Lithium battery BMS low-power-consumption device for electric bicycle, control method and vehicle

Abstract

The invention belongs to the technical field of power management systems, and particularly relates to a lithium battery BMS low-power-consumption device for an electric bicycle, a control method and a vehicle, the device comprises a power distribution network and a microcontroller, and the power distribution network comprises a DC-DC step-down circuit and a voltage stabilizer LDO; a state monitoring unit in the microcontroller obtains battery state information and vehicle use activity characteristics, a parameter generation unit generates dynamic switching parameters according to the battery state information and the vehicle use activity characteristics, and a mode switching unit controls the device to be switched among a working mode, a standby mode, a sleep mode and a deep break mode according to the dynamic switching parameters. And the enabling states of the DC-DC step-down circuit and the voltage stabilizer LDO are controlled. Through hierarchical control of a four-stage power consumption mode and dynamic switching parameter generation based on a multi-factor adaptive decision, step-by-step reduction of BMS static power consumption is realized, and the standby time of the electric bicycle in a static state and the service life of a battery are effectively prolonged.

Description

Low-power lithium battery BMS device, control method and vehicle for electric bicycles Technical Field

[0001] This invention belongs to the field of power management system technology, specifically relating to a low-power lithium battery BMS device, control method, and vehicle for electric bicycles. Background Technology

[0002] The safety, range, and lifespan of lithium-ion battery packs in electric bicycles have become core concerns for consumers. To ensure safe and stable battery operation, a Battery Management System (BMS) has become standard. However, unlike continuously moving electric vehicles, electric two-wheelers are characterized by frequent start-stop cycles and long periods of inactivity (such as at night or on weekdays), making the static power consumption of the BMS itself a particularly prominent issue of "invisible" battery drain. In existing technologies, many BMS designs primarily focus on active functions such as charge / discharge management, status monitoring, and safety protection, while neglecting optimization for power consumption control during long periods of inactivity. Some BMSs may continuously maintain the full functionality of the microcontroller (MCU), analog front-end (AFE) chip, communication modules (such as Bluetooth), and voltage sampling circuits, resulting in a static operating current maintained at the milliampere (mA) level. Patent CN121105902A provides a "Power Consumption Scheduling Control Method for Dynamically Adjusting RTC Wake-up of Truck Lithium Batteries," which achieves BMS sleep scheduling by judging the lithium battery status, configuring RTC interrupt wake-up, and setting wake-up time partitions.

[0003] However, existing technologies have the following drawbacks: First, the management dimension is singular, relying solely on the macroscopic parameter of the total battery pack voltage for power consumption decisions, failing to monitor the risks of over-discharge, voltage abnormalities, or imbalances in individual cells, thus posing safety risks; Second, the wake-up mechanism is rigid, mainly relying on fixed RTC timing and limited hardware signals, failing to meet the immediate interactive needs of modern smart electric two-wheeler users who can connect to an app via Bluetooth to view battery status, perform location checks, or set parameters in real time; Third, the power consumption grading is coarse, mainly focusing on two-level state management around "sleep" and "RTC wake-up cycle," without involving independent, tiered power supply control for different functional circuit modules, making it difficult to achieve multi-level power consumption transitions from hundreds of microamps to tens of microamps, and failing to maximize the extension of standby time. Summary of the Invention

[0004] The purpose of this invention is to provide a low-power lithium battery BMS device, control method, and vehicle for electric bicycles, in order to solve the technical problems of existing electric bicycle BMS having high static power consumption in a stationary state and difficulty in achieving a balance between extremely low power consumption and multi-source wake-up response.

[0005] The present invention achieves the above-mentioned objectives through the following technical solutions: Firstly, the present invention proposes a low-power lithium battery BMS device for electric bicycles, comprising a power distribution network and a microcontroller, wherein: the power distribution network includes a DC-DC buck circuit and a voltage regulator LDO, the enable terminals of the DC-DC buck circuit and the voltage regulator LDO being controlled by the microcontroller; the microcontroller includes: a status monitoring unit for acquiring battery status information and vehicle usage activity characteristics; a parameter generation unit for generating dynamic switching parameters based on the battery status information and the vehicle usage activity characteristics; and a mode switching unit for controlling the device to switch between operating mode, standby mode, sleep mode, and deep sleep mode based on the dynamic switching parameters, and controlling the enable state of the DC-DC buck circuit and the voltage regulator LDO; wherein the DC-DC buck circuit is used to supply power in operating mode and standby mode, and the voltage regulator LDO is used to supply power in sleep mode.

[0006] Furthermore, the status monitoring unit includes: a first monitoring subunit, used to divide a day into multiple time slices according to a preset time division rule, obtain the effective vehicle activity indications within multiple historical time slices, and update the usage probability of the current time slice based on an exponentially weighted moving average; a second monitoring subunit, used to obtain the battery's state of charge and ambient temperature, and determine a power stress factor and a temperature factor based on the state of charge and ambient temperature, respectively; a third monitoring subunit, used to obtain the statistical count of invalid wake-up events, and determine an invalid wake-up factor based on the comparison result of the statistical count of invalid wake-up events with a preset upper limit value; wherein an invalid wake-up event is an event in which the microcontroller, after being woken up, fails to detect effective charging / discharging current, fails to establish effective communication interaction, and fails to handle faults within a preset observation window.

[0007] Furthermore, the parameter generation unit is used to generate the dynamic switching parameters according to the usage probability, the power shortage factor, the temperature factor, and the invalid wake-up factor, in the following manner: the dynamic switching parameters include at least one of the following: the delay time T1 for entering standby mode, the real-time clock (RTC) wake-up interval T2 in sleep mode, and the undervoltage determination confirmation time T3. ; ; ;in, As a factor of power shortage, For temperature factor, Number of times the device cannot be woken up. The probability of use at the current moment. The preset base delay time, The preset base wake-up time, This is the preset undervoltage determination basis confirmation time.

[0008] Furthermore, the mode switching unit is also used to: control the device to enter a working mode when a charging / discharging current is detected or a valid interactive wake-up is achieved; in the working mode, the DC-DC step-down circuit is enabled, and the voltage regulator LDO is disabled; when there is no charging / discharging current and no fault occurs, control the device to enter a standby mode; in the standby mode, the DC-DC step-down circuit is enabled, and the power path of non-core modules is disconnected by controlling the load switch; when there is no charging / discharging current in the standby mode for more than a first preset time, control the device to enter a sleep mode; in the sleep mode, the DC-DC step-down circuit is disabled, the voltage regulator LDO is enabled, the microcontroller enters a deep sleep state and wakes up according to the RTC timer; when any single battery cell voltage is detected to be lower than the undervoltage threshold, control the device to enter a deep sleep mode; in the deep sleep mode, the DC-DC step-down circuit and the voltage regulator LDO are disabled, and the device is powered by the hardware monitoring circuit.

[0009] Furthermore, the step of controlling the device to switch between working mode, standby mode, hibernation mode, and deep sleep mode according to the dynamic switching parameters, and controlling the enable state of the DC-DC buck circuit and the voltage regulator LDO, includes: controlling the device to switch from working mode to standby mode according to the delay time T1 for entering standby mode in the dynamic switching parameters; controlling the microcontroller to wake up from hibernation mode at regular intervals according to the RTC timed wake-up interval T2 in hibernation mode according to the dynamic switching parameters; and controlling the device to enter deep sleep mode when the voltage of a single battery cell is lower than the undervoltage threshold according to the undervoltage determination confirmation time T3 in the dynamic switching parameters.

[0010] Furthermore, the usage probability is obtained through the following steps: dividing a day into several time slices, and defining a binary activity indicator for each time slice k. : The usage probability is updated using EWMA, as shown in the following formula: ;in, The probability of use at the current moment. The probability of use at the previous moment. This is the smoothing coefficient.

[0011] Furthermore, the energy stress factor and temperature factor are obtained through the following steps: the energy stress factor and temperature factor are obtained by monotonically normalizing SOC and temperature, as shown in the following formula: ; ;in, This represents the amplitude limiting function, limiting the amplitude to 0~1; The current state of charge, This is the normal reference boundary for the state of charge; This is the limit of the state of charge; The current temperature; This is the normal temperature reference boundary; This is the low-temperature boundary.

[0012] Furthermore, the invalid wake-up factor is obtained through the following steps: if the MCU is woken up and no effective charging / discharging current is detected, no effective communication interaction is established, and no fault handling occurs within a preset observation window, it is recorded as an invalid wake-up, and the count W is increased; the invalid wake-up factor is determined by the following formula. : ;in, This represents the current number of invalid wake-ups. This is the maximum number of invalid wake-ups; W is reset to zero in response to the detection of a valid charge / discharge event.

[0013] Secondly, the present invention proposes a low-power control method for a lithium battery BMS for electric bicycles, characterized in that, applied to the aforementioned device, the method includes: acquiring battery status information and vehicle usage activity characteristics; generating dynamic switching parameters based on the battery status information and the vehicle usage activity characteristics; controlling the device to switch between working mode, standby mode, hibernation mode and deep sleep mode based on the dynamic switching parameters, and controlling the enabling state of the high-efficiency DC-DC buck circuit and the low-power linear regulator LDO.

[0014] Thirdly, the present invention proposes a vehicle including the aforementioned lithium battery BMS low-power device for electric bicycles.

[0015] The beneficial effects of this invention are as follows: 1. This invention constructs a branch-controllable power distribution network consisting of a DC-DC step-down circuit and a low-power LDO, and combines it with a four-level power consumption mode switching mechanism to achieve a gradual reduction in the static power consumption of the BMS. In working mode and standby mode, DC-DC power supply ensures system response speed. In sleep mode, switching to LDO power supply reduces power consumption to 350μA. In deep sleep mode, cutting off both power supplies and retaining only the hardware monitoring circuit further reduces power consumption to below 30μA, effectively extending the standby time of the electric bicycle in a static state.

[0016] 2. This invention introduces an adaptive decision-making algorithm based on time-slice usage probability, power shortage factor, temperature factor, and invalid wake-up factor to dynamically generate mode switching delay time, RTC wake-up interval, and undervoltage determination confirmation time. The system can maintain rapid response during peak user periods based on user usage patterns, and enter a low-power state more quickly in power shortage or low-temperature environments. Furthermore, the invalid wake-up factor backoff mechanism suppresses false wake-ups caused by noise interference, minimizing unnecessary energy consumption while ensuring safety monitoring. Attached Figure Description

[0017] Figure 1 is a system block diagram of a lithium battery BMS low-power device in an embodiment of the present invention; Figure 2 is another system block diagram of a lithium battery BMS low-power device in an embodiment of the present invention; Figure 3 is a schematic diagram of a power tree with independent and controllable branch circuits in the present invention. Detailed Implementation

[0018] The present application will now be described in further detail with reference to the accompanying drawings. It should be noted that the following specific embodiments are only used to further illustrate the present application and should not be construed as limiting the scope of protection of the present application. Those skilled in the art can make some non-essential improvements and adjustments to the present application based on the above application content.

[0019] The following description, with reference to the accompanying drawings, describes a low-power lithium battery BMS device for electric bicycles according to an embodiment of the present invention.

[0020] Please refer to Figures 1 and 3 for a low-power lithium battery BMS device for electric bicycles, including a power distribution network and a microcontroller (MCU).

[0021] The power distribution network includes a DC-DC buck converter and an LDO regulator, with their enable terminals controlled by a microcontroller. The network employs a decentralized, independently controllable design architecture. The DC-DC buck converter boasts high efficiency and high current output capability, making it suitable for high-load system operation scenarios. The LDO features ultra-low quiescent power consumption, making it suitable for deep system sleep scenarios. By independently controlling the enable terminals of the two power supplies through the microcontroller, the design goal of on-demand power supply from the hardware level is achieved.

[0022] Referring to Figure 3, for example, the power distribution network uses the positive terminal B+ of the battery pack as the input terminal, and a first branch and a second branch are connected in parallel between B+ and ground to provide hierarchical power supply and switching for each functional module under different operating modes.

[0023] The first branch includes a DC-DC step-down circuit, whose enable terminal EN is connected to and controlled by the MCU. The output of the DC-DC circuit forms the system's operating power node and supplies power to each module via a switching circuit and an LDO regulator. Specifically, the switching circuit outputs SW_5V to power the communication circuit; simultaneously, the DC-DC output, after being regulated by the LDO, provides BT_3V3 to power the Bluetooth communication module. Furthermore, the first branch can also provide the necessary power input for operation to the temperature acquisition and communication circuit, as well as the MCU and peripheral circuits, via appropriate voltage regulation or distribution nodes.

[0024] The second branch includes a low-power LDO, whose enable terminal EN is also connected to and controlled by the MCU. The output of the low-power LDO forms a low-power power supply path. This output provides power to the MOS drive circuit on one hand, and after being regulated by the subsequent LDO, it becomes the SYS_3V3 power supply, which is used to provide the power required to maintain the operation of the MCU and peripheral circuits in sleep or low-power mode, and can also provide basic power supply for temperature acquisition and communication circuits in low-power mode.

[0025] A first isolation diode D1 and a second isolation diode D2 are installed between the two branches to isolate the DC-DC branch and the low-power LDO branch, suppress the mutual influence of voltage between the branches and prevent reverse current injection, thereby improving the stability and reliability of the power supply switching process.

[0026] With the above architecture, SW_5V is a controlled power supply controlled by the switching circuit; while whether BT_3V3, SYS_3V3, and other power supplies are supplied depends on the corresponding regulators and their enable states. Both can be controlled by the MCU through the DC-DC converter, low-power LDO, and switching circuit to achieve on-demand start / stop and power consumption management.

[0027] The microcontroller includes a state monitoring unit, a parameter generation unit, and a mode switching unit. The state monitoring unit acquires battery status information and vehicle usage activity characteristics. It collects real-time parameters such as battery voltage, current, and temperature, and monitors the vehicle's charging and discharging activities, communication interactions, and other behaviors. The parameter generation unit generates dynamic switching parameters based on the battery status information and vehicle usage activity characteristics. The mode switching unit controls the device to switch between operating mode, standby mode, hibernation mode, and deep sleep mode according to the dynamic switching parameters, and controls the enable status of the DC-DC buck circuit and the LDO regulator.

[0028] Specifically, the parameter generation unit is based on a lightweight adaptive decision-making algorithm, which transforms the collected multi-dimensional information into specific mode switching parameters, enabling the power management strategy to be dynamically adjusted according to the actual situation, thus avoiding the problem of insufficient adaptability caused by fixed threshold settings.

[0029] The DC-DC step-down circuit provides power in both operating and standby modes, while the LDO regulator powers the microcontroller's deep sleep state and wake-up detection circuit in sleep mode, maintaining basic timed wake-up and external wake-up functions. Through the division of labor between the two power supplies, the system utilizes DC-DC power to ensure fast response during high-performance demands and LDO power to ensure extremely low power consumption during deep sleep, achieving an optimal balance between performance and power consumption.

[0030] Optionally, in deep rest mode, the DC-DC buck circuit and the LDO regulator are disabled. In deep rest mode, the system cuts off power to almost all circuits, leaving only the hardware monitoring circuit operational, reducing the overall board power consumption to below 30uA, maximizing the preservation of residual battery power, and preventing irreversible damage due to over-discharge. Furthermore, deep rest mode ensures recoverability upon exiting deep rest mode through blind charging activation: when the charger is plugged in, the hardware monitoring circuit detects the charging voltage and generates a wake-up signal, re-powering the LDO or MCU, allowing the system to exit deep rest mode normally and resume operation.

[0031] In this disclosure, the DC-DC buck circuit can specifically be a high-efficiency synchronous buck converter used to step down the high voltage (e.g., 48V, 36V, or 24V) of the lithium battery pack to a stable low-voltage power supply (e.g., 3.3V or 5V), providing operating voltage for the MCU, Bluetooth module, temperature sensor, MOSFET driver circuit, etc. In the context of this invention, it specifically undertakes the following power supply tasks: providing power for the MCU to operate at full speed in both working and standby modes; maintaining broadcast or connection power for the Bluetooth module; providing operating power for the voltage / temperature sampling circuit; and providing driving capability for the MOSFET driver circuit, ensuring that the system has complete monitoring, communication, and control functions when activated.

[0032] In addition, the voltage regulator LDO can specifically be an ultra-low quiescent current linear regulator. This circuit also steps down the battery voltage to a stable low-voltage power supply, but its core feature is its extremely low quiescent power consumption (typically in the 1μA-5μA range). In the scenario of this invention, it specifically undertakes the following power supply tasks: powering the MCU's deep sleep state in sleep mode (only retaining RAM and RTC operation), powering the wake-up detection logic circuit (such as external interrupt detection and charging insertion detection), and powering the RTC real-time clock to maintain the timed wake-up function, ensuring that the system can still respond to external wake-up signals even at microampere-level power consumption.

[0033] Preferably, the status monitoring unit includes a first monitoring subunit, a second monitoring subunit, and a third monitoring subunit; wherein, the first monitoring subunit is used to divide a day into multiple time slices (e.g., by hour) according to a preset time division rule, obtain the effective vehicle activity indications in multiple historical time slices, and update the usage probability of the current time slice based on an exponentially weighted moving average; the second monitoring subunit is used to obtain the battery's state of charge and ambient temperature, and determine the power stress factor and temperature factor based on the state of charge and ambient temperature, respectively; the third monitoring subunit is used to obtain the statistical count of invalid wake-up events, and determine the invalid wake-up factor based on the comparison result of the statistical count of invalid wake-up events with a preset upper limit value; wherein an invalid wake-up event is an event in which the microcontroller is woken up and no effective charging / discharging current is detected within a preset observation window, no effective communication interaction is established, and no fault handling is performed.

[0034] Among them, the power stress factor reflects the degree of battery power shortage. When the SOC is low, the factor value increases, and the system will adopt a more aggressive power-saving strategy. The temperature factor reflects the impact of low temperature environment on battery performance. When the temperature is low, the battery's usable capacity decreases and its internal resistance increases, resulting in a higher factor value, and the system will also strengthen power consumption control. The invalid wake-up factor effectively reduces the additional energy consumption caused by noise disturbances without sacrificing fault priority handling. For example, the system may be falsely woken up due to environmental noise, electromagnetic interference, or brief contact. If each false wake-up causes the system to enter full-function state, it will cause unnecessary power waste.

[0035] In this disclosure, the first monitoring subunit can learn users' vehicle usage patterns at different times. An exponentially weighted moving average algorithm is employed, which considers the influence of historical behavior while also enabling rapid responses to new changes. For example, the probability of use naturally increases during weekday morning and evening rush hours, while decreasing during late-night hours. This time-slice-based usage probability learning mechanism allows the system to predict users' vehicle usage needs, maintain faster response times during peak hours, and more actively enter a low-power state during infrequently used periods.

[0036] Preferably, the parameter generation unit is used to generate dynamic switching parameters according to the usage probability, power shortage factor, temperature factor, and invalid wake-up factor, in the following manner: the dynamic switching parameters include at least one of the following: the delay time T1 for entering standby mode, the real-time clock (RTC) wake-up interval T2 in sleep mode, and the undervoltage determination confirmation time T3: ; ; ;in, As a factor of power shortage, For temperature factor, Number of times the device cannot be woken up. The probability of use at the current moment. The preset base delay time, The preset base wake-up time, This is the preset undervoltage determination basis confirmation time.

[0037] Specifically, , , These three custom parameters represent the system's time base under typical operating conditions or factory default settings. Their values ​​should be preset based on the physical characteristics of the battery pack and the target application scenario. The preset time is usually 3 to 5 minutes to balance temporary parking scenarios and the suppression of frequent mode switching; The typical range is 10 seconds to 10 minutes, and 30 seconds can be used in the exemplary implementation to balance the timeliness of monitoring and power consumption. The preset time is usually 5 to 10 seconds, used to filter out transient interference.

[0038] In addition, the calibration of the above-mentioned custom parameters is completed through the following process: First, based on the battery cell specifications, hardware response time and relevant safety standards, the minimum allowable value of each parameter is determined; second, representative user usage modes are selected for laboratory simulation tests, with the goal of minimizing power consumption and meeting monitoring requirements, and recommended parameter values ​​are determined through experiments; finally, real vehicle road tests are conducted and fine-tuned based on user feedback to ensure the adaptability of the parameters in actual applications.

[0039] Specifically, the value of T1 is dynamically adjusted based on the current usage probability, battery level, temperature conditions, and invalid wake-up events. For example, during peak usage periods, T1 is automatically extended to prevent short pauses from causing a disruption to the user experience; when battery is low, T1 is automatically shortened to allow for faster entry into a low-power state and energy saving.

[0040] The value of T2 is also dynamically adjusted based on multiple factors. When invalid wake-ups are frequent, T2 is automatically extended to reduce the number of wake-ups and avoid repeated invalid power consumption; when the usage probability is high, T2 is appropriately shortened to ensure that the system can respond to users' possible vehicle usage needs in a timely manner.

[0041] The T3 timeout is introduced to prevent misjudgments caused by instantaneous voltage fluctuations. When the power is low, the T3 timeout automatically shortens, allowing the system to make an undervoltage judgment more quickly and enter deep sleep mode to protect the remaining power. When invalid wake-ups are frequent, the T3 timeout is also shortened accordingly to reduce repeated judgments caused by interference.

[0042] Understandably, when the vehicle usage probability is high for the current time slice, the delay time T1 for entering standby mode, the RTC timed wake-up interval T2 in hibernation mode, and the undervoltage determination confirmation time T3 are all extended accordingly. This means that the system tends to maintain a standby state for a longer period of time during peak user periods, wake up to check the status more frequently, and confirm undervoltage events with a more cautious attitude, thereby ensuring a rapid response to user needs and avoiding misjudgments. Conversely, when the battery is low, the temperature is too low, or there are frequent invalid wake-ups, the above three time parameters are shortened accordingly. The system switches from working mode to standby mode more quickly, wakes up to monitor battery status at shorter intervals, and makes undervoltage determinations to enter deep sleep mode more quickly, thereby prioritizing battery safety and reducing energy consumption under adverse conditions.

[0043] Referring to Figure 2, this disclosure also integrates a multi-source, priority-based hybrid wake-up network to balance low power consumption and secure response. This hybrid wake-up source system encompasses various wake-up sources, including hardware interrupts (such as charging access detection and AFE fault interrupt), low-current load detection, wireless communication connection requests (such as Bluetooth connection requests), and internal timers (RTC timed wake-up). Depending on the power consumption mode requirements, the system activates different sets of wake-up sources: in operating mode, all wake-up sources are active; in standby mode, key wake-up sources such as charging access and Bluetooth connection requests are retained; in sleep mode, RTC timed wake-up and charging access wake-up are the primary sources; and in deep sleep mode, only blind charging activation wake-up triggered by the hardware monitoring circuit is retained. All wake-up decisions are based on hardware signals, eliminating the need for software polling and ensuring fast and reliable response.

[0044] Preferably, the mode switching unit is also used for: when the charging / discharging current is detected or a valid interactive wake-up is achieved, the control device enters the working mode. In the working mode, the DC-DC buck circuit is enabled, and the LDO regulator is disabled. In the working mode, the system operates at full capacity, the MCU works at full speed, and continuously executes all functions such as voltage / temperature acquisition, status calculation, MOSFET control, and Bluetooth communication. In this mode, the total power consumption of the board is approximately 10mA, ensuring that the battery pack is in a fully controllable and interactive state, meeting the needs of charging / discharging management and real-time user queries.

[0045] When there is no continuous charging / discharging current and no fault, the control device enters standby mode. In standby mode, the DC-DC buck circuit is enabled, and the power path of non-core modules is disconnected by controlling the load switch. In standby mode, the system shuts down the power supply to non-core circuits such as the Bluetooth module and some sensors. The MCU reduces its operating frequency or enters an intermittent working state, but keeps the MOSFET closed to maintain the output path. In this mode, the overall power consumption of the board is reduced to about 800uA, significantly reducing power consumption while maintaining fast response capability. The system can immediately exit standby and return to working mode through events such as small current load access, Bluetooth connection requests, and charger insertion, ensuring a smooth user experience.

[0046] When there is no charging / discharging current in standby mode for more than a first preset time, the control device enters sleep mode. In sleep mode, the DC-DC buck circuit is disabled, the LDO regulator is enabled, and the microcontroller enters deep sleep mode and is woken up by the RTC at regular intervals. In sleep mode, the system is powered by the ultra-low power LDO, the MCU is in deep sleep, and only relies on the RTC for brief wake-ups for simplified system checks. In this mode, the overall power consumption of the board is reduced to about 350uA, which is a further reduction in power consumption compared to standby mode. It should be noted that in sleep mode, the MOSFET drive circuit is de-energized, causing the MOSFET to disconnect and thus cutting off the main circuit. This is to avoid any potential leakage paths when in deep sleep mode, further ensuring safety and energy saving.

[0047] When any single battery cell voltage is detected to be below the undervoltage threshold, the control device enters deep sleep mode. In deep sleep mode, the DC-DC buck converter and LDO regulator are disabled, and power is supplied by the hardware monitoring circuit. Deep sleep mode is the ultimate power-saving protection state, reducing the overall board power consumption to below 30uA to maximize the preservation of the last remaining charge. In this mode, the system can only exit through a special blind charging activation method: when the charger is plugged in, the hardware monitoring circuit detects the charging voltage and generates a wake-up signal to re-power the system. This design effectively prevents irreversible battery damage caused by continuous power consumption by the BMS at low charge levels, and also prevents the system from being unable to be woken up again.

[0048] In this disclosure, this embodiment achieves power management ranging from full-function operation at the 10mA level to deep sleep at the 30μA level through progressive control of four power consumption modes. The system dynamically switches power supply strategies according to the vehicle's usage status: the operating mode ensures full functionality, the standby mode maintains rapid response, the sleep mode relies on the low-power LDO for operation, and the deep sleep mode uses hardware circuitry to safeguard the last remaining battery power. Each mode has a clear function, minimizing idle power consumption and extending battery standby time and lifespan while ensuring safety and responsiveness.

[0049] Preferably, according to dynamic switching parameters, the control device switches between working mode, standby mode, hibernation mode, and deep sleep mode, and controls the enable state of the DC-DC buck circuit and the voltage regulator LDO, including: according to the delay time T1 for entering standby mode in the dynamic switching parameters, the control device switches from working mode to standby mode; according to the RTC timed wake-up interval T2 in hibernation mode in the dynamic switching parameters, the control device controls the microcontroller to wake up in hibernation mode at regular intervals; according to the undervoltage determination confirmation time T3 in the dynamic switching parameters, the control device confirms the voltage of a single battery cell below the undervoltage threshold for a certain period of time, and when the undervoltage condition is continuously met within the confirmation period, the control device enters deep sleep mode.

[0050] In a preferred embodiment, the probability is obtained through the following steps: dividing a day into several time slices, and defining a binary activity indicator for each time slice k. : The usage probability is updated using EWMA, as shown in the following formula: ;in, The probability of use at the current moment. The probability of use at the previous moment. This is the smoothing coefficient.

[0051] In this disclosure, the smoothing coefficient This determines the weighting of historical and new data. The larger the value, the faster the system responds to new changes; The smaller the value, the more persistent the impact of historical data. This can be achieved by adjusting... A certain value can balance the system's response speed and stability. For example, for users with highly regular commuting scenarios, a smaller value can be selected. This value allows the system to more stably remember user habits; for scenarios with significant changes in user behavior, a larger value can be selected. This value enables the system to quickly adapt to new usage patterns. This EWMA-based update mechanism allows the system to gradually "memorize" user usage patterns and form adaptive probability estimates without storing large amounts of historical data, making it particularly suitable for resource-constrained embedded BMS systems.

[0052] In a preferred embodiment, the charge stress factor and temperature factor are obtained through the following steps: the charge stress factor and temperature factor are obtained by monotonically normalizing the state of charge (SOC) and temperature, as shown in the following equation: ; ;in, This represents the amplitude limiting function, limiting the amplitude to 0~1; The current state of charge, This is the normal reference boundary for the state of charge; This is the limit of the state of charge; The current temperature; This is the normal temperature reference boundary; This is the low-temperature boundary.

[0053] The above steps, through normalization, unify physical quantities with different dimensions into the 0-1 range, facilitating subsequent weighted calculations. When SOC approaches its lower limit, E... A value close to 1 indicates extremely low battery levels; when the temperature approaches the low-temperature threshold, A value close to 1 indicates that the environment has a significant impact on battery performance. It should be noted that the above reference boundary values ​​are only examples and can be adjusted according to specific battery characteristics and usage scenarios in actual applications.

[0054] In a preferred embodiment, the invalid wake-up factor is obtained through the following steps: if the MCU is woken up and no valid charging / discharging current is detected within a preset observation window, no valid communication interaction is established, and no fault handling occurs, it is recorded as an invalid wake-up, and the count W is increased; the invalid wake-up factor is determined by the following formula. : ;in, This represents the current number of invalid wake-ups. This is the maximum number of invalid wake-ups; W is reset to zero in response to the detection of a valid charge / discharge event.

[0055] This zeroing mechanism ensures that the invalid wake-up count is reset after each valid use, preventing historical interference from continuing to affect subsequent decisions. When invalid wake-ups occur frequently... As the value approaches 1, the system will correspondingly shorten various time parameters, enter a low-power state more quickly, and reduce the wake-up frequency, thus implementing an adaptive backoff mechanism that effectively suppresses energy waste in noisy environments. For example, in environments with strong electromagnetic interference, the system may be frequently and falsely woken up due to noise. Through the action of the invalid wake-up factor, the system will automatically extend the wake-up interval, shorten the standby delay, and reduce the additional power consumption caused by false wake-ups.

[0056] When battery power is low or there are many invalid wake-ups, the decision-making process for entering deep sleep and the number of invalid wake-ups can be appropriately shortened, thereby conserving remaining battery power. For example, in critical scenarios (such as low battery or harsh environments), the system automatically adjusts its behavior strategy to prioritize battery safety and basic functions, reflecting the concept of intelligent power management. For instance, when the battery power is below 20% and the ambient temperature is below 5°C, the system will significantly shorten various time parameters, enter deep sleep state more quickly, and maximize the battery's usable time.

[0057] Example 2 This disclosure provides a specific embodiment of a low-power control method for a lithium battery BMS for electric bicycles, applied to the device in Example 1. The method includes: acquiring battery status information and vehicle usage activity characteristics; generating dynamic switching parameters based on the battery status information and vehicle usage activity characteristics; controlling the device to switch between working mode, standby mode, hibernation mode and deep sleep mode based on the dynamic switching parameters, and controlling the enabling state of the high-efficiency DC-DC buck circuit and the low-power linear regulator LDO.

[0058] This method embodiment corresponds to the device embodiment. By performing the above steps, the same technical effect as the device embodiment is achieved, that is, by adaptive multi-level power consumption management, static power consumption is reduced to the minimum while ensuring system security and responsiveness.

[0059] Specifically, the method includes: dividing a day into multiple time slices according to a preset time division rule, obtaining the effective vehicle activity indicators within multiple historical time slices, and updating the usage probability of the current time slice based on an exponentially weighted moving average; obtaining the battery's state of charge and ambient temperature, and determining the power shortage factor and temperature factor based on the state of charge and ambient temperature, respectively; obtaining the statistical count of invalid wake-up events, and determining the invalid wake-up factor based on the statistical count; generating dynamic switching parameters based on the usage probability, power shortage factor, temperature factor, and invalid wake-up factor; and controlling the switching of the four-level mode and the power enable state based on the dynamic switching parameters.

[0060] Example 3: A vehicle including a lithium battery BMS low-power device for an electric bicycle as described in Example 1.

[0061] This vehicle embodiment integrates the aforementioned low-power BMS device, minimizing battery power loss during long-term storage of the electric bicycle. This allows users to obtain a longer driving range when using it again, while also effectively extending the overall battery lifespan. This vehicle is particularly suitable for models such as shared electric bicycles and household electric bicycles that involve frequent start-stop operations and long periods of inactivity.

[0062] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0063] In addition, the functional modules in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0064] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A low-power lithium battery BMS device for electric bicycles, characterized in that, The device includes a power distribution network and a microcontroller. The power distribution network includes a DC-DC buck converter and an LDO (Levelized Displacement Regulator), with the enable terminals of both controlled by the microcontroller. The microcontroller includes: a status monitoring unit for acquiring battery status information and vehicle usage activity characteristics; a parameter generation unit for generating dynamic switching parameters based on the battery status information and vehicle usage activity characteristics; and a mode switching unit for controlling the device to switch between operating mode, standby mode, sleep mode, and deep sleep mode based on the dynamic switching parameters, and controlling the enable state of the DC-DC buck converter and the LDO. The DC-DC buck converter provides power in operating and standby modes, and the LDO provides power in sleep mode.

2. The low-power lithium battery BMS device for electric bicycles according to claim 1, characterized in that, The status monitoring unit includes: a first monitoring subunit, used to divide a day into multiple time slices according to a preset time division rule, obtain the effective vehicle activity indications within multiple historical time slices, and update the usage probability of the current time slice based on an exponentially weighted moving average; a second monitoring subunit, used to obtain the battery's state of charge and ambient temperature, and determine a power stress factor and a temperature factor based on the state of charge and ambient temperature, respectively; and a third monitoring subunit, used to obtain the statistical count of invalid wake-up events, and determine an invalid wake-up factor based on the comparison result of the statistical count of invalid wake-up events with a preset upper limit value; wherein an invalid wake-up event is an event in which the microcontroller, after being woken up, fails to detect effective charging / discharging current, fails to establish effective communication interaction, and fails to handle faults within a preset observation window.

3. The low-power lithium battery BMS device for electric bicycles according to claim 2, characterized in that, The parameter generation unit is used to generate the dynamic switching parameters according to the usage probability, the power shortage factor, the temperature factor, and the invalid wake-up factor, in the following manner: the dynamic switching parameters include at least one of the following: the delay time T1 for entering standby mode, the real-time clock (RTC) wake-up interval T2 in sleep mode, and the undervoltage determination confirmation time T3: ； ； ;in, As a factor of power shortage, For temperature factor, Number of times the device cannot be woken up. The probability of use at the current moment. The preset base delay time, The preset base wake-up time, This is the preset undervoltage determination basis confirmation time.

4. The low-power lithium battery BMS device for electric bicycles according to claim 1, characterized in that, The mode switching unit is also used to: control the device to enter the working mode when the charging and discharging current or effective interactive wake-up is detected, and enable the DC-DC buck circuit and disable the regulator LDO in the working mode; When there is no continuous charging / discharging current and no fault, the device is controlled to enter standby mode. In standby mode, the DC-DC buck circuit is enabled, and the power path of non-core modules is disconnected by controlling the load switch. When there is no continuous charging / discharging current in standby mode for more than a first preset time, the device is controlled to enter sleep mode. In sleep mode, the DC-DC buck circuit is disabled, the voltage regulator LDO is enabled, the microcontroller enters deep sleep state and is woken up according to the RTC timer. When the voltage of any single battery cell is detected to be lower than the undervoltage threshold, the device is controlled to enter deep sleep mode. In deep sleep mode, the DC-DC buck circuit and the voltage regulator LDO are disabled, and power is supplied by the hardware monitoring circuit.

5. The low-power lithium battery BMS device for electric bicycles according to claim 4, characterized in that, The step of controlling the device to switch between working mode, standby mode, hibernation mode, and deep sleep mode according to the dynamic switching parameters, and controlling the enable state of the DC-DC buck circuit and the voltage regulator LDO, includes: controlling the device to switch from working mode to standby mode according to the delay time T1 for entering standby mode in the dynamic switching parameters; controlling the microcontroller to wake up from hibernation mode at regular intervals according to the RTC timed wake-up interval T2 in hibernation mode according to the dynamic switching parameters; and controlling the device to enter deep sleep mode when the voltage of a single battery cell is lower than the undervoltage threshold according to the undervoltage determination confirmation time T3 in the dynamic switching parameters.

6. The low-power lithium battery BMS device for electric bicycles according to claim 3, characterized in that, The usage probability is obtained through the following steps: dividing a day into several time slices, and defining a binary activity indicator for each time slice k. ： The usage probability is updated using EWMA, as shown in the following formula: ;in, The probability of use at the current moment. The probability of use at the previous moment. This is the smoothing coefficient.

7. The low-power lithium battery BMS device for electric bicycles according to claim 3, characterized in that, The energy stress factor and temperature factor are obtained through the following steps: The energy stress factor and temperature factor are obtained by monotonically normalizing SOC and temperature, as shown in the following formula: ； ;in, This represents the amplitude limiting function, limiting the amplitude to 0~1; The current state of charge, This is the normal reference boundary for the state of charge; This is the limit of the state of charge; The current temperature; This is the normal temperature reference boundary; This is the low-temperature boundary.

8. The low-power lithium battery BMS device for electric bicycles according to claim 3, characterized in that, The invalid wake-up factor is obtained through the following steps: If the MCU is woken up and no effective charging / discharging current is detected within a preset observation window, no effective communication interaction is established, and no fault handling occurs, it is recorded as an invalid wake-up, and the count W is increased; the invalid wake-up factor is determined by the following formula. ： ;in, This represents the current number of invalid wake-ups. This is the maximum number of invalid wake-ups; W is reset to zero in response to the detection of a valid charge / discharge event.

9. A low-power control method for a lithium battery BMS in an electric bicycle, characterized in that, The method, applied to any one of claims 1-8, comprises: acquiring battery status information and vehicle usage activity characteristics; generating dynamic switching parameters based on the battery status information and the vehicle usage activity characteristics; controlling the device to switch between a working mode, a standby mode, a hibernation mode, and a deep sleep mode based on the dynamic switching parameters, and controlling the enable state of the high-efficiency DC-DC buck circuit and the low-power linear regulator (LDO).

10. A vehicle, characterized in that, Includes the lithium battery BMS low-power device for electric bicycles according to any one of claims 1-8.

Images