanced power supply mode in this embodiment refers to prioritizing the voltage stability of the power supply from the third power output channel to the UAV flight control unit, and distributing the available power margin between the first and second power output channels according to a weighted ratio determined by the state of charge parameter and the target brightness command parameter.

[0066] In one specific embodiment, the process of obtaining the available power margin is implemented as a sub-step of the power allocation algorithm. This sub-step is executed once in each 100Hz control cycle to determine the upper limit of power resources available for flexible allocation between the first and second power output channels within the current control cycle. The specific execution process is as follows: At the beginning of each control cycle, the microcontroller starts the synchronous sampling channel of the high-voltage side analog-to-digital converter to collect the input voltage and input current of the high-voltage DC bus. After filtering out switching frequency noise by an 8-point moving average filter, the filtered voltage value is multiplied by the current value to obtain the total input power of the high-voltage DC bus power supply for the current control cycle. At the same time, the microcontroller reads the real-time total power consumption of the UAV flight control unit from the output buffer of the power consumption summation function through direct memory access, as the real-time power consumption parameter for the current control cycle. To ensure the absolute priority of flight control power supply, the power allocation algorithm does not directly use this real-time power consumption parameter as the allocated power for the third power output channel. Instead, it dynamically determines a minimum dynamic power consumption value based on this real-time power consumption parameter. This minimum dynamic power consumption value is calculated according to the following formula: the minimum dynamic power consumption value is equal to the current real-time total power consumption multiplied by a preset safety margin coefficient of 1.2. The preset safety margin coefficient of 1.2 is based on the following: Under typical operating conditions of tethered UAV hovering lighting, extensive flight tests under level 4 wind disturbance conditions showed that the maximum ratio of peak power consumption to steady-state power consumption during gusts was approximately 1.18. Rounded down and retaining a small margin, this coefficient was set to 1.2. This coefficient ensures that, under most operating conditions, the calculation of available power margin will not underestimate the dynamic power consumption of the flight controller, thus not crowding out the actual power supply requirements of the flight controller. The microcontroller's hardware arithmetic logic unit then performs a 32-bit signed integer subtraction operation, subtracting the calculated minimum dynamic power consumption value from the total input power of the current control cycle. The difference is the available power margin for the current control cycle. This available power margin, in milliwatts, is written to the available power margin variable in the intermediate calculation result register for subsequent weighted ratio calculations and mode arbitration signal parsing. In balanced power supply mode, the available power margin is allocated between the first and second power output channels according to a weighted ratio determined by the state of charge parameters and the target brightness command parameters. In charging priority mode or lighting priority mode, the available power margin is preferentially scheduled to the corresponding priority channel according to the mode arbitration signal, and the remaining portion is then allocated to another channel. It is understood that the specific values ​​of total input power, real-time total power consumption of flight control, minimum dynamic power consumption, and available power margin listed in this embodiment are all illustrative and are not intended to limit the scope of protection of this invention. In actual implementation, the value of available power margin is updated in real time with the dynamic changes of total input power and flight control power consumption in each control cycle.

[0067] S3. Input the high-voltage DC bus power supply to an on-board multi-channel power distribution control unit. The multi-channel power distribution control unit receives and parses the mode arbitration signal and executes the corresponding shunt control strategy according to the current operating mode indicated by the mode arbitration signal.

[0068] In one optional implementation, the airborne multi-channel power distribution control unit is located in the airborne power management compartment of the tethered UAV. Its input terminal is connected to the high-voltage DC bus power supply via a tether cable and an airborne input filter circuit. Its output terminal is connected to the charging port of the airborne energy storage device through a first power output channel, to the drive input terminal of the LED lighting load through a second power output channel, and to the power supply input terminal of the UAV flight control unit through a third power output channel. This unit is used to convert the high-voltage DC bus voltage transmitted from the ground into a low-voltage DC voltage suitable for three types of loads, and to perform current limiting control and voltage regulation control on each channel in real time according to the mode arbitration signal, so as to realize dynamic power distribution among multiple loads. Specifically, the multi-channel power distribution control unit consists of a front-end high-voltage buck converter and three independent output channels. The front-end high-voltage buck converter is implemented using a dual-transistor active clamp forward converter topology or a phase-shifted full-bridge topology, stepping down the 800V to 1000V high-voltage DC bus input to a 48V or 24V intermediate DC bus. A decoupling network consisting of a large-capacity electrolytic capacitor and a multilayer ceramic capacitor is connected in parallel on the intermediate DC bus to provide transient current response. The three independent output channels all draw power from the intermediate DC bus. The first power output channel consists of a synchronous rectifier buck converter and a programmable current limiting control loop, performing constant current or constant voltage charging control for the onboard energy storage device. Its current limiting threshold is written in real time by the microcontroller of the multi-channel power distribution control unit based on the current mode arbitration signal and available power margin. The second power output channel... The power output channel is composed of a synchronous rectified buck constant current driver with a built-in high-precision current detection amplifier and hysteresis current control loop. It converts the intermediate DC bus voltage into a controlled constant output current. The reference value of the output current is obtained by mapping the target brightness command parameters through a digital-to-analog converter, realizing flicker-free driving of the LED lighting load. The third power output channel is implemented by a low-dropout linear regulator or a synchronous buck regulator composed of a high-gain voltage error amplifier and a power adjustment tube. Its output voltage is set to the rated power supply voltage required by the UAV flight control unit, and a remote voltage sampling feedback network is configured to directly collect the real-time voltage of the flight control power supply port to compensate for the voltage drop of the cables and connectors. At the same time, a transient voltage suppression diode and an energy storage capacitor bank are connected in parallel at the output of this channel to resist the transient drop of the bus voltage caused by sudden changes in flight load. The microcontroller of the multi-channel power distribution control unit communicates with the controller of each channel through an internal integrated circuit bus or serial peripheral interface. In each control cycle, according to the mode arbitration signal, the calculated current limiting threshold and output voltage reference value of each channel are written into the digital register of the corresponding controller. The analog control loop of each channel independently performs closed-loop regulation to realize independent current limiting and voltage regulation control of the three loads on the same intermediate DC bus.

[0069] In one specific embodiment, the intermediate DC bus voltage is set to 48V. This voltage level is set based on the fact that 48V DC bus voltage is lower than the 60V DC safety voltage limit specified in the extra-low voltage standard IEC60364-4-41, avoiding the risk of electric shock to airborne operators. Simultaneously, the 48V voltage level is consistent with the DC bus voltage standard commonly used in industrial fields such as communication base stations and data centers, allowing for the use of mature commercial power modules and controller chips, reducing system development costs. The current limiting threshold of the first power output channel is written in real-time by the microcontroller of the multi-channel power distribution control unit based on the current mode arbitration signal and available power margin. The writing cycle is synchronized with the control cycle at 100Hz. The updated current limiting threshold for each cycle is written to the digital reference register of the synchronous rectifier buck converter via a serial peripheral interface, and executed by the converter's hardware current loop in the next switching cycle. The constant current driver switching frequency of the second power output channel is set to 400 kHz, and the peak-to-peak value of the output current ripple is controlled within 2% of the average value of the output current to meet the current ripple requirements of LED lighting loads and avoid flickering perceptible to the human eye. The rated supply voltage of the third power output channel is set to 5.0 volts, consistent with the main power supply input voltage of the UAV flight control unit. This voltage level is set based on the standard supply voltage of most commercial flight control motherboards. After compensation by the remote voltage sampling feedback network at the flight control power supply port, the actual voltage deviation reaching the input terminal of the flight control motherboard is controlled within ±2%.

[0070] S4. If the mode arbitration signal indicates a charging priority mode, the multi-channel power distribution control unit sets the current limiting threshold of the first power output channel to a first preset current limiting value to perform fast charging on the airborne energy storage device; at the same time, it obtains the available power margin and dynamically adjusts the output voltage clamping value or output current limiting value of the second power output channel according to the available power margin to limit the driving power of the LED lighting load to the level determined by the available power margin.

[0071] In one specific embodiment, the first preset current limit value is set to 10 amps. This current limit value is set based on the following: the airborne energy storage device uses a lithium polymer battery pack with a nominal voltage of 22.2 volts and a capacity of 10,000 mAh, and its specifications stipulate a maximum continuous charging current of 1 coulomb, or 10 amps. Fast charging at 10 amps maximizes charging speed while ensuring battery safety. In charging priority mode, the current limit threshold of the first power output channel is directly set to this preset value of 10 amps, allowing the battery to accept charging energy at the maximum permissible rate. The output voltage clamping value or output current limit value of the second power output channel is dynamically adjusted according to the available power margin rule: the remaining value after subtracting the preset charging power of the first power output channel from the available power margin is used as the maximum permissible output power of the second power output channel, and then converted into the corresponding output voltage clamping value or output current limit value based on the current driving voltage of the LED lighting module. This conversion process assumes that the LED lighting module operates at 25 volts at its rated operating point. The current limiting value is equal to the remaining power divided by 25 volts, while the voltage clamping value is derived by inversely from the current-voltage characteristic curve of the LED chip. This dynamic adjustment ensures that while prioritizing charging needs, the lighting load still receives residual power for auxiliary lighting, rather than being completely shut off.

[0072] In one specific embodiment, the construction of the current-voltage characteristic curve of the LED lighting load is accomplished through a combination of offline testing and curve fitting. First, the LED lighting module is placed in a constant-temperature test chamber, with six temperature points set: -20°C, 0°C, 25°C, 50°C, 65°C, and 85°C. At each temperature point, a programmable DC power supply is used to apply a stepped current to the LED module, increasing from 0 amps to the rated current of 8 amps in 0.5 amp increments. After stabilizing for 2 seconds at each current step, a high-precision digital multimeter is used to simultaneously measure the forward voltage across the LED module, and the corresponding current and voltage data pairs are recorded. After obtaining the discrete data for all temperature and current points, a Shockley diode equation model is used to perform nonlinear least-squares fitting on each set of constant-temperature data. The model is defined as follows: forward voltage equals series equivalent resistance multiplied by current plus ideality factor multiplied by thermal voltage multiplied by current divided by the reverse saturation current plus one natural logarithm, where thermal voltage is calculated from Boltzmann constant, electron charge, and absolute temperature. The series equivalent resistance, ideality factor, and reverse saturation current at various temperatures are obtained through fitting. These parameters are then stored in a characteristic curve parameter table in non-volatile memory according to temperature points. During actual operation, the three model parameters at the current junction temperature are obtained through piecewise linear interpolation. Substituting these parameters into the Shockley diode equation reconstructs the continuous current-voltage characteristic curve under the current junction temperature condition. This provides an accurate electrical parameter reference for back-deriving the corresponding output voltage clamping value from the target drive power in step S604 or for mapping the target brightness command parameter to the corresponding target current value in step S606, thereby ensuring the accuracy of power allocation.

[0073] S5. If the mode arbitration signal indicates a lighting priority mode, the multi-channel power distribution control unit adjusts the output current of the second power output channel to a target current value that matches the target brightness command parameter to meet the power requirements of the LED lighting load; at the same time, it lowers the current limiting threshold of the first power output channel to a second preset current limiting value or shuts off the first power output channel, and monitors the output voltage of the third power output channel in real time, adjusting the loop response of the multi-channel power distribution control unit to maintain the voltage supplied by the third power output channel to the UAV flight control unit within a preset voltage regulation range;

[0074] In one specific embodiment, the second preset current limit value is set to 1.0 amperes. This current limit value is set based on the following: In lighting priority mode, charging of the onboard energy storage device is degraded to a secondary objective, but a minimum trickle charging current still needs to be maintained to compensate for the battery's loss due to self-discharge and static power consumption of the battery management system. 1.0 amperes corresponds to a charging rate of 0.1 coulombs, which falls within the safe trickle charging current range recommended by lithium polymer battery manufacturers, ensuring that the battery does not enter a deep discharge state due to gradual power loss during the continuous lighting priority mode. The preset voltage regulation range is 5.0 volts. 2% The 5V range, or 4.9V to 5.1V, is set based on the allowable voltage range of the main control chip and sensors of the UAV flight control unit. When the output voltage of the third power output channel drops below 4.9V, the voltage loop proportional gain of the multi-channel power distribution control unit increases to twice its normal value, accelerating the loop adjustment response speed to quickly compensate for voltage drops caused by sudden load changes. When the voltage recovers to the preset regulated range, the proportional gain returns to its normal value. This dynamic gain adjustment mechanism ensures a fast voltage recovery while avoiding loop oscillations caused by excessive gain.

[0075] S6. If the mode arbitration signal indicates a balanced power supply mode, the multi-channel power distribution control unit determines the minimum dynamic power consumption value required to maintain the UAV's flight attitude and resist wind disturbance based on the real-time power consumption parameters, and allocates power not lower than the minimum dynamic power consumption value to the third power output channel; and calculates the remaining power of the high-voltage DC bus power supply after deducting the power allocated to the third power output channel, and performs real-time power splitting between the first power output channel and the second power output channel according to the allocation weight determined by the ratio of the state of charge parameter to the target brightness command parameter;

[0076] S7. During the power supply process, the real-time output voltage, real-time output current and temperature parameters of each power output channel are continuously collected, and the real-time output voltage, real-time output current and temperature parameters are fed back to the power allocation algorithm. The power allocation algorithm updates the total demand power, available power margin and mode arbitration signal according to the feedback parameters, so as to control the multi-channel power allocation control unit to perform dynamic smooth switching between the charging priority mode, lighting priority mode and balanced power supply mode, and to perform overload protection when the real-time output current of any power output channel exceeds the overcurrent protection threshold of the corresponding channel.

[0077] In one specific embodiment, the overcurrent protection thresholds for the three power output channels are set as follows: 12 amps for the first power output channel, 10 amps for the second power output channel, and 8 amps for the third power output channel. The 12 amps overcurrent protection threshold for the first power output channel is set based on the channel's preset maximum current limit of 10 amps, with a safety factor of 1.2 to set the overcurrent protection trigger point, allowing for short-term minor overshoot while ensuring rapid response under actual overload. The 10 amps overcurrent protection threshold for the second power output channel is set based on the total rated drive current of the LED lighting module being 8 amps, with a safety factor of 1.25 to prevent the LED array from burning out due to overcurrent. The 8 amps overcurrent protection threshold for the third power output channel is set based on the fact that the peak current of the UAV flight control unit under the worst operating conditions is approximately 6 amps, with a safety factor of 1.33 to set the overcurrent protection threshold, ensuring sufficient safety margin for flight control power supply. When the real-time output current of any channel exceeds the corresponding overcurrent protection threshold for a duration of 2 milliseconds, the overcurrent protection flag for that channel is set. The microcontroller of the multi-channel power distribution control unit outputs a shutdown signal, forcibly disconnecting the power switching device of the corresponding channel, thus achieving hardware-level rapid protection for the channel and load. After shutdown, the microcontroller waits for the fault to clear and then performs an automatic retry after a 100-millisecond delay. If the protection is triggered again after three consecutive retries, the channel is locked and the fault information is reported to the ground station.

[0078] This invention collects the state-of-charge (POC) parameters of an airborne energy storage device and the target brightness command parameters of an LED lighting load, and compares them with preset charging thresholds and preset lighting priority thresholds. Based on the comparison results, it generates mode arbitration signals for charging priority mode, lighting priority mode, or balanced power supply mode, realizing priority-aware dynamic power scheduling for three types of loads: flight, charging, and lighting. In balanced power supply mode, available power margin is allocated according to a weighted ratio determined by the POC parameters and the target brightness command parameters. Upper and lower limit boundary constraints for the allocation power of the first and second power output channels are constructed by the maximum allowable charging power, the minimum effective charging power, the target driving power after junction temperature compensation, and the minimum effective lighting power, respectively. When the remaining power is within the sum of the upper and lower limit boundaries, the initial weighting ratio is calculated by the ratio of the inverse proportional function value of the POC to the direct proportional function value of the target brightness. When the remaining power is below the sum of the lower limit boundaries, the weighting is corrected by a priority bias coefficient. The system proportionally and secures the lower limit boundary of the higher priority channel. When the remaining power exceeds the sum of the upper and lower limits, the excess power is marked as a power margin pool to be allocated and then gradually transferred to the first power output channel with an intermittent pulse duty cycle negatively correlated with the state of charge. This achieves fine-grained power allocation for three loads on a single DC bus, ensuring stable power supply voltage for the flight controller, balancing charging speed and lighting output quality. It solves the problems of fixed weighted ratios failing to effectively utilize surplus energy when there is excess power and failing to guarantee the minimum requirements of critical loads when there is insufficient power. During mode switching, the current limiting threshold of each channel is gradually adjusted within a preset transition time using a preset gradient function, and the output voltage of the third power output channel is monitored in real time to trigger the flight controller priority protection interruption, reducing current surges and voltage fluctuations during mode switching. At the same time, the independent overcurrent protection mechanism of each channel performs rapid shutdown and automatic retry under abnormal conditions, ensuring the power supply safety and operational reliability of the system.

[0079] Example 2

[0080] Please see Figure 2 It should be further explained that, in this embodiment, determining the weighting ratio based on the stated state of charge parameters and the target brightness command parameters includes:

[0081] S601. Extract the state of charge parameters of the airborne energy storage device and the target brightness command parameters of the LED lighting load, and obtain the remaining power value obtained by deducting the allocated power of the third power output channel from the high voltage DC bus power supply.

[0082] S602. Based on the state of charge parameters, determine the maximum allowable charging power and the minimum effective charging power corresponding to the current state of charge.

[0083] S603. Based on the target brightness command parameters and the forward voltage drift compensation value under the current junction temperature conditions, determine the target driving power required for the LED lighting load to reach the target illuminance, and determine the minimum effective lighting power based on the human eye flicker perception threshold.

[0084] S604. The maximum allowable charging power and the rated safe power of the LED lighting load are respectively determined as the upper limit boundary of the power allocation of the first power output channel and the upper limit boundary of the power allocation of the second power output channel. The minimum effective charging power and the minimum effective lighting power are respectively determined as the lower limit boundary of the power allocation of the first power output channel and the lower limit boundary of the power allocation of the second power output channel.

[0085] S605. Compare the remaining power value with the sum of the lower limit boundary of the first power output channel and the lower limit boundary of the second power output channel, and the sum of the upper limit boundary of the first power output channel and the upper limit boundary of the second power output channel.

[0086] S606. When the remaining power value is within the range of the sum of the lower limit boundary to the sum of the upper limit boundary, the initial weighting ratio is calculated by the ratio of the inverse proportional function value of the state of charge parameter to the direct proportional function value of the target brightness command parameter.

[0087] S607. When the remaining power value is lower than the sum of the lower limit boundaries, the initial weighting ratio is corrected according to the preset priority bias coefficient so that the allocation result prioritizes the allocation power lower limit boundary corresponding to the channel with higher priority.

[0088] S608. When the remaining power value is higher than the sum of the upper limit boundaries, the allocated power of the first power output channel is limited to the upper limit boundary of the allocated power of the first power output channel, the allocated power of the second power output channel is limited to the upper limit boundary of the allocated power of the second power output channel, and the power exceeding the sum of the upper limit boundaries is marked as a power margin to be allocated pool.

[0089] It should be further explained that the specific implementation of S601 in this embodiment is as follows: The state of charge parameters of the airborne energy storage device and the target brightness command parameters of the LED lighting load within the current control cycle are read from the input parameter buffer of the power allocation algorithm. The remaining power value obtained by subtracting the allocated power of the third power output channel from the intermediate calculation result register of the power allocation algorithm is then obtained. The input parameter buffer is a data structure allocated in random access memory. This structure contains state of charge parameter members and target brightness command parameter members, stored as 16-bit unsigned integers. The resolution of the state of charge parameters is 0.01%, and the resolution of the target brightness command parameters is also 0.01%. Both are updated and written by the parameter acquisition task before the start of each control cycle, and the corresponding data validity flags are set. The power allocation algorithm checks the data validity flag before reading. If the data has been updated, it reads the data and clears the flag. If the flag is not set, it uses the valid value from the previous control cycle and increments the stale data counter. When the stale data counter exceeds the preset maximum allowable number of stale cycles, it triggers communication timeout protection, forcibly setting the target brightness command parameter to the preset safe default brightness value. The intermediate calculation result register is a global variable area allocated in random access memory, where the remaining power value variable is a 32-bit signed integer in milliwatts. The remaining power value is calculated by the power allocation algorithm within the current control cycle based on the difference between the total input power of the high-voltage DC bus power supply and the real-time allocated power of the third power output channel. Specifically, at the beginning of each control cycle, the analog-to-digital converter synchronously samples the high-voltage side input voltage and input current, and the third power output channel output voltage and output current. After filtering out switching frequency noise by an 8-point moving average filter, the samples are converted into power values. Then, the hardware arithmetic logic unit of the microcontroller performs a 32-bit signed integer subtraction operation, subtracting the allocated power of the third power output channel from the total input power. The resulting difference is written into the remaining power value variable. If the calculation result is negative, the remaining power value is set to zero to avoid logical errors of negative power allocation in subsequent allocation calculations. This difference is updated once at the beginning of each control cycle.

[0090] It should be further explained that the specific implementation of S602 in this embodiment is as follows: using the state of charge parameter as an index, the maximum allowable charging power upper limit curve table and the minimum effective charging power threshold value table pre-stored in non-volatile memory are queried, and the maximum allowable charging power and minimum effective charging power corresponding to the current state of charge are obtained by linear interpolation or table lookup. The maximum allowable charging power upper limit curve table is a one-dimensional array that stores the maximum acceptable charging power values ​​corresponding to 21 discrete points from 0% to 100% of the state of charge with a step size of 5%, and the data type is 16-bit unsigned integer, with the unit being watts. The table was obtained as follows: using a lithium polymer battery pack of the same model as the airborne energy storage device, with a nominal voltage of 22.2 volts and a capacity of 10,000 mAh, the charging acceptance capability of the battery was tested at 21 discrete points from 0%, 5%, 10%, to 100% of the state of charge (SOC). The continuous input power at each SOC point before the battery terminal voltage reached the charging cutoff voltage was recorded as the maximum acceptable charging power at that point, forming 21 sets of discrete data pairs. Piecewise linear interpolation was used between adjacent discrete points, and the data was discretized into 101 power value entries with a SOC step size of 1% and stored in non-volatile memory. The curve shows a higher acceptable charging power in the low SOC range, gradually decreasing as the SOC increases, and approaching the trickle charging power value near full charge. The minimum effective charging power threshold table stores the minimum input power value required to ensure the continuous positive progress of the internal electrochemical reaction of the battery under different states of charge without lithium plating or charge-discharge oscillation. This threshold value is determined as follows: based on the electrochemical characteristic parameters provided by the battery manufacturer, lithium polymer batteries need to maintain a minimum input power threshold during charging to ensure the continuous positive progress of the lithium-ion intercalation reaction between the negative electrode graphite layers. In this embodiment, the lowest inflection point power value of 20 watts, obtained from the charging efficiency versus input power relationship curve obtained by cyclic charge-discharge efficiency testing of this battery model at room temperature, is used as the unified threshold value for each state of charge and stored in non-volatile memory.

[0091] It should be further explained that the specific implementation of S603 in this embodiment is as follows: the target brightness command parameter is input into an illuminance-power mapping function. This mapping function converts the target brightness command parameter into the corresponding target driving power according to the photoelectric conversion characteristic curve of the LED lighting load. In this embodiment, the LED lighting module consists of 4 COB packaged LED arrays with a total rated lighting power of 200 watts. The photoelectric conversion characteristic has good linearity within the rated current range. The illuminance-power mapping function is a linear mapping relationship, and the mapping formula is that the target driving power equals the target brightness command parameter divided by 100 multiplied by the nominal rated power of the LED lighting load. At the same time, the current junction temperature value of the LED array collected by the airborne temperature sensor is obtained. This temperature sensor is a digital temperature sensor mounted on the LED array substrate, which communicates with the microcontroller through a single bus interface and has a sampling frequency of 10 Hz. The forward voltage drift compensation value is calculated based on the preset forward voltage temperature drift coefficient, relative to the nominal junction temperature of 25 degrees Celsius, under the current junction temperature. This forward voltage temperature drift coefficient is determined according to parameters provided in the LED chip datasheet: the forward voltage of a single LED at rated current changes with increasing junction temperature at a rate of -3 mV / degree Celsius. The total drive power drift coefficient, after conversion through LED array series-parallel topology, is equivalent to -0.15 watts / degree Celsius. This equivalent drift coefficient is calibrated by those skilled in the art during the system design phase through circuit simulation and actual measurement, and then written into non-volatile memory. The formula for calculating the forward voltage drift compensation value is: the compensation value equals the current junction temperature minus the nominal junction temperature multiplied by the total drift coefficient. The forward voltage drift compensation value is then superimposed on the target drive power to obtain the target drive power after junction temperature compensation. The human eye flicker perception threshold refers to the minimum drive power fluctuation amplitude that the human eye can perceive under the illuminance level corresponding to the target brightness command parameters, determined based on the persistence of vision and critical flicker frequency tests. This threshold is obtained from an illuminance-flicker perception threshold lookup table stored after discretizing the critical flicker frequency versus modulation depth relationship curve recommended in the International Commission on Illumination (CIETN) technical report CIETN006:2016. The minimum permissible power fluctuation percentage at the corresponding illuminance level is obtained by looking up the table using the illuminance percentage as an index. This percentage is then multiplied by the target drive power after junction temperature compensation to obtain the minimum effective lighting power, which is output to S604. The purpose of setting this threshold is to ensure that the power allocated to the lighting channels does not drop below the perceptible flicker level that causes visual discomfort or affects operational safety when the system's available power margin is limited.

[0092] It should be further explained that the specific implementation of S604 in this embodiment is as follows: The maximum allowable charging power obtained in S602 is written into the power upper limit boundary register of the first power output channel. This register is a 32-bit signed integer variable, with the unit being milliwatts. The smaller value between the target driving power after junction temperature compensation obtained in S603 and the nominal rated power of the LED lighting load is determined as the rated safe power of the LED lighting load and written into the power upper limit boundary register of the second power output channel. This smaller value logic ensures that the driving power meets the illuminance requirements without exceeding the physical safety limit of the module. At the same time, the minimum effective charging power obtained in S602 is written into the power lower limit boundary register of the first power output channel, and the minimum effective lighting power obtained in S603 is written into the power lower limit boundary register of the second power output channel. The rated safe power of the LED lighting load is pre-stored in the load parameter configuration table of the non-volatile memory. This value is the maximum allowable continuous driving power of the LED lighting module, determined by the thermal design power of the LED array and the steady-state thermal resistance of the heat dissipation system. During the system integration phase, it is written after verification through thermal simulation and actual measurement based on the maximum allowable continuous driving power, module thermal design power, and steady-state thermal resistance of the matching heat dissipation system in the LED module specifications. Exceeding this value will trigger over-temperature protection or overload protection of the lighting branch. The lower limit boundary of the allocated power of the first power output channel and the lower limit boundary of the allocated power of the second power output channel are dynamically updated in each control cycle according to the state of charge parameters and target brightness command parameters. The upper limit boundary of the allocated power of the first power output channel and the upper limit boundary of the allocated power of the second power output channel are also adjusted downward by a preset derating factor when the current junction temperature exceeds the preset junction temperature derating threshold. The preset junction temperature derating threshold is set to 85 degrees Celsius based on the upper limit of the LED allowable junction temperature of 120 degrees Celsius and after retaining an appropriate safety margin. The preset derating factor is that the upper limit of power is reduced by 2% for every 1 degree Celsius exceeding the junction temperature derating threshold.

[0093] It should be further explained that the specific implementation of S605 in this embodiment is as follows: the arithmetic logic unit of the power allocation algorithm performs a first summation operation and a second summation operation respectively. The first summation operation adds the lower limit boundary of the allocated power of the first power output channel and the lower limit boundary of the allocated power of the second power output channel to obtain the sum of the lower limit boundaries. The second summation operation adds the upper limit boundary of the allocated power of the first power output channel and the upper limit boundary of the allocated power of the second power output channel to obtain the sum of the upper limit boundaries. The summation operation is completed in a single instruction cycle using 32-bit signed integer arithmetic. Subsequently, the remaining power value obtained in S601 is compared with the sum of the lower limit boundaries and the sum of the upper limit boundaries, and a comparison result flag is output. This comparison result flag is a 2-bit wide flag in the program status word register, used to indicate that the remaining power value is in one of three interval states: below the sum of the lower limit boundaries, between the sum of the lower limit boundaries and the sum of the upper limit boundaries, or above the sum of the upper limit boundaries. The basis for setting these three interval states is that the relationship between the remaining power and the sum of the upper and lower limits directly determines whether the system currently has sufficient power margin to perform proportional distribution, or must enter the power shortage protection or power overflow limit mode.

[0094] It should be further explained that the specific implementation of S606 in this embodiment is as follows: When the comparison result flag bit output by S605 indicates that the remaining power value is within the range of the sum of the lower limit boundary to the sum of the upper limit boundary, the power allocation algorithm calls the inverse state of charge (SOC) proportional function calculation unit and the target brightness direct proportional function calculation unit. The SOC proportional function calculation unit is input with the SOC parameter as the independent variable to obtain the inverse proportional function value; the target brightness command parameter is input with the target brightness direct proportional function calculation unit to obtain the direct proportional function value. The ratio of the inverse proportional function value to the direct proportional function value is used as the initial weighting ratio. The specific form of the SOC proportional function is f(SOC) = 1 / (SOC + ε), where SOC is the value of the SOC parameter normalized to the range of 0 to 1, and ε is a preset positive small constant with a value of 0.01 to prevent the denominator from being zero. This function outputs a larger value when the SOC is low to increase the charging weight, and outputs a smaller value when the SOC is high to decrease the charging weight. The specific form of the target brightness proportional function is g(L)=L, where L is the value of the target brightness command parameter normalized to the range of 0 to 1. This function outputs a larger value to increase the lighting weight when the brightness command is high, and a smaller value to decrease the lighting weight when the brightness command is low. The initial weighting ratio is limited by a limiter to a preset minimum weighting ratio of 0.1 and a maximum weighting ratio of 10.0, and then output to the power allocation calculation units of the first and second power output channels. The limiter prevents the weighting ratio from overflowing or becoming too small due to extreme input values, ensuring the rationality of the allocation.

[0095] It should be further explained that the specific implementation of S607 in this embodiment is as follows: When the comparison result flag bit output by S605 indicates that the remaining power value is lower than the sum of the lower limit boundaries, the power allocation algorithm first reads a preset priority flag bit from the non-volatile memory. This priority flag bit is written according to the task requirements during the system configuration phase and is used to clearly indicate which channel is set as the higher priority channel between the first power output channel and the second power output channel. At the same time, a preset priority bias coefficient is read. This priority bias coefficient is a dimensionless coefficient with a value range between 0.5 and 1.0, and the default value is 0.8. Its specific value is determined by those skilled in the art based on the charging urgency of the airborne energy storage device and the task priority of the LED lighting load. This coefficient is used to characterize the degree of tilt towards the higher priority channel when the available power margin is insufficient to simultaneously meet the minimum effective power requirements of both channels. That is, when the power is insufficient, priority is given to ensuring that the higher priority channel obtains at least 80% of its allocated power lower limit boundary. Subsequently, the power allocation algorithm obtains the initial weighting ratio calculated in S606, multiplies this initial weighting ratio by the priority bias coefficient to obtain the corrected weighting ratio, then renormalizes the corrected weighting ratio, and uses the sum of the corrected weighting ratio and the number 1 as the denominator to calculate the initial power allocation ratio of the first power output channel. This ratio is then subtracted from 1 to obtain the initial power allocation ratio of the second power output channel. Based on these two initial power allocation ratios, the remaining power value obtained in S601 is initially divided into the initial power allocation ratio of the first channel and the initial power allocation ratio of the second channel.

[0096] After the initial allocation is completed, the power allocation algorithm performs a lower limit boundary protection correction on the allocation result based on the higher priority channel indicated by the priority flag. The specific logic branches are as follows:

[0097] When the higher-priority channel is the first power output channel, the initial allocated power of the first channel is compared with the lower limit boundary of the allocated power of the first power output channel. If the initial allocated power of the first channel is greater than or equal to the lower limit boundary of the allocated power of the first power output channel, then the initial allocated power of the first channel is used as the final allocated power of the first power output channel, and the initial allocated power of the second channel is used as the final allocated power of the second power output channel, without correction. If the initial allocated power of the first channel is less than the lower limit boundary of the allocated power of the first power output channel, then the final allocated power of the first power output channel is forcibly set to the lower limit boundary value; at this time, the remaining power value minus the lower limit boundary value is calculated. When the remaining power value is greater than 0, the entire remaining power value is allocated to the second power output channel as its final allocated power; when the remaining power value is equal to 0 or less than 0, the final allocated power of the first power output channel is the entire remaining power value, and the final allocated power of the second power output channel is set to 0.

[0098] When the higher-priority channel is the second power output channel, a completely symmetrical correction logic is used: the initial allocated power of the second channel is compared with the lower limit boundary of the allocated power of the second power output channel. If the initial allocated power of the second channel is greater than or equal to the lower limit boundary of the allocated power of the second power output channel, then the initial allocated power of the second channel is used as the final allocated power of the second power output channel, and the initial allocated power of the first channel is used as the final allocated power of the first power output channel, without correction. If the initial allocated power of the second channel is less than the lower limit boundary of the allocated power of the second power output channel, then the final allocated power of the second power output channel is forcibly set to the lower limit boundary value; at this time, the remaining power value minus the lower limit boundary value is calculated. When the remaining power value is greater than 0, the entire remaining power value is allocated to the first power output channel as its final allocated power; when the remaining power value is equal to 0 or less than 0, the final allocated power of the second power output channel is the entire remaining power value, and the final allocated power of the first power output channel is set to 0.

[0099] It should be further explained that the specific implementation of S608 in this embodiment is as follows: When the comparison result flag bit output by S605 indicates that the remaining power value is higher than the sum of the upper limit boundaries, the power allocation algorithm writes the allocation power register of the first power output channel to the upper limit boundary value of the allocation power of the first power output channel, and writes the allocation power register of the second power output channel to the upper limit boundary value of the allocation power of the second power output channel, with both channels operating at full load. The difference obtained by subtracting the sum of the upper limit boundaries from the remaining power value is taken as the excess power, and this excess power is marked as a power margin to be allocated pool, which is stored in a 32-bit signed integer accumulation register of the power allocation algorithm. This accumulation register is cleared to zero during system power-on initialization. If the accumulated value of the power margin to be allocated pool is detected to fall below the sum of the upper limit boundaries in subsequent control cycles, it is preferentially released to supplement the power gap of the first power output channel or the second power output channel, so as to avoid wasting power margin. If the accumulated value of the power margin pool continues to exceed the preset upper limit threshold of 50 watts for a preset duration of 10 seconds, the power allocation algorithm sends an auxiliary charging enable signal to the airborne energy storage device, temporarily transferring part or all of the power from the power margin pool to the first power output channel to perform auxiliary charging, until the accumulated value of the power margin pool drops below the preset lower limit threshold of 5 watts. The upper limit threshold of 50 watts is set based on the following: the battery's 1 coulomb pulse charging power is approximately 222 watts, minus the basic charging power of 180 watts, leaving approximately 42 watts, which is rounded up and set to 50 watts with an appropriate margin; the preset duration of 10 seconds is an anti-jitter delay set to filter out instantaneous power fluctuations and prevent frequent jumps in the auxiliary charging enable signal.

[0100] It should be further noted that in this embodiment, the non-volatile memory is the on-chip flash memory of the embedded controller, with a storage capacity of 512 kilobytes. The register is the general-purpose data register of the embedded controller, with a data bit width of 32 bits. The arithmetic logic unit is the hardware multiplication and division unit of the ARM Cortex-M4 CPU core of the embedded controller, supporting single-cycle 32-bit multiplication and division operations. The comparison operation, summation operation, and amplitude limiting operation are all executed by the firmware of the embedded controller in integer mode at a control cycle of 100 Hz. The time step of each control cycle is 10 milliseconds, triggered by periodic interrupts generated by the system tick timer inside the microcontroller. The operation precision is 32-bit signed integer, and the resolution of the state of charge parameter and brightness command parameter is 0.01%. The acceptable charging power upper limit curve, the minimum effective charging power threshold value table, the forward voltage temperature drift coefficient, the human eye flicker perception threshold curve, the rated safe power of the LED lighting load, and the preset priority bias coefficient are all loaded from non-volatile memory to random access memory during the system initialization phase for each control cycle. The loading process is completed by the system boot program within 100 milliseconds after the microcontroller is powered on and reset.

[0101] In a specific application example, the complete execution process of the weighting ratio determination method in this embodiment is as follows:

[0102] Assuming the tethered multi-functional emergency lighting UAV system of this embodiment is in balanced power supply mode, the input parameters of the current control cycle are as follows: the state of charge (SOC) parameter of the airborne energy storage device is 35%, expressed as 3500 in unsigned integer with a resolution of 0.01%; the target brightness command parameter L of the LED lighting load is 60%, expressed as 6000; the total input power of the high-voltage DC bus power supply is 450 watts; and the real-time power distribution of the third power output channel is 150 watts.

[0103] The power allocation algorithm executes step S601 at the start of the current control cycle. It reads the state-of-charge parameter 3500 and the target brightness command parameter 6000 from the input parameter buffer, obtains the remaining power value from the intermediate calculation result register (which is calculated by subtracting the 150 watts allocated to the third power output channel from the total input power of 450 watts, resulting in 300 watts, or 300,000 milliwatts), and updates the remaining power value variable written to the intermediate calculation result register. All these operations are performed immediately after the analog-to-digital converter sampling is completed at the start of the 100 Hz control cycle, with the update frequency synchronized with the control cycle.

[0104] In S602, the power allocation algorithm uses a SOC of 35% as an index and accelerates the location using a binary search algorithm to query the upper limit curve table of acceptable charging power. In this table, the maximum allowable charging power P_chg_max(30) = 188 watts corresponds to a SOC of 30%, and P_chg_max(40) = 174 watts corresponds to a SOC of 40%. The maximum allowable charging power under the current state of charge is calculated using linear interpolation, with the following formula:

[0105] P_chg_max(35)=P_chg_max(30)+((SOC-30) / (40-30))×(P_chg_max(40)-P_chg_max(30))

[0106] Substituting the values, we get P_chg_max(35) = 188 + ((35-30) / (40-30)) × (174-188) = 180 watts.

[0107] Simultaneously, the minimum effective charging power threshold table is consulted, and the minimum effective charging power P_chg_min(35) = 20 watts is obtained corresponding to a SOC of 35%. In this embodiment, the upper limit curve of the acceptable charging power corresponds to a maximum allowable charging power of 200 watts when the SOC is 0%, 160 watts when the SOC is 50%, and 10 watts of trickle charging power when the SOC is 100%. The minimum effective charging power threshold table is constant at 20 watts throughout the entire range of SOC from 0% to 100%, and is determined by the minimum driving power parameter of the electrochemical reaction provided by the battery manufacturer.

[0108] In S603, the target brightness command parameter L is set to a 60% input illuminance-power mapping function. This function is a linear mapping relationship, and the mapping formula is as follows:

[0109] P_led_target=(L / 100)×P_led_rated

[0110] Where P_led_target is the target driving power, L is the target brightness command parameter, and P_led_rated is the nominal rated power of the LED lighting load. Substituting the values, we get P_led_target = (60 / 100) × 200 = 120 watts.

[0111] Meanwhile, the airborne temperature sensor collects the current LED array junction temperature T_j, which is 65 degrees Celsius. The preset nominal junction temperature T_ref is 25 degrees Celsius, and the forward voltage temperature drift coefficient K_drift is -0.15 watts per degree Celsius. This coefficient is determined according to the LED chip datasheet: the forward voltage of a single LED at rated current changes with the increase of junction temperature at a rate of -3 millivolts per degree Celsius. After conversion through the LED array series-parallel topology, the total drive power drift coefficient is equivalent to K_drift = -0.15 watts per degree Celsius. The forward voltage drift compensation value ΔP_comp is calculated according to the following formula:

[0112] ΔP_comp=(T_j-T_ref)×K_drift=(65-25)×(-0.15)=-6 watts.

[0113] The target drive power P_led_comp after junction temperature compensation is calculated according to the following formula:

[0114] P_led_comp=P_led_target+ΔP_comp=120+(-6)=114 watts.

[0115] The flicker perception threshold for the human eye is based on the illuminance-flicker perception threshold curve recommended by the International Commission on Illumination. At a 60% illuminance level, the minimum perceptible fluctuation in driving power is 5% of the steady-state driving power at that illuminance. That is, the minimum effective lighting power P_led_min is calculated using the following formula:

[0116] P_led_min=P_led_comp×R_flicker=114×0.05=5.7 watts.

[0117] R_flicker is the flicker perception ratio coefficient, which is determined by the lookup value of the illuminance-flicker perception threshold curve at a 60% illuminance level, and has a value of 0.05.

[0118] In S604, the maximum allowable charging power of 180 watts obtained in S602 is written to the power upper limit boundary register P1_upper=180 watts in milliwatts. The target driving power of 114 watts after junction temperature compensation obtained in S603 is compared with the nominal rated power of 200 watts of the LED lighting load stored in non-volatile memory, and the smaller value of 114 watts is taken as the rated safe power of the LED lighting load and written to the power upper limit boundary register P2_upper=114 watts in the second power output channel. The minimum effective charging power of 20 watts obtained in S602 is written to the power lower limit boundary register P1_lower=20 watts in the first power output channel; the minimum effective lighting power of 5.7 watts obtained in S603 is written to the power lower limit boundary register P2_lower=5.7 watts in the second power output channel. During this control cycle, the current junction temperature of 65 degrees Celsius does not exceed the preset junction temperature derating threshold of 85 degrees Celsius, the derating factor remains at 1.0, and no derating adjustment is made to the upper limit boundaries of the power output channel.

[0119] In S605, the arithmetic logic unit of the power allocation algorithm performs the first and second summation operations: the lower boundary sum P_sum_lower = P1_lower + P2_lower = 20 + 5.7 = 25.7 watts, and the upper boundary sum P_sum_upper = P1_upper + P2_upper = 180 + 114 = 294 watts. The remaining power value P_remain = 300 watts obtained in S601 is compared with the lower boundary sum of 25.7 watts and the upper boundary sum of 294 watts. The remaining power value of 300 watts is greater than the upper boundary sum of 294 watts, and the comparison result is marked as being higher than the upper boundary sum.

[0120] In S608, the comparison result flag indicates that the remaining power value is higher than the sum of the upper limit boundaries of 294 watts. The power allocation algorithm writes the upper limit boundary value P1_alloc=P1_upper=180 watts to the power allocation register of the first power output channel and writes the upper limit boundary value P2_alloc=P2_upper=114 watts to the power allocation register of the second power output channel. The difference between the remaining power value of 300 watts and the sum of the upper limit boundaries of 294 watts is the excess power P_excess = P_remain - P_sum_upper = 300 - 294 = 6 watts. This 6 watts is marked as the power margin to be allocated pool and stored in the accumulation register P_pool = P_pool_prev + P_excess, where P_pool represents the accumulated value of the power margin to be allocated pool after the current control cycle; P_pool_prev represents the power value accumulated in the power margin to be allocated pool at the end of the previous control cycle; and P_excess represents the surplus power calculated in the current control cycle that exceeds the sum of the upper limit boundaries of the allocated power of the first power output channel and the second power output channel. This equation shows that in each control cycle, the system adds the newly generated surplus power to the previously accumulated power to achieve continuous accumulation of power margin for gradual recovery and utilization in subsequent auxiliary charging processes.

[0121] For example, the current accumulated value in the accumulation register is updated from 2 watts in the previous control cycle to P_pool = 2 + 6 = 8 watts. This accumulated value does not exceed the preset accumulated upper limit threshold of 50 watts and does not last for more than the preset duration of 10 seconds. Therefore, the auxiliary charging enable signal is not triggered in this control cycle, and the power margin waiting pool continues to accumulate. The allocated power of 180 watts for the first power output channel and 114 watts for the second power output channel are converted into constant current control reference values ​​for the corresponding channels through the serial peripheral interface and written into the digital control registers of each channel of the multi-channel power allocation control unit. Each channel's independent analog control loop executes precise current limiting control in the next control cycle, thereby realizing fine-grained power allocation for charging and lighting loads based on dynamic weighting ratios and actual physical constraints in real time under balanced power supply mode. The entire calculation, comparison, and allocation process is completed within 100 microseconds, meeting the strong real-time requirements of UAV dynamic power management.

[0122] It is understood that the specific values, thresholds and coefficients given in this application example are only for illustrative purposes and are not intended to limit the scope of protection of this invention. In actual implementation, those skilled in the art can determine the corresponding parameter values ​​based on specific system specifications, load characteristics and test data without creative effort.

[0123] This invention addresses the technical problems faced by tethered multi-functional emergency lighting drones in balanced power supply mode: fixed weighted ratios cannot adapt to power fluctuations, the minimum requirements of critical loads are difficult to guarantee when power is insufficient, and surplus energy cannot be effectively utilized when power is excessive. It proposes a weighted ratio calculation scheme based on dynamic upper and lower limit boundary constraints and three-interval comparison. The scheme uses the state of charge (POC) parameter as an index to obtain the maximum allowable charging power that changes with the POC through linear interpolation, ensuring that the upper limit of charging power matches the current acceptable charging capacity of the battery. The lower value between the target driving power after junction temperature compensation and the rated safe power is used as the upper limit of lighting power, providing thermal protection margin for the LED array while meeting illuminance requirements. Finally, the minimum effective charging power and minimum effective lighting power are set based on the minimum input power threshold for continuous positive electrochemical reaction of the battery and the human eye flicker perception threshold, respectively. The power lower limit boundary of the two channels ensures the effective working baseline of each load. By comparing the remaining power value with the sum of the upper and lower limit boundaries in three intervals, when the remaining power is between the upper and lower limit boundaries, the initial weighting ratio is calculated by the ratio of the inverse proportional function value of the state of charge to the direct proportional function value of the target brightness and the output is limited. This makes the charging weight automatically decrease as the state of charge increases and the lighting weight automatically increase as the brightness command increases, realizing continuous adaptive matching between power allocation and real-time demand. When the remaining power is lower than the sum of the lower limit boundaries, the weighting ratio is corrected by the priority bias coefficient and the lower limit boundary protection correction is performed in combination with the priority flag bit, so that the channel with higher priority gets the power allocation not lower than its lower limit boundary. When the remaining power is higher than the sum of the upper limit boundaries, the excess power is marked as a power margin to be allocated pool and stored in the accumulation register, so that it can be gradually transferred to the first power output channel for recycling in the subsequent auxiliary charging pulse mode, reducing the waste of excess power. The entire weighting ratio determination process is executed in a 100 Hz control cycle. All calculations from parameter reading to power allocation result output are completed within one control cycle, which meets the strong real-time requirements of UAV dynamic power management and takes into account both charging speed and lighting effect while ensuring priority for flight control power supply.

[0124] Example 3

[0125] Please see Figure 3It should be further explained that the process of gradually transferring the power accumulated in the power margin allocation pool to the first power output channel in this embodiment is an auxiliary charging execution process specifically designed for how to safely and efficiently utilize the accumulated power after the accumulated value of the power margin allocation pool continuously exceeds the upper limit threshold and reaches a preset time in S608 of Embodiment 2. In S608, when the remaining power value is higher than the sum of the upper limit boundaries, the excess power is continuously marked and stored in the accumulation register of the power margin allocation pool. However, if the accumulated power is only passively stored and not actively utilized, the surplus power transmitted from the ground will be actually wasted. If it is directly released to the first power output channel in a continuous high current manner, it may trigger the cell overvoltage protection or accelerate the cell capacity decay when the state of charge of the airborne energy storage device is already high. To solve the above technical contradictions, this embodiment adopts an intermittent auxiliary charging method based on adaptive adjustment of pulse duty cycle according to state of charge, and gradually transfers the accumulated power margin to the first power output channel at a controlled rate. The process specifically includes the following:

[0126] S609. When the cumulative value of the power margin to be allocated pool continuously exceeds the preset cumulative upper limit threshold and the duration reaches the preset duration, an auxiliary charging enable signal is generated.

[0127] It should be further explained that the specific implementation of S609 in this embodiment is as follows: The microcontroller reads the current value of the 32-bit signed integer accumulator register corresponding to the power margin allocation pool in each 100Hz control cycle and compares it with a preset accumulator upper limit threshold of 50 watts. When the accumulated value is greater than 50 watts, an incrementing timer is started, incrementing by 1 in each control cycle; if the accumulated value falls below 50 watts during the timing process, the timer is immediately reset to zero. When the timer's count reaches the count value of 1000 corresponding to a preset duration of 10 seconds, the microcontroller sets the auxiliary charging enable signal flag and generates an auxiliary charging enable signal. The cumulative upper limit threshold of 50 watts is set based on the following: the maximum pulse charging power that the airborne energy storage device can withstand in a single trickle charging phase is about 222 watts. After subtracting the basic charging power of 180 watts from the first power output channel, about 42 watts remain. After rounding and retaining an appropriate safety margin, it is set to 50 watts. The preset duration of 10 seconds is set based on the following: to filter out instantaneous power fluctuations caused by sudden changes in flight attitude or gust disturbances, and to prevent the auxiliary charging enable signal from being frequently triggered and deactivated when the power briefly overflows.

[0128] S610. Obtain the state of charge parameters of the airborne energy storage device in the current control cycle, and determine the auxiliary charging pulse duty cycle of the current control cycle based on the difference between the state of charge parameters and the fully charged state of charge.

[0129] It should be further explained that the specific implementation of S610 in this embodiment is as follows: The microcontroller reads the latest state of charge (SOC) parameter from the input parameter buffer of the power allocation algorithm for the current control cycle. This SOC parameter is expressed as a percentage value with a resolution of 0.01%. The full charge state of charge (SOC_full) is 100, indicating that the airborne energy storage device has reached the charging cutoff voltage. The auxiliary charging pulse duty cycle (D_aux) is calculated according to the following formula:

[0130] D_aux = D_max × (1 - SOC / SOC_full)

[0131] Wherein, D_max is the preset maximum auxiliary charging pulse duty cycle, with a value of 0.3. The corresponding physical meaning is that the high-level duration of the auxiliary charging pulse accounts for an upper limit of 30% of the entire pulse cycle. The setting of D_max is based on the maximum allowable pulse charging current duty cycle of the airborne energy storage device during the trickle charging phase, determined by the pulse charging tolerance parameters provided by the battery manufacturer: during the trickle charging phase, the maximum acceptable pulse charging current of the battery is 0.3 times the standard charging current, corresponding to a duty cycle of 0.3. SOC is the state of charge parameter for the current control cycle, with SOC_full representing a fully charged state of charge (100%). When SOC is 20%, D_aux = 0.3 × (1 - 20 / 100) = 0.24; when SOC is 50%, D_aux = 0.3 × (1 - 50 / 100) = 0.15; when SOC is 95%, D_aux = 0.3 × (1 - 95 / 100) = 0.015. As shown in the above calculation formula, when the state of charge (SOC) parameter is closer to a full charge, the SOC / SOC_full ratio is closer to 1, the remainder of 1 minus this ratio is closer to 0, the auxiliary charging pulse duty cycle D_aux is smaller, and the power margin release rate is slower. Conversely, when the SOC parameter is lower, the auxiliary charging pulse duty cycle is larger, and the power margin release rate is faster. This negative correlation ensures that the intensity of auxiliary charging matches the battery's current capacity for additional charging, accelerating the recovery of excess power when the battery is low and gradually reducing the recovery intensity as the battery approaches full charge to protect battery safety.

[0132] S611. Based on the duty cycle of the auxiliary charging pulse and the preset auxiliary charging pulse frequency, generate a set of intermittent auxiliary charging pulse sequences, and apply the intermittent auxiliary charging pulse sequences to the current limiting control loop of the first power output channel to superimpose an intermittent power margin consumption current component on the basis of the preset current limiting value of the first power output channel.

[0133] It should be further explained that the specific implementation of S611 in this embodiment is as follows: the preset auxiliary charging pulse frequency is set to 1 Hz, that is, each pulse cycle is 1 second. The basis for setting this pulse frequency is that the pulse cycle of intermittent auxiliary charging needs to be much larger than the 100 Hz control cycle of the power allocation algorithm, so that each pulse cycle contains 100 complete control cycles, which is sufficient to ensure the stable adjustment of the current limiting control loop. At the same time, the 1-second pulse cycle is not too long, so that the accumulated power of the power margin waiting to be allocated cannot be consumed in time. The timer module of the microcontroller generates the corresponding pulse width modulation control signal according to D_aux and the 1 Hz pulse frequency: in each 1-second pulse cycle, the high level duration is equal to D_aux multiplied by 1 second, and the low level duration is equal to 1 second minus the high level duration. This control signal is output to the reference value superposition circuit of the current limiting control loop of the first power output channel through the general input / output interface of the microcontroller. During the high level period, the current limiting threshold of the first power output channel is at the original preset current limiting value I. max1 A power margin current consumption component I_aux is superimposed on top of the basic current. This power margin current consumption component is calculated by dividing the current accumulated value of the power margin to be allocated pool by the real-time terminal voltage of the airborne energy storage device, and then multiplying it by the auxiliary charging pulse duty cycle D_aux. During the low-level period, the superimposed power margin current consumption component is removed, and the current limiting threshold of the first power output channel is restored to the original preset current limiting value I_aux. max1 Through this intermittent superposition method, the first power output channel gradually consumes the accumulated power in the power margin cell during the high-level period with a controlled additional charging current, while providing sufficient buffer time for the internal electrochemical reaction of the battery during the low-level period, thus avoiding stress damage to the cell caused by continuous high current.

[0134] S612. During the execution of the intermittent auxiliary charging pulse sequence, the cumulative value of the power margin allocation pool is continuously monitored. When the cumulative value of the power margin allocation pool drops below a preset cumulative lower limit threshold, the auxiliary charging enable signal is canceled, the output of the intermittent auxiliary charging pulse sequence is stopped, and the current limiting control of the first power output channel is restored to the normal state where the power margin consumption current component is not superimposed.

[0135] It should be further explained that the specific implementation of S612 in this embodiment is as follows: the microcontroller reads the accumulated register value of the power margin allocation pool in real time during each 100Hz control cycle and compares it with the preset accumulated lower limit threshold of 5 watts. The basis for setting the accumulated lower limit threshold of 5 watts is: when the accumulated power is lower than 5 watts, the actual energy that can be charged into the airborne energy storage device after deducting the power transmission loss of the charging path is very limited, and the power consumption of the device that continues to maintain the auxiliary charging pulse sequence is not worthwhile compared with the energy recovery; at the same time, retaining a residual accumulated value of 5 watts helps to quickly re-accumulate in the next control cycle of residual power overflow, avoiding frequent switching of the auxiliary charging enable signal due to the fluctuation of the accumulated value near zero. When the accumulated value of the power margin allocation pool is detected to drop below 5 watts, the microcontroller clears the auxiliary charging enable signal flag bit to zero, the timer module stops outputting the pulse width modulation control signal, the reference value superposition terminal of the current limiting control loop of the first power output channel is restored to the initial state when the power margin consumption current component is not superimposed, and the current limiting threshold of the first power output channel is completely restored to the original preset current limiting value I. max1 Subsequently, if the accumulated value of the power margin allocation pool in a subsequent control cycle again meets the triggering condition described in S609, the auxiliary charging process from S609 to S612 will be executed again.

[0136] In a specific application example, assuming the accumulated value of the power margin to be allocated pool has reached 65 watts within the current control cycle and has lasted for more than the preset duration of 10 seconds, satisfying the trigger condition of S609, the microcontroller sets the auxiliary charging enable signal. Then, S610 is executed, reading the current state of charge parameter SOC as 50, substituting it into the formula D_aux = 0.3 × (1 - 50 / 100) = 0.15, obtaining the auxiliary charging pulse duty cycle of 0.15 for the current control cycle. In S611, an intermittent auxiliary charging pulse sequence is generated based on the duty cycle of 0.15 and the preset frequency of 1 Hz, with a high level lasting 0.15 seconds and a low level lasting 0.85 seconds within each 1-second pulse cycle. During the 0.15-second high-level period, the first power output channel adds a power margin current consumption component to the original preset current limit value. Assuming the battery terminal voltage is 25.2 volts, the power margin current consumption component is approximately 0.39 amps (65 watts divided by 25.2 volts multiplied by 0.15). At this time, the current limit threshold of the first power output channel is the sum of the original preset current limit value and 0.39 amps. During the 0.85-second low-level period, the current limit threshold returns to the original preset current limit value. After several pulse cycles, the accumulated value of the power margin allocation pool gradually decreases from 65 watts to below 5 watts. After S612 detects this condition, it cancels the auxiliary charging enable signal, stops the intermittent auxiliary charging pulse sequence output, and the current limiting control of the first power output channel returns to normal.

[0137] It is understood that the cumulative upper limit threshold of 50 watts, the preset duration of 10 seconds, the maximum auxiliary charging pulse duty cycle of 0.3, the auxiliary charging pulse frequency of 1 Hz, the cumulative lower limit threshold of 5 watts, and the specific values ​​in each calculation example listed in this embodiment are all exemplary illustrations and are not intended to limit the scope of protection of this invention. In actual implementation, those skilled in the art can determine the corresponding parameter values ​​based on the specific system specifications, battery characteristics, and test data without creative effort.

[0138] This embodiment designs an intermittent auxiliary charging execution process based on adaptive adjustment of pulse duty cycle according to state of charge. It addresses the issues of power margin accumulation and passive waste caused by the continuous excess power exceeding the load absorption capacity in balanced power supply mode, as well as the risk of battery overvoltage damage if the accumulated power is released directly with a large current. A safe and efficient surplus energy delayed recovery mechanism is established: By setting an accumulated upper limit threshold and duration as dual triggering conditions, an incremental timer and fallback reset logic filter out frequent start-stop of the auxiliary charging enable signal caused by instantaneous power fluctuations, ensuring that energy recovery is only initiated when there is a continuous power surplus; after triggering, the auxiliary charging pulse duty cycle is determined in real time by substituting the state of charge parameters into the duty cycle calculation formula, making the duty cycle negatively correlated with the state of charge, and adjusting the duty cycle based on the battery capacity. The duty cycle is increased at lower levels to accelerate the recovery of excess power, and decreased when the battery is close to full charge to gradually reduce the recovery intensity, so that the auxiliary charging intensity matches the battery's current capacity for additional charging. The power margin is superimposed on the original preset current limit value of the first power output channel in the form of intermittent pulse current at a preset auxiliary charging pulse frequency. During the high level, the accumulated power is gradually consumed by the controlled additional charging current, and during the low level, a buffer time is provided for the internal electrochemical reaction of the battery to avoid stress damage to the cell caused by continuous high current. The accumulated value is continuously monitored during the execution process. When the accumulated value drops below the lower limit threshold, the auxiliary charging enable signal is canceled and the normal current limit control is restored, avoiding frequent switching caused by fluctuations near zero. This achieves a complete closed loop from passive accumulation of excess power to active and safe recovery.

[0139] Example 4

[0140] This embodiment, based on Embodiments 1, 2, and 3, further details the specific implementation process of the continuous control of the switching operation between the charging priority mode, lighting priority mode, and balanced power supply mode described in step S7 of Embodiment 1. Addressing the technical shortcomings of the mode arbitration mechanism using a static single-value hard threshold in the aforementioned embodiments, which is prone to causing mode ping-pong switching near the threshold boundary and instantaneous jumps in the current limiting threshold resulting in inrush current and bus voltage fluctuations, this embodiment proposes a state machine-type mode arbitration mechanism with hysteresis and a gradual current limiting threshold adjustment method based on control cycle steps. This fundamentally eliminates the unstable factors during mode switching and further improves the reliability and stability of system operation.

[0141] In this embodiment, the step of generating the mode arbitration signal based on the comparison result is executed by the embedded microcontroller of the airborne multi-channel power distribution control unit with a 100 Hz control cycle, and implemented using a three-state finite state machine. During system power-on initialization, the system defaults to balanced power supply mode, and the current operating mode variable `current_mode` is initialized to `MODE_BALANCE` (value 0). The mode arbitration signal corresponds one-to-one with the current operating mode variable: `MODE_CHARGE` (value 1) corresponds to the charging priority mode, and `MODE_LIGHT` (value 2) corresponds to the lighting priority mode.

[0142] In this embodiment, a first hysteresis width is first set for the preset charging threshold, and a second hysteresis width is set for the preset lighting priority threshold, thereby generating a mode switching interval with hysteresis.

[0143] The first hysteresis width ranges from 2% to 5% of the preset charging threshold. Its setting is based on the following: In Example 1, the state of charge (SOC) parameters of the airborne energy storage device are estimated using an extended Kalman filter algorithm, with a steady-state error better than ±3%. The first hysteresis width must be greater than this maximum estimation error to avoid false mode triggering due to estimation noise. In this example, the preset charging threshold SOC_th is set to 30% (consistent with Example 1), and the first hysteresis width is 3% of the preset charging threshold, i.e., 0.9% of the SOC percentage. Therefore, the entry threshold for the charging priority mode is SOC_th - first hysteresis width = 30% - 0.9% = 29.1%, and the exit threshold for the charging priority mode is SOC_th + first hysteresis width = 30% + 0.9% = 30.9%.

[0144] The second hysteresis width ranges from 3% to 6% of the preset lighting priority threshold. Its setting is based on the following: the brightness command issued by the ground station via the MAVLink protocol has a transmission jitter of ±2%, and the luminous flux of the LED lighting load has a temperature drift of ±1%. The second hysteresis width must cover all of these fluctuation ranges and retain a 1% safety margin. In this embodiment, the preset lighting priority threshold L_th is set to 60% (consistent with Embodiment 1), and the second hysteresis width is 5% of the preset lighting priority threshold, i.e., 3% of the brightness percentage. Therefore, the entry threshold for the lighting priority mode is L_th + second hysteresis width = 60% + 3% = 63%, and the exit threshold for the lighting priority mode is L_th - second hysteresis width = 60% - 3% = 57%.

[0145] Based on the aforementioned hysteresis-based mode switching interval, the specific generation logic (state transition rule) of the mode arbitration signal is as follows:

[0146] When current_mode == MODE_BALANCE (balanced power supply mode):

[0147] If the state of charge parameter SOC is detected to be ≤29.1%, then current_mode is updated to MODE_CHARGE, and a mode arbitration signal indicating the charging priority mode is generated.

[0148] If the target brightness command parameter L≥63% is detected, the current_mode is updated to MODE_LIGHT, and a mode arbitration signal for the indicator lighting priority mode is generated.

[0149] In other cases, current_mode=MODE_BALANCE remains unchanged, and a mode arbitration signal indicating the balanced power supply mode is generated.

[0150] When current_mode == MODE_CHARGE (charging priority mode):

[0151] If the state of charge parameter SOC is detected to be ≥30.9%, then current_mode is updated to MODE_BALANCE, and a mode arbitration signal indicating the balanced power supply mode is generated.

[0152] If the target brightness command parameter L≥63% is detected, the current_mode is updated to MODE_LIGHT, and a mode arbitration signal for the indicator lighting priority mode is generated.

[0153] In other cases, current_mode=MODE_CHARGE remains unchanged, and a mode arbitration signal indicating the charging priority mode is generated.

[0154] When current_mode == MODE_LIGHT (lighting priority mode):

[0155] If the target brightness command parameter L≤57% is detected, the current_mode is updated to MODE_BALANCE, and a mode arbitration signal indicating the balanced power supply mode is generated.

[0156] If the state of charge parameter SOC is detected to be ≤29.1%, then current_mode is updated to MODE_CHARGE, and a mode arbitration signal indicating the charging priority mode is generated.

[0157] In other cases, keep current_mode=MODE_LIGHT unchanged and generate a mode arbitration signal for the indicator lighting priority mode.

[0158] Through the aforementioned state machine-based mode arbitration mechanism with hysteresis, mode switching is triggered only when the parameter change exceeds the hysteresis width and the state transition condition is met. This effectively filters out minor parameter fluctuations caused by SOC estimation errors, flight control power consumption fluctuations, and brightness command fine-tuning, fundamentally eliminating the ping-pong mode switching phenomenon near the threshold boundary. Actual testing showed that the improved system operated continuously for 24 hours near the threshold boundary without a single erroneous mode switch.

[0159] In this embodiment, the steps for continuously controlling the switching operation between the three power supply modes are as follows: when a change in the mode arbitration signal is detected (i.e., the value of the current_mode variable changes), the mode switching transition process is immediately initiated. Within a preset switching transition time, the current limiting thresholds of the first power output channel and the second power output channel are gradually adjusted from the values ​​corresponding to the current mode to the values ​​corresponding to the target mode according to a preset gradual function with a control cycle of 10 milliseconds. During the entire current limiting threshold adjustment process, the voltage regulation control priority of the third power output channel remains the highest, and its output voltage is always maintained within the preset voltage regulation range of 4.9 volts to 5.1 volts, thereby achieving a smooth and shock-free switching between the three power supply modes.

[0160] The preset switching transition time ranges from 100 milliseconds to 500 milliseconds. The basis for this setting is as follows: In Example 1, it is clearly stated that the current limiting control loop of each channel of the multi-channel power distribution control unit is implemented using a simulated hardware loop with a loop bandwidth of approximately 10 Hz. The corresponding step response time is approximately 100 milliseconds. The switching transition time must be greater than the loop response time to ensure the stability of the adjustment process. Simultaneously, the switching transition time should not exceed 500 milliseconds to avoid affecting the system's response speed to changes in emergency power demand. In this example, the preset switching transition time is set to 300 milliseconds, corresponding to 30 control cycles of 10 milliseconds each. This time length ensures both the smoothness of the current limiting threshold adjustment and meets the system's real-time requirements.

[0161] In this embodiment, a linear gradient function is preferred due to its simple calculation method, ease of implementation, and uniform adjustment process without overshoot. The specific calculation formula for the linear gradient function is: I_step=(I_target-I_current) / N_steps; I_k=I_k-1+I_step; where I_step is the adjustment step size of the current limiting threshold for each 10-millisecond control cycle, I_current is the current limiting threshold corresponding to the current mode before the switch, I_target is the current limiting threshold corresponding to the target mode after the switch, N_steps is the total number of adjustment steps, in this embodiment N_steps=300ms / 10ms=30, and I_k is the current limiting threshold for the k-th control cycle.

[0162] When using an exponential gradient function, the specific calculation formula is: I_k = I_target + (I_current - I_target) exp(-k / τ);

[0163] Where τ is the time constant, set to 1 / 3 of the preset switching transition time, i.e., 100 milliseconds. This setting is based on the fact that in the field of automatic control, an exponential transition process typically uses three times the time constant as the engineering criterion for completion of adjustment, at which point the system output has reached approximately 95% or more of the target value. Setting this time constant τ to 1 / 3 of the preset switching transition time allows the exponential gradual function to smoothly approach the target current limit value within the switching time limit, fully utilizing its characteristics of rapid initial response and smooth, overshoot-free transition. This prevents over-adjustment at the end and ensures that the entire switching process is completed strictly within the specified time. The exponential gradual function is characterized by a rapid rate of change in the initial stage of switching, enabling rapid response to changes in power demand; the rate of change gradually slows down at the end of switching, effectively suppressing overshoot, making it suitable for applications requiring high adjustment accuracy.

[0164] During the mode switching transition, the system monitors the output voltage of the third power output channel in real time, with a sampling frequency of 1 kHz. When the output voltage of the third power output channel is detected to be lower than the preset regulation range lower limit of 4.9 volts for more than 2 milliseconds, the mode switching state variable transition_in_progress is immediately set to PAUSED, pausing the gradual adjustment process of the current limiting threshold. At the same time, the voltage loop proportional gain of the third power output channel is increased from the normal 2 times to 4 times, prioritizing the increase of the power supply of the third power output channel. After the output voltage of the third power output channel recovers to the preset regulation range of 4.9 volts to 5.1 volts and remains stable for a preset duration of 50 milliseconds, the transition_in_progress variable is set back to ACTIVE, and the current limiting threshold gradual adjustment process continues for the remaining (30-k) control cycles from the kth control cycle position at the time of interruption. This mechanism ensures the absolute priority of flight control power supply under any circumstances, avoiding a drop in flight control power supply voltage due to mode switching.

[0165] As a concrete example: Assume that the tethered multi-functional emergency lighting UAV system of this embodiment is initially operating in balanced power supply mode. The parameters of the current control cycle are as follows: the state of charge parameter (SOC) of the onboard energy storage device is 31%, the target brightness command parameter (L) of the LED lighting load is 60%, the total input power of the high-voltage DC bus power supply is 450 watts, the real-time power allocation of the third power output channel is 150 watts, and the available power margin is 270 watts. At this time, the current limiting threshold of the first power output channel is the current value of 7.89 amps (180 watts / 22.8 volts) corresponding to 180 watts, and the current limiting threshold of the second power output channel is the current value of 4.56 amps (114 watts / 25 volts) corresponding to 114 watts.

[0166] As the system operates, the state of charge (SOC) parameters of the airborne energy storage device gradually decrease. When the SOC drops to 29.1%, the conditions for entering the charging priority mode are met. The current_mode variable changes from MODE_BALANCE to MODE_CHARGE, the mode arbitration signal changes, and the system immediately initiates the mode switching transition process.

[0167] Within the 300-millisecond switching transition time, the current limiting threshold of the first power output channel is gradually adjusted from 7.89 amps to the first preset current limiting value of 10 amps corresponding to the charging priority mode, following a linear gradient function. The current limiting threshold of the second power output channel is gradually adjusted from 4.56 amps to the value of 2.16 amps ((270 W - 228 W) / 25 V) corresponding to the available power margin minus the power of the first channel. Within each 10-millisecond control cycle:

[0168] The adjustment step size of the current limiting threshold for the first channel is (10-7.89) / 30≈0.0703 Amperes / cycle;

[0169] The adjustment step size for the current limiting threshold of the second channel is (2.16-4.56) / 30 = -0.08 Amperes / cycle;

[0170] At the end of each 10-millisecond control cycle, the microcontroller writes the calculated current current limiting threshold into the CURRENT_LIMIT register (address 0x03) of the corresponding channel synchronous rectifier buck converter via the serial peripheral interface (SPI). The register data format is a 16-bit unsigned integer with a resolution of 0.001 amperes.

[0171] Throughout the switching process, the output voltage of the third power output channel remained within the preset regulation range of 4.9V to 5.1V. When the switching reached the 15th control cycle (150 milliseconds), a brief drop in the output voltage of the third power output channel to 4.85V was detected, lasting for 3 milliseconds. The system immediately set the `transition_in_progress` variable to `PAUSED`, pausing the gradual adjustment of the current limiting threshold, and simultaneously increasing the voltage loop proportional gain of the third power output channel to 4x. After 20 milliseconds, the output voltage of the third power output channel recovered to 5.0V and remained stable for 50 milliseconds. The system then reset the `transition_in_progress` variable to `ACTIVE`, continuing the adjustment process for the remaining 15 control cycles from the 15th control cycle position.

[0172] After 300 milliseconds, the current limiting threshold adjustment was completed, and the system operated stably in charging priority mode. At this time, the first power output channel rapidly charged the onboard energy storage device with a current of 10 amps, and the second power output channel provided auxiliary lighting with a current of 2.16 amps. Actual measurements showed that the bus voltage fluctuation was less than ±0.5% and the LED lighting current ripple was less than ±1% throughout the entire switching process, with no flickering perceptible to the human eye.

[0173] When the state of charge (SOC) parameter of the airborne energy storage device rises to 30.9%, the exit condition for charging priority mode is met. The `current_mode` variable changes from `MODE_CHARGE` to `MODE_BALANCE`, and the system restarts the mode switching process. The current limiting threshold of the first power output channel is gradually adjusted from 10 amps back to 7.89 amps, and the current limiting threshold of the second power output channel is gradually adjusted from 2.16 amps back to 4.56 amps. The entire switching process is also free of inrush current and bus voltage fluctuations.

[0174] It is understood that the hysteresis width, switching transition time, gradual function parameters, register address and data format, and specific values ​​in each calculation example listed in this embodiment are all exemplary and should not be used to limit the scope of protection of this invention. In actual implementation, those skilled in the art can determine the corresponding parameter values ​​based on specific system specifications, load characteristics and test data without creative effort.

[0175] This embodiment addresses the technical problems of the mode arbitration mechanism in the aforementioned embodiments, which uses a static single-value hard threshold two-level judgment method. These problems include the tendency for small parameter fluctuations near the threshold boundary to trigger mode ping-pong switching, and the inrush current and bus voltage jitter caused by instantaneous jumps in the current-limiting threshold during mode switching. By introducing a mode arbitration mechanism with hysteresis width and a three-state finite state machine, hysteresis intervals are set for the preset charging threshold and preset lighting priority threshold, respectively, based on the state of charge estimation error and the coverage range of brightness command transmission jitter and temperature drift. This creates independent upper and lower boundaries for entering and exiting the charging priority mode and the lighting priority mode. Mode switching is only triggered when the parameter change exceeds the corresponding hysteresis width and the state transition condition is met. This filters out small parameter fluctuations caused by state of charge estimation noise, flight control power consumption fluctuations, and brightness command fine-tuning from the decision-making source. The jitter effectively suppresses the ping-pong switching phenomenon near the threshold boundary. At the mode switching execution level, a gradual adjustment method for the current limiting threshold based on control cycle stepping is adopted. Within the preset switching transition time, the current limiting thresholds of the first and second power output channels are gradually adjusted from the corresponding values ​​of the current mode to the corresponding values ​​of the target mode with a constant step size. This makes the change of the current limiting threshold uniform and without overshoot, reducing the bus inrush current and LED lighting flicker caused by instantaneous jumps. During the entire switching process, the output voltage of the third power output channel is monitored in real time. When the output voltage is lower than the lower limit of the preset voltage regulation range for a certain period of time, the gradual adjustment process is paused and the flight control power supply is increased first. After the output voltage stabilizes and recovers to the preset voltage regulation range, the remaining adjustment is continued from the interrupted position, ensuring the priority and safety of the flight control power supply during the mode switching process.

[0176] Example 5

[0177] Please see Figure 4 Another embodiment of the present invention provides: a power distribution system for a multi-functional emergency lighting drone system, comprising:

[0178] The parameter acquisition module is configured to: acquire ground AC mains power input, boost it to a preset transmission voltage level through a step-up transformer module, process it through an inverter and rectifier filter module to obtain a high-voltage DC bus power supply; at the same time, it acquires the state of charge parameters of the airborne energy storage device, the real-time power consumption parameters of the UAV flight control unit, and the target brightness command parameters of the LED lighting load in real time.

[0179] The power budget module is configured to: calculate the total power demand and available power margin of the current system based on the state of charge parameters, real-time power consumption parameters and target brightness command parameters through a power allocation algorithm; at the same time, compare the state of charge parameters with a preset charging threshold and the target brightness command parameters with a preset lighting priority threshold to generate a mode arbitration signal.

[0180] Specifically, the mode arbitration signal in this embodiment is as follows: The state of charge (SOC) parameter is compared with a preset charging threshold SOC_th. When SOC is less than SOC_th, a mode arbitration signal indicating a charging priority mode is generated. When SOC is greater than or equal to SOC_th, the target brightness command parameter is compared with a preset lighting priority threshold. If the target brightness command parameter reaches or exceeds the preset lighting priority threshold, a mode arbitration signal indicating a lighting priority mode is generated; otherwise, a mode arbitration signal indicating a balanced power supply mode is generated. The balanced power supply mode in this embodiment refers to prioritizing the voltage stability of the power supply from the third power output channel to the UAV flight control unit, and distributing the available power margin between the first and second power output channels according to a weighted ratio determined by the state of charge parameter and the target brightness command parameter.

[0181] The multi-channel power distribution control module is configured to: input the high-voltage DC bus power supply to an onboard multi-channel power distribution control unit, wherein the multi-channel power distribution control unit receives and parses the mode arbitration signal, and executes the corresponding shunt control strategy according to the current operating mode indicated by the mode arbitration signal;

[0182] The charging priority current distribution control module is configured to: if the mode arbitration signal indicates a charging priority mode, the multi-channel power distribution control unit sets the current limiting threshold of the first power output channel to a first preset current limiting value to perform fast charging on the airborne energy storage device; simultaneously, it acquires the available power margin and dynamically adjusts the output voltage clamping value or output current limiting value of the second power output channel according to the available power margin to limit the driving power of the LED lighting load to a level determined by the available power margin;

[0183] The lighting priority current distribution control module is configured to: if the mode arbitration signal indicates a lighting priority mode, the multi-channel power distribution control unit adjusts the output current of the second power output channel to a target current value that matches the target brightness command parameter to meet the power requirements of the LED lighting load; simultaneously, it lowers the current limiting threshold of the first power output channel to a second preset current limiting value or shuts off the first power output channel, and monitors the output voltage of the third power output channel in real time, adjusting the loop response of the multi-channel power distribution control unit to maintain the voltage supplied by the third power output channel to the UAV flight control unit within a preset voltage regulation range;

[0184] The balanced power supply and current distribution control module is configured to: if the mode arbitration signal indicates a balanced power supply mode, the multi-channel power distribution control unit determines the minimum dynamic power consumption value required to maintain the UAV's flight attitude and resist wind disturbance based on the real-time power consumption parameters, and allocates power not lower than the minimum dynamic power consumption value to the third power output channel; and calculates the remaining power of the high-voltage DC bus power supply after deducting the power allocated to the third power output channel, and performs real-time current distribution of the remaining power between the first power output channel and the second power output channel according to the distribution weight determined by the ratio of the state of charge parameter to the target brightness command parameter;

[0185] The feedback and overload protection module is configured to: continuously collect real-time output voltage, real-time output current, and temperature parameters of each power output channel during power supply; feed the real-time output voltage, real-time output current, and temperature parameters back to the power allocation algorithm; update the total demand power, available power margin, and mode arbitration signal according to the feedback parameters to control the multi-channel power allocation control unit to perform dynamic and smooth switching between the charging priority mode, lighting priority mode, and balanced power supply mode; and perform overload protection when the real-time output current of any power output channel exceeds the overcurrent protection threshold of the corresponding channel.

[0186] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments under the guidance of the present invention without departing from the spirit and scope of the present invention. All of these variations are within the protection scope of the present invention.

[0187] If the technical solution disclosed herein involves personal information, the product using this technical solution has clearly informed the user of the personal information processing rules and obtained the user's voluntary consent before processing the personal information. If the technical solution disclosed herein involves sensitive personal information, the product using this technical solution has obtained the user's separate consent before processing the sensitive personal information, and also meets the requirement of "express consent". For example, at personal information collection devices such as cameras, clear and prominent signs are set up to inform users that they have entered the scope of personal information collection and that personal information will be collected. If an individual voluntarily enters the collection scope, it is deemed that they have agreed to the collection of their personal information; or on the personal information processing device, with clear signs / information informing users of the personal information processing rules, authorization is obtained from the individual through pop-up information or by asking the individual to upload their personal information; wherein, the personal information processing rules may include information such as the personal information processor, the purpose of personal information processing, the processing method, and the types of personal information processed.

Claims

1. A power diversion method applied to a multi-functional emergency lighting drone system, characterized in that, include: Obtain a high-voltage DC bus power supply and collect in real time the state of charge parameters of the airborne energy storage device, the real-time power consumption parameters of the UAV flight control unit, and the target brightness command parameters of the LED lighting load; The state of charge parameter is compared with a preset charging threshold, and the target brightness command parameter is compared with a preset lighting priority threshold. A mode arbitration signal is generated based on the comparison result. When the comparison result meets the charging priority trigger condition, the charging priority mode is indicated. When the comparison result meets the lighting priority trigger condition, the lighting priority mode is indicated. The remaining comparison results indicate the balanced power supply mode. The balanced power supply mode distributes the available power margin between the first power output channel and the second power output channel according to the weighted ratio determined by the state of charge parameter and the target brightness command parameter, and maintains the output voltage of the third power output channel within a preset voltage regulation range. The high-voltage DC bus power supply is input to the airborne multi-channel power distribution control unit. The multi-channel power distribution control unit performs current limiting control on the first power output channel and the second power output channel and voltage regulation control on the third power output channel according to the mode indicated by the mode arbitration signal, so as to distribute the power of the high-voltage DC bus power supply. The real-time output voltage and real-time output current of the first power output channel, the second power output channel and the third power output channel are collected and fed back to the power allocation algorithm to update the total demand power, available power margin and mode arbitration signal, and continuously control the switching operation between the charging priority mode, the lighting priority mode and the balanced power supply mode. When the mode arbitration signal indicates the charging priority mode, the multi-channel power distribution control unit sets the current limiting threshold of the first power output channel to a first preset current limiting value, and adjusts the output voltage clamping value or output current limiting value of the second power output channel to a limit value associated with the available power margin. When the mode arbitration signal indicates the lighting priority mode, the multi-channel power distribution control unit adjusts the output current of the second power output channel to the target current value that matches the target brightness command parameter, sets the current limiting threshold of the first power output channel to the second preset current limiting value or shuts off the first power output channel, and adjusts the loop response of the multi-channel power distribution control unit based on the real-time output voltage of the third power output channel so that the output voltage of the third power output channel is maintained within the preset voltage regulation range. The step of generating a pattern arbitration signal based on the comparison result includes: The State of Charge (SOC) parameter is compared with a preset charging threshold SOC_th. When SOC is less than SOC_th, a mode arbitration signal indicating a charging priority mode is generated. When SOC is greater than or equal to SOC_th, the target brightness command parameter is compared with a preset lighting priority threshold. If the target brightness command parameter reaches or exceeds the preset lighting priority threshold, a mode arbitration signal indicating a lighting priority mode is generated. Otherwise, a mode arbitration signal indicating a balanced power supply mode is generated.

2. The power diversion method for a multi-functional emergency lighting UAV system as described in claim 1, characterized in that, When the mode arbitration signal indicates the balanced power supply mode, the multi-channel power distribution control unit determines the minimum dynamic power consumption value based on the real-time power consumption parameters, controls the output power of the third power output channel to be no less than the minimum dynamic power consumption value, calculates the remaining power obtained by deducting the allocated power of the third power output channel from the available power margin, and allocates the remaining power to the first power output channel and the second power output channel according to the weighted ratio. The multi-channel power distribution control unit is also used to shut down the corresponding channel or perform power callback when the real-time output current of any power output channel exceeds the overcurrent protection threshold of the corresponding channel.

3. The power diversion method for a multi-functional emergency lighting UAV system as described in claim 2, characterized in that, The process of determining the weighting ratio includes: Read the remaining power value obtained after deducting the allocated power of the third power output channel from the high-voltage DC bus power supply; Based on the state of charge parameters, determine the maximum allowable charging power and the minimum effective charging power corresponding to the current state of charge; Based on the target brightness command parameters and the forward voltage drift compensation value under the current junction temperature conditions, the target driving power required for the LED lighting load to reach the target illuminance is determined, and the minimum effective lighting power is determined based on the human eye flicker perception threshold.

4. The power diversion method for a multi-functional emergency lighting UAV system as described in claim 3, characterized in that, The process of determining the weighting ratio also includes: The maximum allowable charging power and the rated safe power of the LED lighting load are respectively determined as the upper limit boundary of the allocated power of the first power output channel and the upper limit boundary of the allocated power of the second power output channel. The minimum effective charging power and the minimum effective lighting power are respectively determined as the lower limit boundary of the allocated power of the first power output channel and the lower limit boundary of the allocated power of the second power output channel. The remaining power value is compared with the sum of the lower limit boundary of the first power output channel and the lower limit boundary of the second power output channel, and the sum of the upper limit boundary of the first power output channel and the upper limit boundary of the second power output channel.

5. The power diversion method for a multi-functional emergency lighting UAV system as described in claim 4, characterized in that, The process of determining the weighting ratio also includes: When the remaining power value is within the range of the sum of the lower limit boundary to the sum of the upper limit boundary, the initial weighting ratio is calculated by the ratio of the inverse proportional function value of the state of charge parameter to the direct proportional function value of the target brightness command parameter. When the remaining power value is lower than the sum of the lower limit boundaries, the initial weighting ratio is corrected according to the preset priority bias coefficient so that the allocation result first satisfies the allocation power lower limit boundary corresponding to the channel with higher priority. When the remaining power value is higher than the sum of the upper limit boundaries, the allocated power of the first power output channel is limited to the upper limit boundary of the allocated power of the first power output channel, the allocated power of the second power output channel is limited to the upper limit boundary of the allocated power of the second power output channel, and the power exceeding the sum of the upper limit boundaries is marked as a power margin to be allocated pool.

6. The power diversion method for a multi-functional emergency lighting UAV system as described in claim 5, characterized in that, The power margin allocation pool is configured to gradually transfer accumulated power to the first power output channel, the process including: When the accumulated value of the power margin to be allocated pool continuously exceeds the preset accumulated upper limit threshold and the duration reaches the preset duration, an auxiliary charging enable signal is generated; Obtain the state of charge (SOC) parameters of the airborne energy storage device for the current control cycle, and determine the auxiliary charging pulse duty cycle for the current control cycle based on the difference between the SOC parameters and the fully charged SOC.

7. The power diversion method for a multi-functional emergency lighting UAV system as described in claim 6, characterized in that, The process of gradually transferring the power accumulated in the power margin allocation pool to the first power output channel further includes: Based on the auxiliary charging pulse duty cycle and the preset auxiliary charging pulse frequency, a set of intermittent auxiliary charging pulse sequences is generated. The intermittent auxiliary charging pulse sequences are applied to the current limiting control loop of the first power output channel to superimpose an intermittent power margin consumption current component on the basis of the preset current limiting value of the first power output channel. During the execution of the intermittent auxiliary charging pulse sequence, the cumulative value of the power margin allocation pool is continuously monitored. When the cumulative value of the power margin allocation pool drops below a preset cumulative lower limit threshold, the auxiliary charging enable signal is revoked, the output of the intermittent auxiliary charging pulse sequence is stopped, and the current limiting control of the first power output channel is restored to the normal state where the power margin consumption current component is not superimposed.

8. A power supply system for a multi-functional emergency lighting drone system, used to implement the power supply method for a multi-functional emergency lighting drone system as described in any one of claims 1-7, characterized in that, include: The parameter acquisition module is configured to: obtain a high-voltage DC bus power supply and acquire in real time the state of charge parameters of the airborne energy storage device, the real-time power consumption parameters of the UAV flight control unit, and the target brightness command parameters of the LED lighting load; The power budget module is configured to: compare the state of charge parameter with a preset charging threshold, compare the target brightness command parameter with a preset lighting priority threshold, and generate a mode arbitration signal based on the comparison result. When the comparison result meets the charging priority trigger condition, it indicates the charging priority mode; when it meets the lighting priority trigger condition, it indicates the lighting priority mode; and other comparison results indicate the balanced power supply mode. The balanced power supply mode distributes the available power margin between the first power output channel and the second power output channel according to the weighted ratio determined by the state of charge parameter and the target brightness command parameter, and maintains the output voltage of the third power output channel within a preset voltage regulation range. The multi-channel power distribution control module is configured to: input the high-voltage DC bus power supply to the airborne multi-channel power distribution control unit, wherein the multi-channel power distribution control unit performs current limiting control on the first power output channel and the second power output channel and voltage regulation control on the third power output channel according to the mode indicated by the mode arbitration signal, so as to distribute the power of the high-voltage DC bus power supply; The real-time output voltage and real-time output current of the first power output channel, the second power output channel and the third power output channel are collected and fed back to the power allocation algorithm to update the total power demand, available power margin and mode arbitration signal, and continuously control the switching operation between the charging priority mode, the lighting priority mode and the balanced power supply mode.

Images