Unmanned aerial vehicle parallel battery control system and method with redundancy and inter-pack balancing

Abstract

The application provides a redundancy and inter-battery equalization unmanned aerial vehicle parallel battery control system and method, and belongs to the technical field of multi-rotor unmanned aerial vehicles; a plurality of high-voltage battery packs are included, each high-voltage battery pack is internally provided with a second BMS, the output ends of the high-voltage battery packs are connected to the power distribution bus of the unmanned aerial vehicle in a parallel mode through power lines, and are used for providing an energy source for the unmanned aerial vehicle; each high-voltage battery pack is also electrically connected with an equalization bus, the equalization bus is independently arranged from the power distribution bus, and is used for carrying energy transmission in an inter-battery equalization process between the battery packs; a first BMS is in communication connection with the unmanned aerial vehicle flight control system and the second BMS in the plurality of high-voltage battery packs; before the start of a flight task of the unmanned aerial vehicle, the number of high-voltage battery packs that need to be connected to the power distribution bus is determined according to a task load and an estimated flight range; and the high-voltage battery packs are also in communication connection with the unmanned aerial vehicle flight control system through a redundant CAN bus.

Description

Technical Field

[0001] This invention relates to the field of multi-rotor unmanned aerial vehicle (UAV) technology, and in particular to a UAV parallel battery control system and method with redundancy and packet balance. Background Technology

[0002] Heavy-duty multi-rotor drones, as aerial work platforms with large payloads and vertical takeoff and landing capabilities, are widely used in logistics transportation, agricultural and forestry plant protection, emergency fire fighting, and special hoisting. Their power battery system, as the sole energy source for the aircraft, directly determines the overall payload capacity, endurance, and flight safety. As drones develop towards larger takeoff weights (>150kg) and longer flight times (>30 minutes), the battery pack system not only needs extremely high energy density to support long-endurance operations but also requires a power supply architecture with electrical redundancy, safety, and flexible mission configuration capabilities. Furthermore, the power battery pack of heavy-duty drones is a high-value core component, requiring dozens to hundreds of charge-discharge cycles, multiple battery swaps, and long-term storage throughout its lifecycle, posing an urgent need for continuous monitoring and full lifecycle management by the battery management system (BMS).

[0003] In conventional solutions, the power supply system architecture for heavy-duty drone battery packs includes the following options: 1. Low-voltage multi-pack parallel power supply architecture; This solution uses battery packs with a nominal voltage of less than 60V to supply power to all ESCs and motors through physical parallel connection. However, under the power requirements of tens of kilowatts for heavy-duty drones weighing over 150kg, this solution has the following inherent defects: First, under high current conditions of over 300A, the I²R heat loss of wires and connectors is extremely high, significantly reducing the system's endurance efficiency. Moreover, the large cross-section cables and additional heat dissipation structures lead to a sharp increase in system weight, severely restricting the effective payload capacity. Second, the parallel circuit is prone to generating uncontrollable internal circulating currents due to the inconsistency between voltage difference and internal resistance. Even with diode isolation, the on-state voltage drop under high current will cause huge heat loss and energy waste, severely limiting the engineering practicality of this solution in heavy-duty scenarios. 2. High-voltage series centralized power supply architecture, drawing on the 400V / 800V high-voltage platform technology in the new energy vehicle field, connects a large number of battery cells in series to obtain a high voltage platform of 400V to 800V. This solution significantly reduces operating current by increasing the voltage platform, reducing wiring weight and line losses, and significantly improving range. However, it has a fatal single-point failure problem: any open circuit in a battery cell, false triggering of BMS protection, or abnormal disconnection of a high-voltage relay will cause the entire aircraft to instantly lose all power, which means an unacceptable crash risk for large, heavy-duty aircraft. In addition, this architecture cannot flexibly increase or decrease the battery pack capacity according to the mission load, limiting operational economics. 3. Multi-pack independent power supply architecture: To avoid the risk of single-point failure, some solutions adopt a power supply method where each battery pack has an independent power branch. However, the independent power supply architecture is constrained by gravity balance, and the number of battery packs is limited by the number of motor shafts, resulting in poor configuration flexibility. When a battery in a branch fails, the corresponding power loss can easily cause severe flight attitude imbalance. The system needs to cut off additional power supply battery packs according to the overall force balance state, leading to a decrease in load-bearing efficiency or even failure to operate normally. 4. Multi-pack parallel high-voltage architecture: The flight controller centrally controls the on / off of the pre-charge relays and main relays of each battery pack. However, this solution has the potential for single-point failure due to centralized control: First, when the flight controller resets or reconnects due to an abnormality, the main relays of all battery packs will disconnect simultaneously, and the entire aircraft will instantly lose all power. Second, when communication between the flight controller and a battery pack is interrupted, if a serious fault such as cell overvoltage or overtemperature occurs inside that battery pack, the flight controller cannot issue a cut-off command in time, and the faulty battery pack will continue to threaten system safety. This approach relies entirely on the availability of the flight controller to isolate battery pack faults, thus failing to achieve autonomous safety protection at the battery pack level.

[0004] Furthermore, existing technologies generally lack a multi-level BMS management architecture. The battery pack acts only as a passive execution unit, lacking local autonomous decision-making capabilities. When the battery pack is detached from the main unit, the internal BMS loses its power supply, or the onboard BMS ceases operation when the battery is disconnected, resulting in the inability to perform status monitoring, fault diagnosis, and data recording during offline periods, making it difficult to meet the needs of full battery lifecycle management. The inter-pack circulating current and balancing issues in high-voltage multi-pack parallel scenarios cannot be ignored. High-voltage battery packs consist of dozens to hundreds of cells connected in series. Due to differences in internal resistance, polarization characteristics, self-discharge rate, and aging degree, the total voltage and SOC of different battery packs often differ. Direct parallel connection to the bus will generate uncontrollable circulating currents, leading to energy loss and accelerated cell aging, and in severe cases, even triggering overcurrent protection.

[0005] Therefore, it is necessary to provide a parallel battery control system and method for UAVs with redundancy and inter-packet balancing, which can reduce power line losses and system weight while achieving single-point failure protection at the battery pack level, local autonomous fault isolation, cross-level interaction and fault-tolerant operation, active inter-packet balancing control, offline self-sustaining management and flexible capacity configuration. Summary of the Invention

[0006] In view of this, the present invention proposes a parallel battery control system and method for UAVs that provides parallel application and fault tolerance capabilities for multiple battery packs, enables inter-packet balancing, and has redundancy and inter-packet balancing.

[0007] On one hand, the present invention provides a parallel battery control system for unmanned aerial vehicles (UAVs) with redundancy and inter-packet equalization, comprising: Multiple high-voltage battery packs are installed in the drone's battery compartment via mechanical locking devices. Each high-voltage battery pack has a built-in second BMS. The output terminals of each high-voltage battery pack are connected to the drone's power distribution bus via power lines in parallel to provide power to the drone. Each high-voltage battery pack is also electrically connected to the balancing bus, which is set up independently of the power distribution bus and is used to carry out energy transfer during the balancing process between the battery packs. The first BMS communicates with the UAV flight control system and the second BMS in the multiple high-voltage battery packs via the first CAN communication bus. The first BMS is also electrically connected to the multiple high-voltage battery packs one by one via the high-voltage interlock detection line to detect the physical connection status of each high-voltage battery pack. Before the start of the UAV flight mission, the number of high-voltage battery packs that need to be connected to the power distribution bus is determined based on the mission load and expected flight range. The high-voltage battery packs are electrically connected to the power distribution bus and the equalization bus respectively. The high-voltage battery packs are also directly connected to the UAV flight control system through the second CAN communication bus.

[0008] Based on the above technical solution, preferably, the multiple high-voltage battery packs further include several battery groups, a high-voltage circuit breaker (BDU), a smart junction box (BJB), several third battery management systems (BMS), and a bidirectional DC / DC converter. The battery groups are sequentially and electrically connected. The negative terminal of the first battery group is electrically connected to the positive terminal of the adjacent battery group. The positive terminal of the first battery group and the negative terminal of the last battery group are both electrically connected to the high-voltage circuit breaker (BDU). The smart junction box (BJB) is located on the circuit between the negative terminal of the last battery group and the high-voltage circuit breaker (BDU), and is electrically connected to the last battery group, the high-voltage circuit breaker (BDU), and the second BMS, respectively. It is used to acquire shunt sampling signals and temperature sensing signals, and send them to the second BMS. The second BMS then processes the shunt sampling signals and temperature sensing signals according to the data. The high-voltage battery pack's real-time charging and discharging current, after temperature drift correction, is obtained from the temperature sensor signal. This is further combined with time integration to obtain the state of charge (SOC) of the high-voltage battery pack. A fusion algorithm combining ampere-hour integration and open-circuit voltage correction is used to improve the accuracy of the SOC estimation. A bidirectional DC / DC converter is electrically connected to the positive terminal of the first battery pack, the negative terminal of the last battery pack, the second BMS, and the balancing bus, respectively. The bidirectional DC / DC converter receives control signals from the second BMS to achieve inter-pack balancing. Several third BMSs are electrically connected to the two ends of several battery packs, one-to-one. Each third BMS is equipped with a passive balancing circuit and is used to acquire the cell voltage and temperature data of each battery pack.

[0009] Preferably, the second BMS is directly connected to the first BMS to form a first CAN communication bus; several third BMSs are connected to each other and to the second BMS via a daisy-chain communication system; several third BMSs are also directly connected to the UAV flight control system via the second CAN communication bus; when the first BMS is working normally, the first BMS communicates with the second BMS via the first CAN communication bus and sends a polling command to the second BMS; the second BMS obtains the voltage and temperature data of the cells of several battery packs collected by each third BMS through polling, and then returns the obtained voltage and temperature data of the cells to the first BMS via the first CAN communication bus; when it is detected that the first BMS has a timeout in sending a polling command, it automatically switches to the second CAN communication bus, and each third BMS directly sends the voltage and temperature data of the cells of several battery packs to the UAV flight control system; When the voltage difference between different battery packs exceeds a preset threshold, the second BMS or the UAV flight control system sends an equalization command to the third BMS, closes the passive equalization circuit corresponding to the battery pack with the highest voltage, and discharges the battery pack through the equalization resistor until the voltage difference between the battery packs does not exceed the preset threshold.

[0010] Preferably, both the second and third BMS are equipped with independent CAN controllers; the second BMS also has a built-in offline self-sustaining joint correction module and a real-time power capacity estimation module; after the high-voltage battery pack is removed from the drone, if the removed high-voltage battery pack still has residual power, the second BMS performs offline self-sustaining SOC and SOH joint correction on the removed high-voltage battery pack through the built-in offline self-sustaining joint correction module, and distinguishes between self-discharge abnormalities and SOH by combining the offline voltage drop and charging voltage rise rate at the next power-on. The system estimates the offset and performs multi-cycle sliding calibration. The second or third BMS of the detached high-voltage battery pack sends the status information of the battery pack or battery group to the outside world through the CAN controller. The status information includes the voltage, temperature, SOC, SOH, cumulative cycle count, and historical fault records of the high-voltage battery pack or battery group. The real-time power capability estimation module outputs power capability values ​​at different time scales based on the preset SOC-temperature-SOH multi-dimensional power table and combined with real-time voltage change rate dynamic correction. The first BMS determines whether the power capability value meets the requirements of the flight mission based on the power capability values ​​reported by multiple high-voltage battery packs, and performs graded resusceptibility judgment and decision when a high-voltage battery pack fails.

[0011] Preferably, the second BMS performs offline self-sustaining joint correction of SOC and SOH on the detached high-voltage battery pack through a built-in offline self-sustaining joint correction module. Upon the next power-on, it integrates the offline voltage drop and charging voltage rise rates to distinguish between self-discharge anomalies and SOH estimation offsets, and performs multi-cycle sliding calibration, including the following: Autonomous wake-up and OCV correction in offline state: When the high-voltage battery pack is detached from the drone and not connected to the charging equipment, the second BMS enters sleep mode, but still maintains RTC timing. The second BMS automatically wakes up once every fixed time interval and performs the following operations: the voltage of each cell of all battery packs is collected through the third BMS, and the voltage value, voltage extreme value and average value of each cell are recorded; if the time since the last time the battery pack was powered off from the high-voltage bus or charging equipment is greater than the preset time, it is considered that the cell has been sufficiently rested and meets the open-circuit OCV correction conditions. Then, based on the current measured voltage of each cell, the corresponding offline OCV-SOC mapping value is obtained by looking up the table, and an open-loop correction is performed on the current SOC of each cell. The SOC correction magnitude is obtained by subtracting the SOC of each cell recorded before the last sleep from the offline OCV-SOC mapping value. The timestamp of the current open-loop correction, the SOC before the open-loop correction, the SOC after the open-loop correction, the SOC correction magnitude, and the voltage drop of each cell in the first hour after going offline are used to form the offline SOC OCV correction history record. Joint calibration upon next power-on: When the high-voltage battery pack is reconnected to the drone, connected to the charging equipment, or actively woken up by the handheld terminal, the second BMS first reads the offline SOC OCV correction history stored during offline operation and executes the joint calibration logic: The second BMS extracts the cell voltage and SOC recorded at the last power-off before offline operation, the cell voltage drop and SOC obtained from the first OCV correction after offline operation, the voltage rise curve and charging capacity of each cell during the most recent M complete charging processes, and the static open-circuit voltage and current SOC of each cell measured at the current power-on moment, calculates the SOC attenuation of each cell, and... The decay of cell SOC is calculated by subtracting the SOC of each cell at the current power-on moment from the SOC recorded at the last power-down before offline. The theoretical self-discharge expected decay is calculated by obtaining the estimated discharge capacity of the cell from the typical self-discharge rate and offline duration in the cell specification, dividing it by the rated capacity of the cell, and then multiplying by 100%. The absolute value of the result of subtracting the theoretical self-discharge expected decay from the decay of each cell SOC is compared with a preset capacity threshold. If the absolute value is less than or equal to the preset capacity threshold, the cell is considered to have normal self-discharge. If the absolute value is greater than the preset capacity threshold, the cell is judged to have an abnormality, and attribution analysis and multi-cycle sliding calibration are performed.

[0012] Preferably, the real-time power capability estimation module outputs power capability values ​​at different time scales based on a preset SOC-temperature-SOH multidimensional power table and dynamic correction using real-time voltage change rate. The first BMS determines whether the power capability values ​​reported by multiple high-voltage battery packs meet the requirements of the flight mission, including the following: Each high-voltage battery pack's second BMS has a pre-set multi-dimensional power capability table SOP. Using cell parameters provided by the cell manufacturer and after the high-voltage battery packs are assembled, the system performs measured power capability calibration. This includes using temperature, the high-voltage battery pack's SOC, and SOH as input dimensions, and using the maximum stable power output of the high-voltage battery pack over 10 seconds, 60 seconds, and 10 minutes as the output. During flight, the second BMS performs the following real-time estimations at a preset cycle: acquiring the high-voltage battery pack's SOC and SOH; reading the current minimum cell voltage and maximum cell temperature from the third BMS; obtaining the basic power capability through linear interpolation in the multi-dimensional power capability table SOP; and incorporating voltage change rate and instantaneous current for real-time correction to obtain the final power capability of the high-voltage battery pack. The first BMS polls all online high-voltage battery packs' second BMSs at a preset cycle to obtain the final power capability of each high-voltage battery pack. These final power capabilities are then summed to obtain the total system power capability. The total system power capability is compared with the power requirements for the current flight state issued by the UAV flight control system to determine whether the current total system power capability meets the flight mission requirements.

[0013] Preferably, the step of performing graded reseverability judgment and decision-making when a high-voltage battery pack fails involves the following: before disconnecting the faulty high-voltage battery pack from the distribution bus due to cell overheating, insulation failure, or communication loss, the first BMS obtains the predicted power demand value of the current flight mission for a future period from the flight control system, evaluates the final power capability of the remaining high-voltage battery packs and compares it with the predicted power demand value of the current flight mission, and executes instructions based on the comparison result: safe removal and continuation of mission, return to base after disconnection, immediately reduce altitude and fly to the landing point after disconnection, or prohibit disconnection; after the high-voltage battery pack is disconnected, the first BMS updates the online list of high-voltage battery packs and reports the final power capability of the remaining high-voltage battery packs to the flight control system, and the flight control system dynamically updates the flight boundary parameters.

[0014] Preferably, the flight boundary parameters include maximum climb rate, maximum flight speed, and return-to-home threshold battery level.

[0015] On the other hand, the present invention also provides a parallel battery control method for UAVs with redundancy and packet balance, comprising the following steps: S1: Configure the above-mentioned parallel battery control system for drones with redundancy and packet balance; S2: After each high-voltage battery pack is mechanically installed and connected to the drone, the first BMS detects the connection status of each high-voltage battery pack interface through the high-voltage interlock detection line. After confirming that the physical connection status is reliable, the first BMS sends a wake-up command through the first CAN bus to wake up the second BMS in each high-voltage battery pack. Each second BMS performs a self-test. After the self-test is passed, the first BMS polls each second BMS to obtain the status information of each high-voltage battery pack. If a high-voltage battery pack has a fault state that prohibits access, the first BMS marks the high-voltage battery pack as unavailable and only includes high-voltage battery packs that meet the conditions in the power-on sequence. If the maximum voltage difference between each high-voltage battery pack exceeds 10V, the equalization process is entered. S3: After receiving the power-on command from the flight control system, the first BMS sends access commands to the second BMS of each high-voltage battery pack to be connected in sequence, so that the high-voltage battery packs are smoothly connected to the power distribution bus; if the time interval between the connection of two adjacent high-voltage battery packs to the power distribution bus is not less than 500ms, until all high-voltage battery packs are connected to the power distribution bus. S4: After all high-voltage battery packs are connected to the grid, they enter normal operation. The first BMS polls the second BMS of each high-voltage battery pack according to a preset cycle to obtain real-time operating data of each high-voltage battery pack, including total voltage, voltage of each individual cell, charging and discharging current, cell temperature, and cell SOC and SOH status information. The first BMS periodically reports the obtained status information to the UAV flight control system. Inside the high-voltage battery pack, each third BMS collects the voltage and temperature of each cell and reports it to the second BMS through a daisy chain. The second BMS activates passive equalization within the pack to eliminate voltage differences between different battery packs based on the degree of voltage consistency of the cells in different battery packs, until the voltage of the battery packs is consistent or the maximum equalization time is reached. S5: During system operation, the first BMS, second BMS, and third BMS continuously perform fault diagnosis on the high-voltage battery pack and system status. Based on the severity of the fault diagnosis, they perform graded resusceptibility judgment and decision-making. Before disconnecting the faulty high-voltage battery pack from the power distribution bus, the first BMS obtains the predicted power demand value of the current flight mission for a period of time from the flight control system, evaluates and compares the final power capacity of the remaining high-voltage battery pack with the predicted power demand value of the current flight mission, and executes instructions such as safe removal and continuation of mission, return to base after disconnection, immediately reduce altitude and fly to the landing point after disconnection, or prohibit disconnection based on the comparison results. The third BMS communicates with the UAV flight control system via the second CAN communication bus. When the first BMS fails, the second CAN communication bus is activated to directly report the status information of the high-voltage battery pack and the status of the first BMS to the UAV flight control system. The UAV flight control system determines the predicted power demand based on the status information reported by each high-voltage battery pack and adjusts the flight strategy to ensure the UAV returns safely.

[0016] Preferably, when the high-voltage battery pack is disconnected from the drone for charging, the second BMS communicates with the charging equipment via the CAN communication interface. The second BMS enters the charging management mode and, in conjunction with the temperature and voltage of each cell in the battery pack collected by the third BMS and the preset charging strategy, calculates the maximum allowable charging current in real time. It also monitors the difference between the maximum voltage and the charging cutoff voltage of each cell, the highest temperature of each cell, and the SOC and SOH of each cell. When any cell reaches the charging cutoff voltage, the second BMS notifies the charging equipment to reduce the charging current. When any cell reaches the maximum temperature limit, the second BMS notifies the charging equipment to stop charging. When all cells have reached the charging cutoff voltage and the charging current has dropped below the cutoff current, the second BMS determines that charging is complete, notifies the charging equipment to stop charging, and updates the SOC of the current high-voltage battery pack to 100%. When the fully charged high-voltage battery pack is reconnected to the power distribution bus, steps S2 and S3 are repeated.

[0017] The parallel battery control system and method for unmanned aerial vehicles (UAVs) with redundancy and inter-packet equalization provided by this invention have the following advantages compared to the prior art: 1. By connecting to an independent power distribution bus in parallel, the number of high-voltage battery packs used by the UAV can be configured according to mission requirements. In the event of a high-voltage battery pack failure, the power capacity of the remaining high-voltage battery packs can be matched with the mission power requirements to determine the flight mission content, such as continuing the mission, returning to base after disconnecting the faulty high-voltage battery pack, flying to the landing point, or prohibiting disconnection operations. This achieves linkage fault tolerance and redundancy between flight control and faults, avoiding in-flight engine failure.

[0018] 2. By configuring a first CAN communication bus and a redundant second CAN communication bus, in conjunction with a three-level BMS, cross-level redundant interaction capability is achieved; when the first BMS and its first CAN communication bus fail, the second CAN communication bus can be started in time, and the UAV flight control system can still adjust the flight strategy or return safely based on the remaining battery status.

[0019] 3. After going offline, the high-voltage battery pack undergoes open-loop correction via a second BMS using a fusion of ampere-hour integration and open-circuit voltage (OCV) method. Upon power-up, it distinguishes between abnormal self-discharge and overestimated State of Health (SOH) estimates, making timely corrections. Through real-time power capacity lookup and dynamic voltage correction, along with a tiered power cut-off decision mechanism based on remaining power capacity, accurate assessment and safe cut-off control of power supply capacity under fault conditions are achieved without hardware constraints such as active DC / DC allocation. This avoids secondary accidents caused by blind cut-off and ensures flight safety.

[0020] 4. Active pack equalization and circulating current suppression: Each battery pack has a built-in bidirectional DC / DC converter connected to the equalization bus. Before being connected to the power bus, the voltage difference between the packs is reduced to within the safe threshold through active voltage compensation, effectively suppressing the impact of parallel circulating current from the source and ensuring system efficiency and cell life. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a schematic diagram of the overall architecture of the parallel battery control system and method for unmanned aerial vehicles with redundancy and packet balance of the present invention. Figure 2 This is a block diagram showing the internal structure of a single high-voltage battery pack in the parallel battery control system and method for unmanned aerial vehicles with redundancy and inter-packet balance of the present invention. Figure 3 This is a schematic diagram of the internal circuit topology of the BDU of the UAV parallel battery control system and method with redundancy and inter-packet balance of the present invention. Figure 4 This invention provides a flowchart of multiple high-voltage battery packs connected to the power distribution bus and the inter-packet balancing process in a parallel battery control system and method for unmanned aerial vehicles with redundancy and inter-packet balancing. Figure 5 This is a flowchart illustrating the power-on connection process of the parallel battery control system and method for unmanned aerial vehicles with redundancy and balanced cell spacing according to the present invention. Figure 6 This is a flowchart illustrating the hierarchical reseverability judgment and decision-making process for the parallel battery control system and method for unmanned aerial vehicles (UAVs) with redundancy and inter-packet balance, as described in this invention. Detailed Implementation

[0023] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0024] In existing parallel power supply architectures for high-voltage battery packs in drones, the high-voltage series centralized power supply architecture has the risk of single-point failure at the topology level. A single cell failure can lead to the loss of power for the entire drone, making it difficult to meet the power redundancy and safety requirements of heavy-duty drones. The multi-pack parallel high-voltage system lacks a multi-level BMS management architecture, and there is a single-point failure risk at the control level, which means that when a single control node fails, each battery pack cannot achieve autonomous fault isolation. Existing battery packs cannot work independently after being separated from the drone, making it difficult to achieve full life cycle management and offline status monitoring of the battery. In high-voltage multi-pack parallel scenarios, there is a lack of inter-pack balancing management methods, and direct parallel connection of each battery pack can easily generate uncontrollable circulating currents, affecting system efficiency and cell life.

[0025] In view of this, such as Figure 1 As shown, on one hand, the present invention provides a parallel battery control system for unmanned aerial vehicles (UAVs) with redundancy and inter-packet equalization, comprising: Multiple high-voltage battery packs are installed in the drone's battery compartment via mechanical locking devices. Each high-voltage battery pack has a built-in second BMS. The output terminals of each high-voltage battery pack are connected to the drone's power distribution bus via power lines in parallel to provide power to the drone. Each high-voltage battery pack is also electrically connected to the balancing bus, which is set up independently of the power distribution bus and is used to carry out energy transfer during the balancing process between the battery packs. The first BMS communicates with the UAV flight control system and the second BMS in the multiple high-voltage battery packs via the first CAN communication bus. The first BMS is also electrically connected to the multiple high-voltage battery packs one by one via the high-voltage interlock detection line to detect the physical connection status of each high-voltage battery pack. Before the start of the UAV flight mission, the number of high-voltage battery packs that need to be connected to the power distribution bus is determined based on the mission load and expected flight range. The high-voltage battery packs are electrically connected to the power distribution bus and the equalization bus respectively. The high-voltage battery packs are also directly connected to the UAV flight control system through the second CAN communication bus. The first BMS is also electrically connected to the high-voltage battery packs through the high-voltage interlock line.

[0026] Figure 1 The illustrated UAV parallel battery control system structure with redundancy and inter-packet balancing includes a first BMS (Level 1 BMS in the diagram) and N high-voltage battery packs, where N is an integer greater than or equal to 2. The first BMS is installed inside the UAV fuselage and is powered by the UAV system's low-voltage auxiliary power supply. The first BMS connects upwards to the UAV flight control system and power distribution unit via power lines and CAN communication lines, and downwards to communicate with each high-voltage battery pack via the CAN communication bus. It also detects the connection status of each high-voltage battery pack via a high-voltage interlock detection line. The first BMS is used to summarize the status information of each high-voltage battery pack and report it to the flight control system, as well as to send control information to each high-voltage battery pack according to the flight control system's scheduling instructions.

[0027] like Figure 2As shown, multiple high-voltage battery packs each include a second BMS, several battery groups, a high-voltage circuit breaker (BDU), a smart junction box (BJB), several third BMSs, and a bidirectional DC / DC converter. The battery groups are sequentially and electrically connected. The negative terminal of the first battery group is electrically connected to the positive terminal of the adjacent battery group. The positive terminal of the first battery group and the negative terminal of the last battery group are both electrically connected to the high-voltage circuit breaker (BDU). The smart junction box (BJB) is located in the loop between the negative terminal of the last battery group and the high-voltage circuit breaker (BDU), and is electrically connected to the last battery group, the high-voltage circuit breaker (BDU), and the second BMS. It is used to acquire shunt sampling signals and temperature sensing signals, and send them to the second BMS. The second BMS then processes the shunt sampling signals and temperature sensing signals... The system obtains the real-time charging and discharging current of the high-voltage battery pack after temperature drift correction, and further combines it with time integration to obtain the state of charge (SOC) of the high-voltage battery pack. The accuracy of the SOC estimation is improved by a fusion algorithm of ampere-hour integration and open-circuit voltage correction. A bidirectional DC / DC converter is electrically connected to the positive terminal of the first battery pack, the negative terminal of the last battery pack, the second BMS, and the balancing bus, respectively. The bidirectional DC / DC converter receives control signals from the second BMS to realize the inter-pack balancing process. Several third BMSs are electrically connected to the two ends of several battery packs, one-to-one. The third BMS is equipped with a passive balancing circuit and is used to obtain the cell voltage and battery pack temperature data of each battery pack. The ampere-hour integration method and the open-circuit voltage method are both conventional techniques in this field and will not be elaborated further here.

[0028] The first and second BMS form a master-slave multi-machine BMS management architecture. The second BMS is directly connected to the first BMS, establishing a first CAN communication bus. Several third BMSs are connected to each other, and to the second BMS, via a daisy-chain communication system. Several third BMSs also communicate directly with the UAV flight control system via the second CAN communication bus. When the first BMS is working normally, it communicates with the second BMS via the first CAN communication bus, sending polling commands to the second BMS. The second BMS obtains the voltage and temperature data of the battery cells of several battery packs collected by each third BMS through polling, and then returns the obtained voltage and temperature data to the first BMS via the first CAN communication bus. When a timeout is detected in the first BMS's polling command communication, the system automatically switches to the second CAN communication bus, and each third BMS directly sends the voltage and temperature data of the battery cells of several battery packs to the UAV flight control system.

[0029] The second BMS is the master control BMU, serving as the main control unit of the high-voltage battery pack. It includes at least the following power supply modules: 1. A CAN communication interface, comprising at least two independent CAN controllers. One controller connects to the first BMS, and the other automatically switches to allow direct communication between the second BMS and the flight control system in case of a communication timeout. The other two CAN controllers can also be used for cross-packet interaction with other high-voltage battery packs' second BMSs. 2. A relay control module, including two H-bridge drive circuits connected to the coils of the main relay K1 and the pre-charge relay K2, respectively. The second BMS applies positive or negative pulses to the main relay K1 via the H-bridge drive circuits to control its opening and closing. The pre-charge relay K2 is controlled by a high-side switch, requiring only unidirectional drive. 3. A high-voltage detection signal receiving module, acquiring the battery pack's total voltage, bus voltage, and high-voltage interlock status information through an isolated communication interface. 4. A current detection module, performing high-precision sampling of the shunt voltage using a Delta-Sigma ADC, simultaneously reading shunt temperature sensor data for temperature drift compensation, and calculating the real-time charging and discharging current. A bidirectional DC / DC control interface is used to send start / stop commands, operating modes, target voltages, or target currents to the bidirectional DC / DC converter of the high-voltage battery pack, and to receive feedback information from the bidirectional DC / DC converter. The second BMS also has a built-in DC power module that converts the input from the battery pack into 5V, 12V, or 24V DC voltage to power the various modules within the second BMS.

[0030] The second BMS is electrically connected to the high-voltage circuit breaker BDU via a signal cable to control the main relay K1 and the pre-charge relay K2, as well as to acquire high-voltage detection information and insulation detection information; the second BMS is electrically connected to the intelligent junction box BJB via a shielded sampling line for acquiring charging and discharging current detection information.

[0031] The internal structure of a high-voltage circuit breaker (BDU) is as follows: Figure 3 As shown, the main relay K1 is connected in series in the positive charging and discharging circuit of the battery pack to control the power output of the current high-voltage battery pack. The main relay K1 is a magnetic latching relay, where a permanent magnet holds the contact state internally, requiring the coil to be energized only at the moment of activation. When the main relay K1 is closed, even if the control power supply is abnormally disconnected, the contacts remain closed, preventing accidental power outages due to abnormal power supply from the second BMS, thus meeting the fault holding requirements of the UAV system for battery pack power supply. When disconnection is required, the second BMS applies a reverse pulse to the coil through the relay drive circuit, causing the contacts to switch to the open state and be held by the permanent magnet. Furthermore, the high-voltage circuit breaker (BDU) has a reserved external direct drive interface, allowing the first BMS or flight control system to directly control the main relay K1 to disconnect in the event of a second BMS failure, ensuring the final execution path for fault isolation.

[0032] The branch formed by the series connection of pre-charge relay K2 and pre-charge resistor R1 is connected in parallel across the main relay K1. Pre-charge relay K2 is a normally open relay, closing only briefly when the high-voltage battery pack is connected to the parallel bus. The resistance value of pre-charge resistor R1 is selected based on the bus support capacitor capacity and the maximum allowable pre-charge current. The pre-charge circuit is activated before the main relay K1 when the high-voltage battery pack is connected. The battery pack charges the support capacitor on the parallel bus through the current-limiting effect of pre-charge resistor R1. Once the voltage of the support capacitor approaches the battery pack voltage, the main relay K1 closes, and then the pre-charge relay K2 opens, completing a smooth connection.

[0033] The high-voltage circuit breaker (BDU) also includes multiple high-voltage detection circuits and insulation detection circuits. The high-voltage detection circuit includes a voltage divider sampling resistor network and an isolation operational amplifier. The input terminals of the voltage divider sampling resistor network are connected to the positive and negative terminals of the battery pack, the output terminal of the main relay K1, and the high-voltage interlock interface, respectively. The input terminal of the voltage divider sampling resistor network is electrically connected to the input terminal of the isolation operational amplifier, and the output terminal of the isolation operational amplifier is electrically connected to the second BMS. The high-voltage detection circuit also has a relay contact status detection function. By judging the voltage at the input and output terminals of the main relay K1, it determines whether the main relay is normally closed or open, thus detecting main relay contact sticking faults. The insulation detection circuit measures the insulation resistance between the positive and negative terminals of the battery pack and the UAV casing in real time. When the insulation resistance drops to a preset safety threshold, such as below 100kΩ / V, the second BMS triggers an insulation fault alarm and automatically disconnects the main relay K1 as needed.

[0034] The intelligent junction box (BJB) is connected in series in the charging and discharging circuit on the negative side of the battery pack. It has a high-precision shunt inside, and the shunt sampling line is connected to the current detection module of the second BMS. At the same time, an NTC temperature sensor is attached to the shunt to obtain the temperature sampling signal of the voltage sampling signal and send it to the second BMS.

[0035] The third BMS draws power directly from both ends of the battery pack. Taking an 18-cell battery pack as an example, the battery pack voltage is approximately 48.6V to 75.6V. The third BMS internally uses a linear regulator to convert the battery pack voltage to a 3.3V or 5V operating voltage. Since each third BMS draws power independently from its corresponding battery pack, voltage differences between battery packs do not affect the normal operation of each third BMS.

[0036] Each high-voltage battery pack provides energy for the entire aircraft's power system. These battery packs are connected in parallel to the UAV's power distribution bus via power lines, to the balancing bus via balancing lines, and to the first BMS via CAN communication lines and high-voltage interlock lines. They are also directly connected to the flight control system via a backup CAN communication line, forming a multi-pack parallel high-voltage power supply architecture. The balancing bus is independent of the power bus and is used to handle energy transfer during the inter-pack balancing process, preventing the balancing current from affecting the power circuit. Before the flight mission begins, operators determine the number of high-voltage battery packs to be connected based on the mission payload weight and expected range. Each high-voltage battery pack is installed in the UAV's battery compartment using a mechanical locking mechanism. Its power output interface is connected to the UAV's power distribution bus, and its balancing output interface is connected to the balancing bus.

[0037] The battery pack consists of several lithium-ion cells connected in series. In this embodiment, 84 ternary lithium cells are used, with a nominal voltage of 310.8V after series connection, and a voltage platform in the 400V range. The battery pack is connected to the high-voltage circuit breaker (BDU) via positive and negative busbars, and the positive and negative leads of the cell pack are connected to the terminals of the bidirectional DC / DC converter. The third BMS is a slave controller CSU. In this embodiment, five third BMSs are configured. The first four third BMSs manage 18 cells per series, and the last third BMS manages 12 cells per series. Each third BMS is connected to the tab of the cell it manages via a voltage sampling line to collect the voltage of each cell in real time with a sampling accuracy of ±5mV. The temperature of key locations in the battery pack is collected via an NTC thermistor. Eight temperature sampling points are set for each battery pack, evenly distributed inside the battery pack.

[0038] Each third BMS is connected to the others, and the third BMS is connected to the second BMS, using a daisy-chain communication method. The second BMS polls each third BMS via a differential communication bus to obtain voltage and temperature data for all cells. The daisy-chain communication method significantly reduces the number of wiring harnesses within the package, while also providing transformer isolation and high safety.

[0039] When the voltage difference between different battery packs exceeds the preset equalization threshold, the second BMS or the UAV flight control system sends an equalization command to the third BMS, closes the passive equalization circuit corresponding to the battery pack with the highest voltage, and discharges the battery pack through the equalization resistor until the voltage difference between the battery packs does not exceed the preset equalization threshold, such as 20mV-100mV.

[0040] Each third BMS also integrates a passive balancing circuit. Each cell corresponds to a balancing switch and a balancing resistor, which are connected in series. When the second-level BMS detects that the voltage difference between the cells in the pack exceeds a preset balancing threshold, it sends a balancing command to the corresponding third BMS. The third BMS closes the balancing switch of the cell with the highest voltage and discharges the cell through the balancing resistor (33Ω in this embodiment) with a discharge current of approximately 100mA until the voltage difference is reduced to within the balancing threshold.

[0041] Both the second and third BMS are equipped with independent CAN controllers. The second BMS also has a built-in offline self-sustaining joint correction module and a real-time power capacity estimation module. After the high-voltage battery pack is removed from the drone, if the removed high-voltage battery pack still has residual power, the second BMS uses its built-in offline self-sustaining joint correction module to perform offline self-sustaining SOC and SOH joint correction on the removed high-voltage battery pack. Upon the next power-on, it combines the offline voltage drop and the charging voltage rise rate to distinguish between self-discharge abnormalities and SOH abnormalities. The system estimates the offset and performs multi-cycle sliding calibration. The second or third BMS of the detached high-voltage battery pack sends the status information of the battery pack or battery group to the outside world through the CAN controller. The status information includes the voltage, temperature, SOC, SOH, cumulative cycle count, and historical fault records of the high-voltage battery pack or battery group. The real-time power capability estimation module outputs power capability values ​​at different time scales based on the preset SOC-temperature-SOH multi-dimensional power table and combined with real-time voltage change rate dynamic correction. The first BMS determines whether the power capability value meets the requirements of the flight mission based on the power capability values ​​reported by multiple high-voltage battery packs, and performs graded resusceptibility judgment and decision when a high-voltage battery pack fails.

[0042] In the above scheme, the second BMS uses a built-in offline self-sustaining joint correction module to perform offline self-sustaining SOC and SOH joint correction on the detached high-voltage battery pack. Upon the next power-on, it integrates the offline voltage drop and charging voltage rise rates to distinguish between self-discharge anomalies and SOH estimation offsets, and performs multi-cycle sliding calibration, including the following: Autonomous wake-up and OCV correction in offline mode: When the high-voltage battery pack is detached from the drone and not connected to the charging equipment, the second BMS enters sleep mode while maintaining RTC timing. The second BMS automatically wakes up once every fixed time interval, which is 1-4 hours, and performs the following operations: The voltage of each cell in the battery pack is collected by the third BMS, and the voltage value, voltage extreme value and average value of each cell are recorded; if the time since the last time the battery pack was disconnected from the high-voltage bus or charging equipment is greater than a preset time, it is considered that the cell has been sufficiently rested and meets the open-circuit OCV correction condition. Then, based on the currently measured cell voltage, the corresponding offline OCV-SOC mapping value is obtained by looking up the table. An open-loop correction is performed on the current SOC of each cell. This is done by subtracting the SOC of each cell recorded before the last sleep cycle from the offline OCV-SOC mapping value, and marking it as... The SOC correction magnitude is obtained. The timestamp of the current open-loop correction, the SOC before the open-loop correction, the SOC after the open-loop correction, the SOC correction magnitude, and the voltage drop of each cell in the first hour after going offline are stored in the non-volatile memory of the second BMS to form the OCV correction history of the offline SOC. Joint calibration upon next power-on: When the high-voltage battery pack is reconnected to the drone, connected to the charging device, or actively woken up by the handheld terminal, the second BMS first reads the offline SOC OCV correction history stored during offline operation and executes the joint calibration logic: The second BMS extracts the cell voltage and SOC recorded during the last power-off before offline operation, and the cell voltage drop value obtained from the first OCV correction after offline operation. Given the SOC, the voltage rise curves and charging capacity of each cell during the most recent M complete charging processes, the current open-circuit voltage and current SOC of each cell measured at the current power-on time, calculate the SOC decay of each cell. The SOC decay of each cell is the cell SOC recorded at the last power-off before disconnection, marked as... Subtract the SOC of each cell at the current power-on moment, and mark it as... ; Calculate the expected decay of theoretical self-discharge The estimated discharge capacity of the battery cell is obtained by using the typical self-discharge rate and offline duration in the cell specifications. Divide by the rated capacity of the battery cell Multiply by 100% to obtain the absolute value of the result, then subtract the theoretical self-discharge expected degradation from the degradation amount of each cell's SOC. and preset capacity threshold Compare, if the absolute value ≤Preset capacity threshold This indicates that the battery cell is undergoing normal self-discharge. If the absolute value... Preset capacity threshold If the cell is found to be abnormal, attribution analysis and multi-cycle sliding calibration are performed. In this embodiment, a preset capacity threshold is used. Take 2% of the rated capacity of the battery cell.

[0043] Attribution analysis: Based on the voltage drop values ​​of each cell obtained from the first OCV correction after the second BMS integration offline and the voltage rise rate during charging, A) if the cell i The cell voltage drop value obtained after the first OCV correction after offline operation ≥Preset voltage threshold And the battery celli If the voltage rise rate during the most recent M complete charging processes deviates by ≤5% from the voltage rise rate of other cells, then the cell is considered... i Excessive self-discharge; mark the battery cell. i To maintain the SOH of each cell and the high-voltage battery pack constant, if the cell i If the self-discharge rate continues to exceed the limit, a maintenance alarm will be triggered; in this embodiment, a preset voltage threshold will be used. The value is 20mV, and M is an integer greater than 2; B) If the battery cell i The cell voltage drop value obtained after the first OCV correction after offline operation Preset voltage threshold But the battery cell i If the voltage rise rate deviation is >5% during the most recent M complete charging processes, the SOH estimate is considered too high, and the cell will be adjusted accordingly. i Marked as requiring capacity decay calibration, a multi-cycle sliding calibration process is performed. In the new round of SOH calculations for the high-voltage battery pack, the cells are given... i With greater weighting, the average of the weighted sums of the state of harmonics (SOH) of each cell is used as the current state of harmonics (SOH) of the high-voltage battery pack. If both A) and B) exist in the same high-voltage battery pack, the cells are marked and processed accordingly, and the minimum SOH of a single cell is taken as the SOH of the high-voltage battery pack. For high-voltage battery packs with excessively high State of Harm (SOH) estimates, multi-cycle sliding calibration is performed. This involves using the second Battery Management System (BMS) to record the charging capacity of the current charge-discharge cycle and the subsequent two complete charge-discharge cycles (a total of three cycles). Discharge capacity SOC correction range The voltage rise rate of each cell during charging is used to recalculate the actual usable capacity of the high-voltage battery pack using a weighted average method, and the SOH value is updated.

[0044] This also uses the cell voltage drop value obtained from the first OCV correction after offline operation. With preset voltage threshold The relationship between the voltage rise rate and the deviation of the voltage rise rate from that of other cells should theoretically result in four combinations. However, the cell voltage drop value obtained after the first OCV correction after offline testing... Preset voltage threshold And the battery cell i If the voltage rise rate during the most recent M full charge cycles deviates by ≤5% from the voltage rise rate of other cells, this is considered normal and requires no marking or SOH adjustment. The cell voltage drop value obtained after the first OCV correction after offline charging... ≥Preset voltage threshold And the battery cell i The voltage rise rate deviation is >5% during the most recent M complete charging processes, which falls under the worst combination of both A) and B). The handling measures have been mentioned above, so it is not expanded into four scenarios.

[0045] Actual usable capacity of high-voltage battery pack , For the first J The actual capacity is obtained by converting the average charging and discharging capacity or the coulombic efficiency over each cycle. As a weighted term, The weights of the three most recent complete charge-discharge cycles The values ​​are 0.2, 0.3, and 0.5 respectively. Update the SOH value to... This sliding calibration strategy avoids interference from single anomalous data and improves the stability and reliability of SOH estimation.

[0046] The offline self-sustaining function specifically supports the following application scenarios: Warehouse management: During the storage of high-voltage battery packs in the warehouse, the second BMS is periodically woken up, such as once every 24 hours, to collect cell voltage and temperature, update the self-discharge rate estimate, record abnormal events, and ensure the traceability of battery health status during storage.

[0047] Transportation monitoring: During the transportation of high-voltage battery packs, the second BMS monitors vibration and impact events by configuring acceleration sensors, and records the timestamps and amplitudes of out-of-limit events, providing data support for transportation safety management.

[0048] Battery swapping wait: After the high-voltage battery pack is replaced from the drone, the second BMS continues to work. Ground maintenance personnel can read the SOC, SOH and fault status of the battery pack through a handheld terminal to determine whether charging or repair is needed, and optimize battery swapping scheduling decisions.

[0049] Charging negotiation: When the high-voltage battery pack is connected to the charging equipment, the second BMS calculates the maximum allowable charging current in real time based on the voltage and temperature information of each cell collected by the third BMS and the preset charging strategy, and sends it to the charging equipment through the CAN bus to achieve safe fast charging control.

[0050] In one embodiment, the real-time power capability estimation module outputs power capability values ​​at different time scales based on a preset SOC-temperature-SOH multidimensional power table and dynamic correction using real-time voltage change rate. The first BMS determines whether the power capability values ​​reported by multiple high-voltage battery packs meet the requirements of the flight mission, including the following: Each high-voltage battery pack's second BMS has a pre-set multi-dimensional power capability table SOP. This SOP uses cell parameters provided by the cell manufacturer and is calibrated through actual power capability measurements after the high-voltage battery packs are assembled. This includes input dimensions such as temperature (e.g., 0℃, 25℃, 45℃), and the high-voltage battery pack's SOC and SOH, in 5% increments. The output is the maximum stable power output of the high-voltage battery pack over continuous timescales of 10 seconds, 60 seconds, and 10 minutes. P 10s , P 60s , P 10min ; P 10s It provides peak power for 10 seconds, designed for high-power instantaneous applications such as takeoff and emergency obstacle avoidance. P 60s It provides short-duration power for 60 seconds, used for climbing and heavy-load maneuvers. P 10min Provides continuous power for 10 minutes, suitable for drone cruising and hovering.

[0051] During flight, the second BMS performs the following real-time SOP estimations at preset intervals: acquiring the SOC and SOH of the high-voltage battery pack, reading the current minimum cell voltage and maximum cell temperature from the third BMS; and obtaining the base power capability through linear interpolation in the multi-dimensional power capability table SOP. , , Introducing the rate of voltage change Real-time correction is performed on the instantaneous current to reflect the battery polarization efficiency and instantaneous power status; if the voltage change rate... A negative increase indicates a rapid voltage drop, suggesting that the high-voltage battery pack is nearing its limit and the estimated power capacity needs to be reduced. (Correction factor) , For calibration parameters, a value of 0.7V / s is used. When the drop rate exceeds... At that time, the power capability is linearly reduced to 70%. Simultaneously, the ratio of the current to the lookup table current is considered to avoid short-term overshoot. A derating factor is also introduced when the temperature approaches the upper limit. The final power capability of the high-voltage battery pack is obtained: , P 60s , P 10min Using a similar formula, the second BMS reports the current final power capability of this high-voltage battery pack to the first BMS via the CAN bus, and then the first BMS reports it to the UAV flight control system.

[0052] For the synthesis of the total power capacity of a system with multiple high-voltage battery packs in parallel, the first BMS polls the second BMS of all online high-voltage battery packs at a preset time period, such as 100ms, to obtain the final power capacity of each high-voltage battery pack in three levels. The total power capacity of the system is then obtained by summing these values. , , ; to increase the total power capacity of the system Compare the power requirements of the current flight status issued by the UAV flight control system to determine whether the current total power capacity of the system meets the flight mission requirements.

[0053] like Figure 6 As shown, when a high-voltage battery pack fails, a graded cutoff capability judgment and decision is executed. This is because any high-voltage battery pack cell overheating, insulation failure, or communication loss occurs. Before cutting off the faulty high-voltage battery pack from the distribution bus, the first BMS obtains the predicted power demand value of the current flight mission for a future period from the flight control system. It evaluates and compares the final power capability of the remaining high-voltage battery pack with the predicted power demand value of the current flight mission. Based on the comparison result, it executes instructions to safely remove and continue the mission, return to base after cutting off, immediately reduce altitude and fly to the landing point after cutting off, or prohibit cutting off. After the high-voltage battery pack is cut off, the first BMS updates the online list of high-voltage battery packs and reports the final power capability of the remaining high-voltage battery packs to the flight control system. The flight control system dynamically updates the flight boundary parameters, including the maximum climb rate, maximum flight speed, and return-to-base threshold battery capacity.

[0054] The first step is to calculate the remaining power capacity after disconnecting the faulty high-voltage battery pack: Let the high-voltage battery pack to be disconnected be... j , , , , High-voltage battery pack to be removed j The maximum stable output power over timescales of 10 seconds, 60 seconds, and 10 minutes. , , The remaining power capacity after disconnecting the faulty high-voltage battery pack; The second step is to obtain the current flight power requirements. The first BMS requests the current and short-term future power requirement forecasts from the flight control system, including continuous cruise power. Corresponding to 10-minute scale, short-term maneuver power Corresponding to a 60s timescale and peak power This corresponds to a 10-second timescale. If the flight control system cannot provide this information, the maximum power measured within the most recent 10 seconds will be used as a conservative estimate.

[0055] The third step is tiered assessment and decision-making. Based on the comparison between remaining power capacity and power demand, four scenarios are identified:

[0056] Case 1, when , ,and When all conditions are met, it is determined to be a safe disconnection. The first BMS immediately sends a disconnection command to the faulty battery pack, executes the disconnection command, updates the system's remaining power capability after disconnection, and notifies the UAV flight control system. Scenario 2, when ,but and This indicates that after disconnecting the faulty high-voltage battery pack, long-term cruise cannot be maintained, but return-to-home maneuvers can still be supported. At this point, the first BMS sends a return-to-home suggestion signal to the flight control system, indicating that cruise cannot be maintained after disconnecting the faulty high-voltage battery pack, but return-to-home can be supported. After confirmation, the flight control system disconnects the high-voltage battery pack and automatically plans the return-to-home path and reduces cruise power.

[0057] Scenario 3, when ,but At this point, it indicates that even short-term maneuvering power cannot be met after the faulty battery pack is cut off, and it can only support instantaneous peak power. The first BMS reports a severe power shortage and recommends that the flight controller perform an emergency return to home or land at the nearest airport. After confirmation, the flight controller cuts off the faulty high-voltage battery pack, and the drone immediately descends and flies towards the landing point.

[0058] Scenario 4, when Removing the faulty high-voltage battery pack would render any safe maneuvers impossible, therefore removal is prohibited. The first BMS will not report a removal request, but instead report a faulty but unremovable status. The flight control system will then execute an on-site emergency landing or an in-flight emergency stop based on the overall situation. Removal will only be permitted after the drone has landed. An in-flight emergency stop requires maintaining a safe altitude and ensuring there are no obstacles or bodies of water below the drone.

[0059] The fourth step is updating the power boundary and adjusting the flight strategy after the faulty high-voltage battery pack is disconnected. When the faulty high-voltage battery pack is disconnected, the first BMS immediately updates the online high-voltage battery pack list, recalculates the total power capacity of the remaining system, and reports it to the flight control system. The flight control system then dynamically limits the flight envelope based on this, including limiting the maximum rate of climb, limiting the maximum flight speed, and triggering the low battery return-to-home threshold in advance. The low battery return-to-home threshold requires recalculating the remaining available capacity according to the new power capacity.

[0060] On the other hand, such as Figure 4 As shown, the present invention also provides a parallel battery control method for UAVs with redundancy and inter-packet balancing, comprising the following steps: S1: Configure the above-mentioned parallel battery control system for drones with redundancy and packet balance; S2: After each high-voltage battery pack is mechanically installed and connected to the drone, the first BMS checks the connection status of each high-voltage battery pack interface via the high-voltage interlock detection line. Once the physical connection is confirmed to be reliable, the first BMS sends a wake-up command via the first CAN bus to wake up the second BMS within each high-voltage battery pack. Each second BMS performs a self-test. After passing the self-test, the first BMS polls each second BMS to obtain status information including the total voltage, individual cell voltage extremes, and cell temperature of each high-voltage battery pack and its battery group. If no fault condition prohibits connection to the high-voltage battery pack, the first BMS will connect the high-voltage battery pack... Marked as unavailable, only high-voltage battery packs that meet the conditions are included in the power-on sequence; if the maximum voltage difference between each high-voltage battery pack exceeds 10V, the equalization process is initiated. The first BMS sets the equalization target voltage based on the average or median voltage of each high-voltage battery pack. Each second BMS controls the bidirectional DC / DC converter to draw power from or release energy from the equalization bus based on the deviation between the voltage of its high-voltage battery pack and the equalization target voltage, until the voltage difference between each high-voltage battery pack does not exceed the neighborhood range of the equalization target voltage, thus completing the inter-pack equalization. Each second BMS then stops the operation of the bidirectional DC / DC converter. S3: As Figure 5 As shown, after receiving the power-on command from the flight control system, the first BMS sequentially sends access commands to the second BMS of each high-voltage battery pack to be connected. The second BMS of the high-voltage battery pack closes the pre-charge relay and detects the supporting capacitor on the power distribution bus for pre-charging. The voltage rise of the supporting capacitor is detected with a detection period of 10ms. When the voltage of the high-voltage battery pack reaches 95%, the second BMS determines that the pre-charging is complete, then closes the main relay and opens the pre-charge relay, completing the smooth connection of the high-voltage battery pack to the power distribution bus. If the voltage of the supporting capacitor does not reach 95% of the high-voltage battery pack voltage within 5 seconds during the pre-charging process, the second BMS determines that there is a short circuit or a pre-charging circuit fault, opens the pre-charge relay, and reports a pre-charging failure fault. The time interval between the connection of two adjacent high-voltage battery packs to the power distribution bus is not less than 500ms, until all high-voltage battery packs are connected to the power distribution bus. S4: After all high-voltage battery packs are connected to the grid, they enter normal operation. The first BMS polls the second BMS of each high-voltage battery pack according to a preset cycle to obtain real-time operating data of each high-voltage battery pack, including total voltage, voltage of each individual cell, charging and discharging current, cell temperature, and cell SOC and SOH status information. The first BMS periodically reports the obtained status information to the UAV flight control system. Inside the high-voltage battery pack, each third BMS collects the voltage and temperature of each cell and reports it to the second BMS through a daisy chain. The second BMS activates passive equalization within the pack to eliminate voltage differences between different battery packs based on the degree of voltage consistency of the cells in different battery packs, until the voltage of the battery packs is consistent or the maximum equalization time is reached. S5: During system operation, the first, second, and third BMS continuously perform fault diagnosis on the high-voltage battery pack and system status. The fault diagnosis includes cell overvoltage, cell undervoltage, cell overheating, insulation fault, main circuit abnormality, and communication loss. Based on the severity of the fault diagnosis, a graded resusceptibility judgment and decision are executed. Before disconnecting the faulty high-voltage battery pack from the distribution bus, the first BMS obtains the predicted power demand value of the current flight mission for a future period from the flight control system, evaluates the final power capability of the remaining high-voltage battery pack and compares it with the predicted power demand value of the current flight mission, and executes instructions based on the comparison results: safe removal and continuation of mission, return to base after disconnection, immediately reduce altitude and fly to the landing point after disconnection, or prohibit disconnection. After the high-voltage battery pack is disconnected, the first BMS updates the online list of high-voltage battery packs and reports the final power capability of the remaining high-voltage battery packs to the flight control system. The flight control system dynamically updates the flight boundary parameters. The third BMS communicates with the UAV flight control system via the second CAN communication bus. When the first BMS fails, the second BMS of each high-voltage battery pack detects a timeout in communication with the first CAN communication bus of the first BMS, and if the second CAN communication bus does not receive a heartbeat packet sent by the first BMS, it determines that the first BMS has failed. The second BMS in each high-voltage battery pack continues to exchange status information with the first CAN communication bus through the CAN transceiver to determine the operating status of the high-voltage battery pack. At the same time, it enables the second CAN communication bus to directly report the status information of the high-voltage battery pack and the status of the first BMS to the UAV flight control system. The UAV flight control system determines the predicted power demand value based on the status information reported by each high-voltage battery pack and adjusts the flight strategy to ensure the safe return of the UAV.

[0061] When the high-voltage battery pack is detached from the drone for charging, the second BMS communicates with the charging equipment via the CAN communication interface. The second BMS enters the charging management mode, combining the temperature and voltage of each cell in the battery pack collected by the third BMS with the preset charging strategy CC-CV charging curve. This is a conventional constant current charging and constant voltage charging strategy, that is, first charging with constant current until the cell voltage reaches the charging cutoff voltage, and then gradually reducing the charging current until the charging current drops to the cutoff current or reaches the maximum charging time; the maximum allowable charging current is calculated in real time, and the maximum voltage of each cell and the charging cutoff voltage are monitored. The second BMS measures the difference in charging cutoff voltage, the maximum temperature of each cell, the SOC of each cell, and the SOH of the cell. When any cell reaches the charging cutoff voltage, the second BMS notifies the charging equipment to reduce the charging current. When any cell reaches the maximum temperature limit, the second BMS notifies the charging equipment to stop charging. When all cells have reached the charging cutoff voltage and the charging current has dropped below the cutoff current, the second BMS determines that charging is complete, notifies the charging equipment to stop charging, and updates the SOC of the current high-voltage battery pack to 100%. When the fully charged high-voltage battery pack is reconnected to the distribution bus, steps S2 and S3 are repeated.

[0062] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A parallel battery control system for unmanned aerial vehicles (UAVs) with redundancy and compartment balancing, characterized in that, include: Multiple high-voltage battery packs are installed in the drone's battery compartment via mechanical locking devices. Each high-voltage battery pack has a built-in second BMS. The output terminals of each high-voltage battery pack are connected to the drone's power distribution bus via power lines in parallel to provide power to the drone. Each high-voltage battery pack is also electrically connected to the balancing bus, which is set up independently of the power distribution bus and is used to carry out energy transfer during the balancing process between the battery packs. The first BMS communicates with the UAV flight control system and the second BMS in the multiple high-voltage battery packs via the first CAN communication bus. The first BMS is also electrically connected to the multiple high-voltage battery packs one by one via the high-voltage interlock detection line to detect the physical connection status of each high-voltage battery pack. Before the start of the UAV flight mission, the number of high-voltage battery packs that need to be connected to the power distribution bus is determined based on the mission load and expected flight range. The high-voltage battery packs are electrically connected to the power distribution bus and the equalization bus respectively. The high-voltage battery packs are also directly connected to the UAV flight control system through the second CAN communication bus.

2. The UAV parallel battery control system with redundancy and inter-packet balancing as described in claim 1, characterized in that, The multiple high-voltage battery packs also include several battery groups, high-voltage circuit breakers (BDUs), intelligent junction boxes (BJBs), several third-level battery management systems (BMSs), and bidirectional DC / DC converters. The battery groups are sequentially and electrically connected. The negative terminal of the first battery group is electrically connected to the positive terminal of the adjacent battery group. The positive terminal of the first battery group and the negative terminal of the last battery group are both electrically connected to the high-voltage circuit breakers (BDUs). The intelligent junction boxes (BJBs) are located in the circuit between the negative terminal of the last battery group and the high-voltage circuit breaker (BDU), and are electrically connected to the last battery group, the high-voltage circuit breaker (BDU), and the second BMSs. They are used to acquire shunt sampling signals and temperature sensing signals, and send them to the second BMSs. The second BMSs obtain the temperature based on the shunt sampling signals and temperature sensing signals. The real-time charging and discharging current of the high-voltage battery pack after drift correction is further combined with time integration to obtain the state of charge (SOC) of the high-voltage battery pack. The SOC estimation accuracy of the high-voltage battery pack is improved by a fusion algorithm of ampere-hour integration and open-circuit voltage correction. A bidirectional DC / DC converter is electrically connected to the positive terminal of the first battery pack, the negative terminal of the last battery pack, the second BMS, and the balancing bus, respectively. The bidirectional DC / DC converter receives the control signal input from the second BMS to realize the inter-pack balancing process between battery packs. Several third BMSs are electrically connected to the two ends of several battery packs one-to-one. The third BMS is equipped with a passive balancing circuit and is used to obtain the voltage of the cells and the temperature data of the battery pack.

3. A parallel battery control system for unmanned aerial vehicles with redundancy and inter-packet equalization as described in claim 2, characterized in that, The second BMS communicates directly with the first BMS, forming a first CAN communication bus. Several third BMSs communicate with each other, and with the second BMS, via a daisy-chain connection. Several third BMSs also communicate directly with the UAV flight control system via the second CAN communication bus. When the first BMS is working normally, it communicates with the second BMS via the first CAN communication bus, sending polling commands to the second BMS. The second BMS obtains the voltage and temperature data of the battery cells from several battery packs collected by each third BMS through polling, and then returns the obtained data to the first BMS via the first CAN communication bus. When a timeout is detected in the first BMS's polling command communication, the system automatically switches to the second CAN communication bus, allowing each third BMS to directly send the voltage and temperature data of the battery cells from several battery packs to the UAV flight control system. When the voltage difference between different battery packs exceeds a preset threshold, the second BMS or the UAV flight control system sends an equalization command to the third BMS, closes the passive equalization circuit corresponding to the battery pack with the highest voltage, and discharges the battery pack through the equalization resistor until the voltage difference between the battery packs does not exceed the preset threshold.

4. A parallel battery control system for unmanned aerial vehicles with redundancy and inter-packet equalization as described in claim 2, characterized in that, Both the second and third BMS are equipped with independent CAN controllers. The second BMS also has a built-in offline self-sustaining joint correction module and a real-time power capacity estimation module. After the high-voltage battery pack is removed from the drone, if the removed high-voltage battery pack still has residual power, the second BMS uses its built-in offline self-sustaining joint correction module to perform offline self-sustaining SOC and SOH joint correction on the removed high-voltage battery pack. Upon the next power-on, it combines the offline voltage drop and the charging voltage rise rate to distinguish between self-discharge abnormalities and SOH abnormalities. The system estimates the offset and performs multi-cycle sliding calibration. The second or third BMS of the detached high-voltage battery pack sends the status information of the battery pack or battery group to the outside world through the CAN controller. The status information includes the voltage, temperature, SOC, SOH, cumulative cycle count, and historical fault records of the high-voltage battery pack or battery group. The real-time power capability estimation module outputs power capability values ​​at different time scales based on the preset SOC-temperature-SOH multi-dimensional power table and combined with real-time voltage change rate dynamic correction. The first BMS determines whether the power capability value meets the requirements of the flight mission based on the power capability values ​​reported by multiple high-voltage battery packs, and performs graded resusceptibility judgment and decision when a high-voltage battery pack fails.

5. A parallel battery control system for unmanned aerial vehicles with redundancy and inter-packet balancing as described in claim 4, characterized in that, The second BMS uses a built-in offline self-sustaining joint correction module to perform offline self-sustaining SOC and SOH joint correction on the detached high-voltage battery pack. Upon the next power-on, it integrates the offline voltage drop and charging voltage rise rate to distinguish between self-discharge anomalies and SOH estimation offsets, and performs multi-cycle sliding calibration, including the following: Autonomous wake-up and OCV correction in offline state: When the high-voltage battery pack is detached from the drone and not connected to the charging equipment, the second BMS enters sleep mode, but still maintains RTC timing. The second BMS automatically wakes up once every fixed time interval and performs the following operations: the voltage of each cell of all battery packs is collected through the third BMS, and the voltage value, voltage extreme value and average value of each cell are recorded; if the time since the last time the battery pack was powered off from the high-voltage bus or charging equipment is greater than the preset time, it is considered that the cell has been sufficiently rested and meets the open-circuit OCV correction conditions. Then, based on the current measured voltage of each cell, the corresponding offline OCV-SOC mapping value is obtained by looking up the table, and an open-loop correction is performed on the current SOC of each cell. The SOC correction magnitude is obtained by subtracting the SOC of each cell recorded before the last sleep from the offline OCV-SOC mapping value. The timestamp of the current open-loop correction, the SOC before the open-loop correction, the SOC after the open-loop correction, the SOC correction magnitude, and the voltage drop of each cell in the first hour after going offline are used to form the offline SOC OCV correction history record. Joint calibration upon next power-on: When the high-voltage battery pack is reconnected to the drone, connected to the charging equipment, or actively woken up by the handheld terminal, the second BMS first reads the offline SOC OCV correction history stored during offline operation and executes the joint calibration logic: The second BMS extracts the cell voltage and SOC recorded at the last power-off before offline operation, the cell voltage drop and SOC obtained from the first OCV correction after offline operation, the voltage rise curve and charging capacity of each cell during the most recent M complete charging processes, and the static open-circuit voltage and current SOC of each cell measured at the current power-on moment, calculates the SOC attenuation of each cell, and... The decay of cell SOC is calculated by subtracting the SOC of each cell at the current power-on moment from the SOC recorded at the last power-down before offline. The theoretical self-discharge expected decay is calculated by obtaining the estimated discharge capacity of the cell from the typical self-discharge rate and offline duration in the cell specification, dividing it by the rated capacity of the cell, and then multiplying by 100%. The absolute value of the result of subtracting the theoretical self-discharge expected decay from the decay of each cell SOC is compared with a preset capacity threshold. If the absolute value is less than or equal to the preset capacity threshold, the cell is considered to have normal self-discharge. If the absolute value is greater than the preset capacity threshold, the cell is judged to have an abnormality, and attribution analysis and multi-cycle sliding calibration are performed.

6. A parallel battery control system for unmanned aerial vehicles with redundancy and inter-packet equalization as described in claim 4, characterized in that, The real-time power capability estimation module outputs power capability values ​​at different time scales based on a preset SOC-temperature-SOH multidimensional power table and dynamic correction using real-time voltage change rate. The first BMS determines whether the power capability values ​​reported by multiple high-voltage battery packs meet the requirements of the flight mission, including the following: Each high-voltage battery pack's second BMS has a pre-set multi-dimensional power capability table SOP. Using cell parameters provided by the cell manufacturer and after the high-voltage battery packs are assembled, the system performs measured power capability calibration. This includes using temperature, the high-voltage battery pack's SOC, and SOH as input dimensions, and using the maximum stable power output of the high-voltage battery pack over 10 seconds, 60 seconds, and 10 minutes as the output. During flight, the second BMS performs the following real-time estimations at a preset cycle: acquiring the high-voltage battery pack's SOC and SOH; reading the current minimum cell voltage and maximum cell temperature from the third BMS; obtaining the basic power capability through linear interpolation in the multi-dimensional power capability table SOP; and incorporating voltage change rate and instantaneous current for real-time correction to obtain the final power capability of the high-voltage battery pack. The first BMS polls all online high-voltage battery packs' second BMSs at a preset cycle to obtain the final power capability of each high-voltage battery pack. These final power capabilities are then summed to obtain the total system power capability. The total system power capability is compared with the power requirements for the current flight state issued by the UAV flight control system to determine whether the current total system power capability meets the flight mission requirements.

7. A parallel battery control system for unmanned aerial vehicles with redundancy and inter-packet balancing as described in claim 6, characterized in that, The process of performing graded reseverability judgment and decision-making when a high-voltage battery pack fails involves the following steps: Before disconnecting any high-voltage battery pack from the distribution bus due to cell overheating, insulation failure, or communication loss, the first BMS obtains the predicted power demand for the current flight mission over a future period from the flight control system. It then compares the final power capability of the remaining high-voltage battery packs with the predicted power demand for the current flight mission. Based on the comparison result, it executes commands such as safe removal and continuation of the mission, return to base after disconnection, immediately descending altitude and flying towards the landing point after disconnection, or prohibiting disconnection. After the high-voltage battery pack is disconnected, the first BMS updates the online list of high-voltage battery packs and reports the final power capability of the remaining high-voltage battery packs to the flight control system. The flight control system then dynamically updates the flight boundary parameters.

8. A parallel battery control system for unmanned aerial vehicles with redundancy and inter-packet equalization as described in claim 7, characterized in that, The flight boundary parameters include maximum climb rate, maximum flight speed, and return-to-home threshold battery level.

9. A parallel battery control method for unmanned aerial vehicles (UAVs) with redundancy and inter-packet balancing, characterized in that, Includes the following steps: S1: Configure a UAV parallel battery control system with redundancy and packet balance as described in any one of claims 2-8; S2: After each high-voltage battery pack is mechanically installed and connected to the drone, the first BMS detects the connection status of each high-voltage battery pack interface through the high-voltage interlock detection line. After confirming that the physical connection status is reliable, the first BMS sends a wake-up command through the first CAN bus to wake up the second BMS in each high-voltage battery pack. Each second BMS performs a self-test. After the self-test is passed, the first BMS polls each second BMS to obtain the status information of each high-voltage battery pack. If a high-voltage battery pack has a fault state that prohibits access, the first BMS marks the high-voltage battery pack as unavailable and only includes high-voltage battery packs that meet the conditions in the power-on sequence. If the maximum voltage difference between each high-voltage battery pack exceeds 10V, the equalization process is entered. S3: After receiving the power-on command from the flight control system, the first BMS sends access commands to the second BMS of each high-voltage battery pack to be connected in sequence, so that the high-voltage battery packs are smoothly connected to the power distribution bus; if the time interval between the connection of two adjacent high-voltage battery packs to the power distribution bus is not less than 500ms, until all high-voltage battery packs are connected to the power distribution bus. S4: After all high-voltage battery packs are connected to the grid, they enter normal operation. The first BMS polls the second BMS of each high-voltage battery pack according to a preset cycle to obtain real-time operating data of each high-voltage battery pack, including total voltage, voltage of each individual cell, charging and discharging current, cell temperature, and cell SOC and SOH status information. The first BMS periodically reports the obtained status information to the UAV flight control system. Inside the high-voltage battery pack, each third BMS collects the voltage and temperature of each cell and reports it to the second BMS through a daisy chain. The second BMS activates passive equalization within the pack to eliminate voltage differences between different battery packs based on the degree of voltage consistency of the cells in different battery packs, until the voltage of the battery packs is consistent or the maximum equalization time is reached. S5: During system operation, the first BMS, second BMS, and third BMS continuously perform fault diagnosis on the high-voltage battery pack and system status. Based on the severity of the fault diagnosis, they perform graded resusceptibility judgment and decision-making. Before disconnecting the faulty high-voltage battery pack from the power distribution bus, the first BMS obtains the predicted power demand value of the current flight mission for a period of time from the flight control system, evaluates and compares the final power capacity of the remaining high-voltage battery pack with the predicted power demand value of the current flight mission, and executes instructions such as safe removal and continuation of mission, return to base after disconnection, immediately reduce altitude and fly to the landing point after disconnection, or prohibit disconnection based on the comparison results. The third BMS communicates with the UAV flight control system via the second CAN communication bus. When the first BMS fails, the second CAN communication bus is activated to directly report the status information of the high-voltage battery pack and the status of the first BMS to the UAV flight control system. The UAV flight control system determines the predicted power demand based on the status information reported by each high-voltage battery pack and adjusts the flight strategy to ensure the UAV returns safely.

10. A parallel battery control method for UAVs with redundancy and packet balance according to claim 9, characterized in that, When the high-voltage battery pack is disconnected from the drone for charging, the second BMS communicates with the charging equipment via the CAN communication interface. The second BMS enters the charging management mode and, in conjunction with the temperature and voltage of each cell in the battery pack collected by the third BMS and the preset charging strategy, calculates the maximum allowable charging current in real time. It also monitors the difference between the maximum voltage and the charging cut-off voltage of each cell, the highest temperature of each cell, and the SOC and SOH of each cell. When any cell reaches the charging cut-off voltage, the second BMS notifies the charging equipment to reduce the charging current. When any cell reaches the maximum temperature limit, the second BMS notifies the charging equipment to stop charging. When all cells have reached the charging cut-off voltage and the charging current has dropped below the cut-off current, the second BMS determines that charging is complete, notifies the charging equipment to stop charging, and updates the SOC of the current high-voltage battery pack to 100%. When the fully charged high-voltage battery pack is reconnected to the power distribution bus, steps S2 and S3 are repeated.

Images