A dual-mode power supply adaptive control method and system based on lithium batteries

Abstract

This invention relates to the field of satellite communication terminal power supply management technology, and discloses a dual-mode power supply adaptive control method and system based on lithium batteries. The method includes acquiring a continuous voltage sequence of an external power source and performing multi-window sliding comparison to obtain a jitter count; performing load adaptive ampere-hour integration on battery state of charge and communication load data to obtain the estimated remaining available power; making a hysteresis-based hierarchical switching decision based on the jitter count and the estimated remaining available power to obtain a power supply switching decision; and performing synchronous switching after pre-charging the battery-side bus to obtain the corresponding power supply state. This invention can distinguish between intermittent disconnections and actual faults of the external power source, and makes reliable power supply switching decisions based on the remaining battery power, avoiding false switching and communication interruptions.

Description

Technical Field

[0001] This invention relates to the field of power supply management technology for satellite communication terminals, and in particular to a dual-mode power supply adaptive control method and system based on lithium batteries. Background Technology

[0002] As a critical infrastructure for global coverage, satellite communication plays an irreplaceable role in scenarios such as remote areas, maritime operations, and emergency rescue. Low-Earth orbit satellite communication terminals have become an important part of this field due to their portability and rapid deployment capabilities. The intelligent power supply scheduling system of these terminal devices is a key link in ensuring communication continuity.

[0003] Current technologies for power supply scheduling in portable communication terminals mostly employ a simple strategy of prioritizing external power or supplementing with batteries. They use fixed voltage thresholds to determine if the external power supply is abnormal and trigger power switching accordingly. However, in border areas where the mains voltage is unstable or in vehicle-mounted scenarios, external voltage fluctuations can take many forms, including momentary drops, persistent undervoltage, and sudden spikes. Fixed voltage thresholds cannot distinguish between transient interference and genuine power supply failures, easily triggering incorrect switching during normal fluctuations or delaying response in the initial stages of an actual power outage. Furthermore, the state of charge of lithium batteries continuously changes during discharge. If the matching relationship between the remaining battery capacity and the communication load is not comprehensively assessed, the system may rashly switch to battery power when the battery is insufficient, leading to communication interruption.

[0004] In summary, existing technologies have the problem of failing to distinguish between intermittent disconnections of external power sources and actual faults, and failing to make reliable power supply switching decisions based on the remaining battery power. Summary of the Invention

[0005] This invention provides a dual-mode power supply adaptive control method and system based on lithium batteries to solve the problem of being unable to distinguish between intermittent disconnection of external power supply and real fault, and to make reliable power supply switching decisions based on the remaining battery power.

[0006] In a first aspect, to solve the above-mentioned technical problems, the present invention provides a dual-mode power supply adaptive control method based on a lithium battery, comprising: A continuous voltage sequence of an external power supply is obtained. Anomalies are detected in the continuous voltage sequence using a multi-window sliding comparator. Instantaneous voltage changes in the continuous voltage sequence are captured and the number of abnormal events per unit time is counted to obtain a jitter count. Obtain battery state of charge and communication load data, perform load-adaptive ampere-hour integration on the battery state of charge and the communication load data, and obtain the estimated remaining available power. Based on the jitter count and the estimated remaining available power, a hysteresis-based hierarchical switching decision is made to obtain a power supply switching decision; Based on the power supply switching decision, the battery-side bus is pre-charged, and a power supply switching control command is sent to the physical layer power supply control module through pre-charging synchronous switching to obtain the corresponding power supply status.

[0007] Secondly, the present invention provides a dual-mode power supply adaptive control system based on a lithium battery, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the dual-mode power supply adaptive control method based on a lithium battery as described above.

[0008] Compared with the prior art, the present invention has the following beneficial effects: (1) This invention captures the instantaneous changes of external voltage by a multi-window sliding comparator and counts the number of abnormal events to obtain jitter count. The jitter count replaces the fixed voltage threshold to distinguish between intermittent disconnection of external power supply and real fault. Thus, it does not trigger switching during normal fluctuations and responds in time during real power outages, solving the problem of false switching caused by inaccurate power supply abnormality identification.

[0009] (2) The present invention obtains the dynamic discharge rate by performing load adaptive ampere-hour integration on the battery state of charge and communication load data, and then calculates the expected remaining power, so that the switching decision is based on dynamic power assessment rather than static threshold, avoiding communication interruption caused by rash switching when the battery power is insufficient.

[0010] (3) The present invention uses jitter counting to determine the level and coordinates with pre-charging synchronous switching to perform a graded response with hysteresis between maintaining external power supply, reducing power to maintain and switching to battery power supply, and closes the switching switch after pre-charging the battery side bus, thereby realizing the disturbance-free switching of power supply mode and improving the operational reliability of the terminal under harsh power supply conditions. Attached Figure Description

[0011] Figure 1 A flowchart illustrating a dual-mode power supply adaptive control method based on a lithium battery, provided in an embodiment of the present invention; Figure 2 This is a schematic diagram illustrating the expected changes in remaining battery power under different operating states of the communication module, provided in an embodiment of the present invention. Figure 3 This is a schematic diagram of a hysteresis-based hierarchical handover decision-making process provided in an embodiment of the present invention. Figure 4 This is a schematic diagram of pre-charging synchronization switching provided in an embodiment of the present invention. Detailed Implementation

[0012] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0013] Reference Figure 1 The first embodiment of the present invention provides a dual-mode power supply adaptive control method based on lithium battery, comprising the following steps: S11, acquire the continuous voltage sequence of the external power supply, perform anomaly detection on the continuous voltage sequence through a multi-window sliding comparator, capture the instantaneous voltage changes of the continuous voltage sequence and count the number of abnormal events per unit time to obtain the jitter count; S12, acquire battery state of charge and communication load data, perform load adaptive ampere-hour integration on the battery state of charge and the communication load data to obtain the estimated available remaining power; S13, Based on the jitter count and the estimated remaining available power, perform a hysteresis-based hierarchical switching decision to obtain a power supply switching decision; S14, based on the power supply switching decision, pre-charge the battery-side bus, and send a power supply switching control command to the physical layer power supply control module through pre-charging synchronous switching to obtain the corresponding power supply status.

[0014] In one implementation, the step of proceeding from steps S11 to S14 further includes: During normal power supply, the supercapacitor is charged by an external power source; during voltage drop when the jitter count is less than a first preset threshold, the supercapacitor discharges to the load to maintain the load voltage at a level not lower than a preset undervoltage threshold; during the transition period after a real fault is determined and before switching to battery power supply, the supercapacitor maintains power supply to the load at the terminal power after reducing terminal power consumption.

[0015] In step S11, a continuous voltage sequence of an external power supply is obtained, and anomaly detection is performed on the continuous voltage sequence using a multi-window sliding comparator. The instantaneous voltage changes of the continuous voltage sequence are captured, and the number of abnormal events per unit time is counted to obtain a jitter count.

[0016] In one implementation, the step of performing anomaly detection on the continuous voltage sequence using a multi-window sliding comparator, capturing instantaneous voltage changes in the continuous voltage sequence, and counting the number of abnormal events per unit time to obtain a jitter count includes: Set up fast and slow windows; The instantaneous voltage changes of the continuous voltage sequence are captured through the fast window; The number of abnormal events is obtained by counting the number of abnormal events per unit time using the slow window. The jitter count is determined by a jitter counter based on the number of abnormal events.

[0017] It should be noted that a voltage sensor continuously collects data from the DC input terminal of the external power supply to form a continuous voltage sequence for anomaly detection. Specifically, the external power supply bus voltage is sampled at a sampling frequency of 1 kHz, with a time interval of 1 ms between adjacent sampling points, and arranged in chronological order to form the continuous voltage sequence.

[0018] Specifically, the fast window and the slow window are two sliding windows with different time scales. The fast window is set to a length of 10 ms, covering 10 consecutive sampling points, and is used to capture instantaneous changes in the bus voltage; the slow window is set to a length of 500 ms, covering 500 consecutive sampling points, and is used to count the number of abnormal events per unit time.

[0019] It should be noted that the abnormal event refers to the instantaneous event in which the bus voltage drops below a preset undervoltage threshold within the fast window. The preset undervoltage threshold is set to 90% of the rated voltage. For example, if the rated voltage of the external power supply is 12 V, the preset undervoltage threshold is set to 10.8 V accordingly.

[0020] Specifically, within each fast window, the minimum value of the bus voltage within that fast window is compared with the preset undervoltage threshold. If the minimum value is lower than the preset undervoltage threshold, an abnormal event is determined to have occurred within that fast window. Within the slow window, the number of abnormal events determined by the fast windows that have been slid through is accumulated to obtain the number of abnormal events.

[0021] It should be noted that the jitter counter determines the jitter count based on the number of abnormal events. The jitter count is used to distinguish between intermittent disconnections and actual faults of the external power supply in subsequent graded switching decisions. For example, in an external power supply bus with a rated voltage of 12 V, if the bus voltage repeatedly drops briefly to below 10.8 V twice within 200 ms and then recovers to the rated value, the number of abnormal events counted by the slow window is 2, and the jitter count is 2. When the bus voltage drops continuously and cannot recover for a long time, the number of abnormal events counted by the slow window including the drop period is the number of fast windows covered by the drop period. For example, when the continuous drop covers 12 fast windows, i.e., 120 ms, the number of abnormal events counted by the slow window is 12, and the jitter count is 12. As the drop continues, the number of abnormal events counted by the subsequent slow window increases to the total number of fast windows within the slow window, which is 50, and remains at this upper limit.

[0022] For example, the sampling frequency is set to 1 kHz, the fast window length is 10 ms, the slow window length is 500 ms, and the preset undervoltage threshold is 10.8 V. The fast window slides along the continuous voltage sequence in 10 ms steps, extracting the minimum voltage value within each 10 ms window. The slow window slides along the continuous voltage sequence in 500 ms steps, counting the number of fast windows within the 500 ms window whose minimum voltage value is lower than 10.8 V at each step. For example, within the 500 ms covered by a certain slow window, there are 50 fast windows. The minimum voltage values ​​of the 3rd, 7th, and 12th fast windows are 10.2 V, 10.5 V, and 10.1 V, respectively, all lower than the preset undervoltage threshold of 10.8 V. The minimum voltage values ​​of the remaining 47 fast windows are all not lower than 10.8 V. Therefore, the number of abnormal events counted by this slow window is 3, resulting in a jitter count of 3. As can be seen, by capturing instantaneous voltage changes with a fast window and counting the number of abnormal events per unit time with a slow window, the multi-window sliding comparator can convert the voltage fluctuations of the external power supply into quantifiable jitter counts, providing a basis for subsequent graded switching decisions.

[0023] In step S12, the battery state of charge and communication load data are acquired, and the load adaptive ampere-hour integration is performed on the battery state of charge and the communication load data to obtain the estimated remaining available power.

[0024] In one implementation, the step of performing load-adaptive ampere-hour integration on the battery state of charge and the communication load data to obtain the estimated available remaining power includes: Determine the corresponding discharge coefficient based on the operating status of the communication module; Based on the discharge coefficient, the dynamic discharge rate is calculated using a state-by-state discharge model; wherein the state-by-state discharge model uses the operating state of the communication module as segmented variables, and each operating state corresponds to a constant discharge rate. The estimated remaining available power is calculated by ampere-hour integration based on the dynamic discharge rate.

[0025] It should be noted that the initial battery state of charge and real-time communication load data of the terminal are obtained. The initial battery state of charge is estimated by the battery management chip through the ampere-hour integration method and the open-circuit voltage method. The real-time communication load data is obtained from the current working status reported by the communication baseband module.

[0026] For example, the initial state of charge (SOC) of the battery is estimated by the battery management chip using a combination of the ampere-hour integration method and the open-circuit voltage method. The open-circuit voltage method maps the open-circuit voltage of the lithium battery to the initial SOC value by looking up the table based on the monotonic correspondence between the open-circuit voltage and the SOC of the lithium battery (this correspondence varies depending on the cell system; the lookup table is pre-calibrated based on the cell system of the actual battery used. For example, the cell in this embodiment is a ternary lithium battery cell). The ampere-hour integration method starts from this initial SOC value and continuously updates the SOC based on the integral of the current over time during the charging and discharging process. For example, after the battery has stabilized at rest, the open-circuit voltage of a single cell is measured to be 3.95 V. Looking up the corresponding initial state of charge (SOC) value in a table, it is found to be 85%. The battery pack consists of three lithium batteries connected in series, with an open-circuit voltage of 11.85 V and a rated total capacity of 10 Ah. Subsequently, the battery enters the discharge process. The ampere-hour integral method updates the SOC in real time based on the integral of the discharge current over time, thus providing an initial SOC of 85% for load-adaptive ampere-hour integral. It should be noted that the open-circuit voltage method has higher accuracy when the battery is at rest, while the ampere-hour integral method has higher accuracy during dynamic charging and discharging. The combined estimation of both methods can balance static accuracy and dynamic tracking capability, providing a reliable input for calculating the estimated remaining usable capacity.

[0027] Specifically, the communication module's operating states include a search state, a lock state, and a standby state. The search state corresponds to the high-power operation of the communication module continuously scanning satellite signals; the lock state corresponds to the normal transmission operation of the communication module after it has locked onto a satellite; and the standby state corresponds to the low-power operation of the communication module when there is no data transmission.

[0028] It should be noted that the discharge coefficient is the discharge current of the battery per unit time under different working conditions. The discharge coefficient is determined by looking up a table. The discharge current is 3.0 A for the search state, 2.0 A for the locked state, and 1.0 A for the standby state. The discharge coefficient is calibrated based on the measured average discharge current of the same model of battery under different communication working conditions.

[0029] Specifically, the state-based discharge model determines the dynamic discharge rate based on the operating state of the communication module, meaning the discharge rate is segmented according to the communication operating state. For example, when the communication module is in a locked state, the discharge current is 2.0 A obtained by looking up the discharge coefficient in a table, and the dynamic discharge rate is 2.0 A.

[0030] For example, assuming a battery's rated total capacity of 10 Ah, an initial state of charge of 85%, and a discharge loss compensation value of 0.3 A, when the communication module is in search mode, the corresponding discharge coefficient is 3.0 A, the dynamic discharge rate is 3.0 A, and the predicted power consumption is 3.0 A plus 0.3 A equals 3.3 A. The predicted power consumption within 1 hour is 3.3 Ah, representing 33% of the rated total capacity of 10 Ah. Subtracting 33% from the remaining capacity baseline of 85% yields an estimated remaining power of 52%. When the communication module is in locked mode, the corresponding discharge coefficient is 2.0 A, the dynamic discharge rate is 2.0 A, and the predicted power consumption is 2.0 A plus 0.3 A equals 2.3 A. The predicted power consumption within 1 hour is 2.3 Ah, representing 23% of the rated total capacity of 10 Ah. Subtracting 23% from the remaining capacity baseline of 85% yields an estimated remaining power of 62%. When the communication module is in standby mode, the corresponding discharge coefficient is 1.0 A, the dynamic discharge rate is 1.0 A, and the predicted power consumption is 1.0 A plus 0.3 A equals 1.3 A. The predicted power consumption within 1 hour is 1.3 Ah, which is 13% of the rated total capacity of 10 Ah. Subtracting 13% from the remaining capacity baseline of 85% yields an estimated remaining power of 72%. It can be seen that under the same initial battery charge state, the estimated remaining power increases sequentially for the search, lockout, and standby operating states. The state-specific discharge model enables the discharge rate to adaptively adjust with the communication operating state, thus reflecting the actual power consumption under different communication loads in the switching decision.

[0031] In one implementation, calculating the estimated remaining available power capacity by ampere-hour integration based on the dynamic discharge rate includes: Based on the communication load data, load fluctuation characteristics are extracted; based on the load fluctuation characteristics and the dynamic discharge rate, a discharge loss compensation value is determined; based on the dynamic discharge rate and the discharge loss compensation value, a power consumption prediction value is calculated. The predicted power consumption is multiplied by a preset prediction time to obtain the power consumption; the proportion of the power consumption to the battery's rated total capacity is subtracted from the remaining capacity baseline to obtain the estimated remaining power.

[0032] It should be noted that a sliding window mechanism is used to extract load fluctuation characteristics for real-time communication load data. Specifically, the time window length is set to 500 ms with a step size of 100 ms, and the number of times the communication module switches its working state within this window is calculated as the load fluctuation characteristic. This characteristic reflects the frequency of the communication module switching between search, lock, and standby states.

[0033] Specifically, a discharge loss compensation value is determined based on the load fluctuation characteristics and the dynamic discharge rate. The communication module generates additional nonlinear energy loss when frequently switching operating states. This compensation value quantifies this hidden loss and is determined using a lookup table method. A lookup table is pre-calibrated based on measured losses of the same battery model under different combinations of state switching frequencies and discharge rates. When looking up the table, the dynamic discharge rate corresponds to the discharge coefficient values ​​for the search, lock, and standby operating states, directly matching the corresponding level. When the number of state switching cycles falls between the calibrated levels, the number of state switching cycles is linearly interpolated using the compensation values ​​of adjacent levels. When the number of state switching cycles exceeds the highest level, the compensation value of the highest level is used. For example, using 8 state switching cycles and a dynamic discharge rate of 2.0 A as the index, a discharge loss compensation value of 0.3 A is obtained by looking up the table.

[0034] For example, the discharge loss compensation value is used to quantify the additional nonlinear power loss generated by the communication module when frequently switching operating states. During calibration, a battery of the same model as the terminal battery is selected. Under standard dynamic operating conditions, and using statistical methods consistent with the online load fluctuation characteristics, the number of state switching times within a 500 ms time window is set to four levels: 0, 4, 8, and 12 times. Within each level, the discharge rate is set to three levels: 1.0 A, 2.0 A, and 3.0 A, resulting in a total of 12 calibration operating conditions. For each operating condition, the battery is continuously run for 1 hour, and the actual power consumption of the battery within 1 hour is measured using the battery management chip. The theoretical power consumption corresponding to the discharge rate under that operating condition is then subtracted to obtain the discharge loss compensation value under that operating condition. The 12 calibration results are then written into a compensation value lookup table. For example, under the calibration conditions of 8 state transitions within a 500 ms time window and a discharge rate of 2.0 A, the actual power consumption measured in 1 hour is 2.3 Ah. After deducting the theoretical power consumption of 2.0 Ah corresponding to the discharge rate of 2.0 A, the discharge loss compensation value is 0.3 A. This value is written into the compensation value lookup table for online table lookup.

[0035] Specifically, based on the dynamic discharge rate and the discharge loss compensation value, the predicted power consumption value is calculated. The dynamic discharge rate of 2.0 A and the discharge loss compensation value of 0.3 A are directly added together to obtain the predicted power consumption value of 2.3 A.

[0036] It should be noted that the rated total capacity of the terminal battery is obtained, which is 10 Ah. Combined with the current initial battery state of charge of 85%, the remaining capacity benchmark is established as 85%.

[0037] Specifically, the predicted power consumption is subtracted from the remaining capacity baseline by integrating the ampere-hours, that is, by multiplying the predicted power consumption by the predicted time to obtain the power consumption, and then the proportion of the power consumption to the rated total capacity is subtracted from the remaining capacity baseline to obtain the expected available remaining power.

[0038] The following example, using the combined calculation of the state-based discharge model and the ampere-hour integral, further illustrates the relationship between the dynamic discharge rate and the estimated remaining power. The state-based discharge model divides the discharge rate into several segments based on the communication module's operating state. Each segment corresponds to a constant discharge coefficient. The ampere-hour integral accumulates the power consumption within each segment at a constant discharge rate. For example, if the terminal is in a search state for the first 30 minutes within a 1-hour prediction period, corresponding to a discharge coefficient of 3.0 A, and in a locked state for the next 30 minutes, corresponding to a discharge coefficient of 2.0 A, then the dynamic discharge rate output by the state-based discharge model is 3.0 A and 2.0 A for the two periods, respectively. The ampere-hour integral accumulates the power consumption within this prediction period as 3.0 A multiplied by 30 minutes plus 2.0 A multiplied by 30 minutes, which is equivalent to 2.5 Ah. This represents 25% of the rated total capacity of 10 Ah. Subtracting 25% from the remaining capacity baseline of 85%, the estimated remaining power is 60%.

[0039] For example, Figure 2 This diagram illustrates the expected remaining battery power under different communication module operating states. The initial battery state of charge is 85%. The discharge coefficients for the search, locked, and standby states are 3.0 A, 2.0 A, and 1.0 A, respectively. Based on load-adaptive ampere-hour integration calculations, the expected remaining battery power after 60 minutes is 52%, 62%, and 72%, respectively. The calculated results for each state are all higher than the preset safe battery power threshold of 20%, demonstrating the impact of communication load changes on the expected remaining battery power.

[0040] In step S13, a hysteresis-based hierarchical switching decision is made based on the jitter count and the estimated remaining available power, resulting in a power supply switching decision.

[0041] In one implementation, the step of performing a hysteresis-based tiered switching decision based on the jitter count and the estimated remaining available power to obtain a power supply switching decision includes: If the jitter count is less than the first preset threshold, then the external power supply is maintained; If the jitter count is not less than the first preset threshold and less than the second preset threshold, then the terminal power consumption is reduced to maintain the external power supply. If the jitter count is not less than the second preset threshold, then switch to battery power to obtain the power supply switching decision.

[0042] It should be noted that the tiered switching decision adopts a hysteresis-based judgment method. Different thresholds are used for entering and exiting the power supply mode. When the exit threshold is lower than the entry threshold, the jitter count rises from low to high, entering a higher level; and then falls back to below the exit threshold before reverting to a lower level. This avoids repeated power supply mode switching when the jitter count fluctuates around the threshold. Specifically, the first preset threshold is set to 3 times, and the second preset threshold is set to 10 times. For example, the hysteresis is set to 1 time: to switch from maintaining external power supply to reducing terminal power consumption, a jitter count of at least 3 is required; to switch back to maintaining external power supply, a jitter count of less than 2 is required. Switching from reducing terminal power consumption to battery power supply, a jitter count of at least 10 is required; to switch back from battery power supply, a jitter count of less than 8 is required. The first and second preset thresholds are calibrated based on the measured jitter count distribution under intermittent power outages and real fault scenarios.

[0043] Specifically, when the jitter count is less than 3 times, it is determined that the external power supply is only intermittently disconnected, and the supercapacitor provides short-term energy support to maintain the external power supply; when the jitter count is not less than 3 times and less than 10 times, it is determined that the external power supply is fluctuating continuously, and the terminal power consumption is reduced to maintain the external power supply; when the jitter count is not less than 10 times, it is determined that the external power supply has a real fault, and the system switches to battery power.

[0044] It should be noted that the dual-mode described in this invention refers to two power supply modes: external power supply and lithium battery power supply. Reducing terminal power consumption to maintain external power supply is a degraded operation state under the external power supply mode. Limited discharge and low power standby when the power is insufficient are data protection states that prioritize the security of core data. Neither of them constitutes a third power supply mode independent of external power supply and lithium battery power supply.

[0045] For example, taking short-term supercapacitor power support as an example, when the external power supply only experiences intermittent disconnection, i.e., the jitter count is less than a first preset threshold, the terminal relies on the built-in supercapacitor to provide short-term power support. For instance, during normal external power supply operation, the external power supply continuously charges the supercapacitor to its rated voltage of 12V. When the bus voltage repeatedly drops below 10.8V twice within 200ms, the supercapacitor discharges to the load during each drop, maintaining the load voltage above 10.8V to compensate for the instantaneous drop in external power supply, thereby maintaining external power supply and preventing battery switching. It should be noted that the supercapacitor's energy support is used to cover the instantaneous drops caused by intermittent disconnections, and also to maintain the load power supply during the transition period between the occurrence of a real fault and the completion of battery takeover. The capacity of the supercapacitor is calibrated based on the power of the terminal after gradual power reduction and the longest transition time required from the occurrence of a real fault to the completion of battery takeover. For example, the slow window updates the jitter count in 500 ms increments, and the transition time required for the switching determination and pre-charging synchronization does not exceed 1 s. The supercapacitor is calibrated to support the terminal power continuously for 1 s after power reduction, so that the load power supply is not interrupted before the real fault determination and the completion of the switching. After switching to battery power, the supercapacitor resumes charging.

[0046] In one implementation, the step of switching to battery power if the jitter count is not less than the second preset threshold includes: If the estimated remaining available power is greater than the preset safe power threshold, then the battery power supply link is determined, and the system switches to battery power supply based on the battery power supply link. If the estimated remaining available power is not greater than the preset safe power threshold, the system will not switch to battery power, but will gradually reduce the terminal power consumption and perform limited discharge on the core data storage module to complete the core data write-to-disk operation. Then, it will enter a low-power standby state and resume normal power supply after the external power supply recovers to above the preset undervoltage threshold.

[0047] It should be noted that the preset safe power threshold represents the minimum power required to maintain the basic data storage of the terminal, which is set to 20%. This threshold is based on the power required for the battery to complete a full data write to disk at the minimum safe depth of discharge, and is the factory configuration value of the battery management chip.

[0048] For example, the system reads the previously predicted remaining battery power of 62% and the preset safe battery power threshold of 20%. The numerical comparison logic determines that 62% is greater than 20%, indicating that the current battery has sufficient energy reserves to support a smooth switching of the power supply link. The system then determines the battery power supply link and switches to battery power.

[0049] Taking the protection scenario when the battery power is low as an example, a preset safe power threshold is set to 20%. When an external power supply experiences a real fault, with a jitter count not less than the second preset threshold of 10 times, but the expected remaining power is only 15%, since the expected remaining power is less than the preset safe power threshold of 20%, it is determined that the battery power is insufficient to support a complete switchover. In this case, the terminal does not switch to battery power, but gradually reduces the terminal's power consumption, and the battery management chip performs limited discharge on the core data storage module within a range not lower than the minimum safe discharge depth, prioritizing the completion of core data write-to-disk. After the data write-to-disk is completed, the terminal shuts down the peripheral function modules and communication modules, enters a low-power standby state, and only keeps the voltage monitoring circuit of the battery management chip and the external power bus operating with a small current. Normal power supply is restored only after the external power recovers to above the preset undervoltage threshold. It should be noted that the above-mentioned limited discharge is the controlled power supply of the core data storage module by the battery management chip and does not constitute a switchover to battery power. By setting a safe power gating in the branch switching to battery power, communication interruptions caused by abruptly switching to battery power when the battery power is low are avoided, further improving the reliability of the power supply switching decision.

[0050] In one implementation, the step of reducing terminal power consumption to maintain external power supply if the jitter count is not less than the first preset threshold and less than the second preset threshold includes: The power reduction level is determined based on the jitter count; Based on the power reduction level, the peripheral function modules of the corresponding power reduction level in the terminal are turned off to obtain the power supply power of the terminal after power reduction. Based on the reduced power consumption of the terminal, the external power supply is maintained.

[0051] It should be noted that the power reduction level is determined based on the range of the jitter count. For example, when the jitter count is 5 times, the power reduction level is determined to be Level 1, and the peripheral display module of the terminal is turned off; when the jitter count is 9 times, the power reduction level is determined to be Level 2, and the peripheral display module and the high-speed data transmission module of the terminal are turned off, while the external power supply is maintained.

[0052] Specifically, the interval between the first preset threshold of 3 times and the second preset threshold of 10 times is divided into two levels. When the jitter count is not less than 3 times and less than 6 times, it corresponds to power reduction level one; when the jitter count is not less than 6 times and less than 10 times, it corresponds to power reduction level two. For example, in power reduction level one, only the peripheral display module is turned off, and the average power consumption of the terminal decreases from 30 W to 26 W; in power reduction level two, both the peripheral display module and the high-speed data transmission module are turned off, and the average power consumption of the terminal decreases from 30 W to 20 W. This reduces the overall power consumption of the terminal, ensuring that the external power supply can still meet the basic power needs of the terminal during voltage fluctuations and maintain external power supply.

[0053] Taking the power reduction level under continuous external power fluctuation as an example, the first preset threshold is set to 3 times, and the second preset threshold is set to 10 times. When the jitter count is 5 times, the jitter count is not less than the first preset threshold of 3 times and less than the second preset threshold of 10 times, indicating that there is continuous fluctuation in the external power supply. Based on the jitter count of 5 times, the power reduction level is determined to be Level 1, and the peripheral display module of the terminal is turned off. At this time, the average power consumption of the terminal decreases from 30 W to 26 W, and the external power supply is maintained. When the jitter count is 9 times, the jitter count is still not less than the first preset threshold of 3 times and less than the second preset threshold of 10 times. Based on the jitter count of 9 times, the power reduction level is determined to be Level 2, and the peripheral display module and high-speed data transmission module of the terminal are turned off. At this time, the average power consumption of the terminal decreases from 30 W to 20 W, and the external power supply is maintained. It should be noted that by dividing the jitter count into different intervals and corresponding to different power reduction levels, and by using a threshold with hysteresis to determine the entry and exit of the power supply mode, a hysteresis-based hierarchical response is achieved. This avoids frequent switching of the power supply mode when the jitter count fluctuates near the threshold, enabling the terminal to maintain communication continuity by gradually reducing power consumption during periods of continuous fluctuation in external power supply.

[0054] For example, Figure 3 This is a schematic diagram of a tiered switching decision with hysteresis. The jitter count is obtained from the number of abnormal events within a 500-millisecond slow window. As external power fluctuations intensify, the jitter count successively exceeds the first and second preset thresholds, causing the power switching decision to shift from maintaining external power supply to reducing terminal power consumption, and then to switching to battery power. When the jitter count falls back, it must further drop below the corresponding exit threshold to exit the current level, thereby avoiding repeated switching of power supply modes near the threshold.

[0055] In step S14, based on the power supply switching decision, the battery-side bus is pre-charged, and a power supply switching control command is sent to the physical layer power supply control module through pre-charging synchronous switching to obtain the corresponding power supply status.

[0056] In one implementation, the pre-charging of the battery-side bus, and the sending of a power supply switching control command to the physical layer power supply control module via pre-charging synchronization switching to obtain the corresponding power supply state, includes: The battery-side bus is pre-charged to obtain the pre-charged battery-side bus voltage. Based on the pre-charged battery-side bus voltage, close the switching switch to obtain the battery power supply link; Disconnect the external power input and determine the corresponding power supply status based on the battery power supply link.

[0057] It should be noted that the pre-charge target voltage is the smaller of the product of the external power supply voltage and a preset ratio, and the voltage obtained by subtracting a preset margin from the current voltage of the battery pack. The preset ratio is set to 95%, and for example, the preset margin is set to 0.2 V. Before switching, the battery-side bus is pre-charged to bring the battery-side bus voltage to the pre-charge target voltage, and then the switching switch is closed to achieve a smooth switching. For example, if the external power supply voltage is 12 V and the current voltage of the battery pack is higher than 11.4 V, the pre-charge target voltage corresponding to the preset ratio of 95% is 11.4 V. The pre-charge circuit gradually raises the battery-side bus voltage from 0 V to 11.4 V.

[0058] Specifically, the pre-charging is achieved through a pre-charging circuit, which consists of a current-limiting resistor and a pre-charging switch connected in series and in parallel across the battery output switch. The battery is connected to the battery-side bus via the battery output switch. During the period when the terminal is powered by an external power source, both the battery output switch and the main switching switch remain open. The battery-side bus is an isolated bus segment that is not connected to the battery or the external power source bus, and the voltage on its bus capacitor is 0 V. During pre-charging, the pre-charging switch is closed before the main switching switch is closed, allowing the charging current to charge the bus capacitor of the battery-side bus through the current-limiting resistor. For example, when the capacitance of the bus capacitor is 100 μF and the resistance of the current-limiting resistor is 100 Ω, the pre-charging time constant is 10 ms. The battery-side bus voltage rises to more than 99% of the pre-charging target voltage of 11.4 V after approximately 50 ms. After this, the battery output switch and the switching switch are closed to obtain the battery power supply link, and then the external power input is disconnected, completing a seamless switching process.

[0059] Specifically, after the battery-side bus voltage reaches 11.4 V, the battery output switch is first closed to connect the battery to the battery-side bus, and then the switching switch is closed to keep the voltage difference between the battery power supply and the external power supply within a small range to avoid surge current at the moment of switching. Then the external power input is disconnected to achieve a smooth switching.

[0060] It should be noted that after the switching is completed, the actual transmission status of the power supply link is monitored, and the voltage at the battery output terminal and the voltage at the load terminal are collected. For example, the collected voltage at the battery output terminal is 11.2 V, and the collected voltage at the load terminal is 11.0 V. The difference between the two is 0.2 V, which reflects the energy loss caused by the parasitic parameters of the line and the on-resistance of the switching switch.

[0061] Specifically, the preset compensation range is set to 0 V to 0.5 V. The voltage drop within this range is considered normal loss caused by line parasitic parameters and the switching impedance. The voltage drop of the power supply link is determined based on the difference: if the difference is within the preset compensation range, the voltage drop is considered normal, the terminal's power supply status is set accordingly, and a seamless switching of the power supply mode is completed; if the difference exceeds the preset compensation range, the voltage drop is considered abnormal, the terminal's power supply status is set as abnormal, and an alarm is output.

[0062] Taking the complete timing sequence of pre-charge synchronous switching as an example, the external power supply voltage is set to 12V, the preset ratio is 95%, and the pre-charge target voltage is 11.4V. Before switching, both the battery output switch and the switching switch are kept open, the battery-side bus is an isolated bus segment, not connected to the battery or the external power supply bus, and the battery-side bus voltage is 0V. When switching begins, the pre-charge circuit charges the bus capacitor of the battery-side bus through the current-limiting resistor, and the battery-side bus voltage rises from 0V. During pre-charging, the pre-charging current decays exponentially according to the battery-side bus capacitor value and the current-limiting resistor value, and the charging rate gradually decreases as the voltage rises. Specifically, when the battery-side bus voltage reaches 95% of the external power supply voltage, i.e., 11.4V, the pre-charging process ends. At this time, the difference between the battery-side bus voltage and the external power supply bus voltage is 0.6V, and the surge current corresponding to this difference is much smaller than the surge current when the switching switch is directly closed. Subsequently, the battery output switch is closed, connecting the battery to the battery-side bus. Then, the switching switch is closed to maintain the voltage difference between battery power and external power supply within a small range of 0.6 V. The physical layer power supply control module then disconnects the external power input, completing the switch from external power supply to battery power. During the switching process, the load-side voltage does not experience a significant drop, achieving a smooth switching operation. After the external power supply is restored, the external power input branch is first closed to restore the external power bus to its rated voltage. Then, the switching switch and battery output switch are disconnected, and the load is powered by the external power supply via the external power bus.

[0063] The following example of a complete power supply switching process further illustrates the technical solution of this invention. The rated voltage of the external power supply is set to 12 V, the preset undervoltage threshold is set to 90% of the rated voltage (10.8 V), the first preset threshold is set to 3 times, the second preset threshold is set to 10 times, the rated total capacity of the battery is 10 Ah, the initial battery state of charge is 85%, and the preset safe charge threshold is 20%. First, voltage anomaly detection is performed by sampling the external power supply bus voltage at a sampling frequency of 1 kHz. The fast window length is set to 10 ms, and the slow window length is set to 500 ms. For example, at a certain moment, the bus voltage briefly drops below 10.8 V twice within 200 ms before returning to the rated value. The number of abnormal events counted in the slow window is 2, resulting in a jitter count of 2. Since the jitter count is less than the first preset threshold of 3 times, it is determined to be an intermittent disconnection. A supercapacitor provides short-term energy support to maintain the external power supply. Subsequently, the bus voltage dropped again and failed to recover for an extended period. The slow window statistics recorded 12 abnormal events (the slow window contained 12 consecutive drops covering 12 fast windows, or 120 ms), resulting in a jitter count of 12. Since the jitter count was not less than the second preset threshold of 10, it was determined to be a genuine fault. Before the power assessment and pre-charging synchronization were completed, the supercapacitor maintained power supply to the load. Next, a power assessment was performed. The communication module was currently in a locked state, with a corresponding discharge coefficient of 2.0 A and a dynamic discharge rate of 2.0 A. The communication module switched states 8 times within a 500 ms time window. Looking up the table, the discharge loss compensation value was 0.3 A, and the predicted power consumption value was 2.0 A plus 0.3 A equals 2.3 A. Multiplying the predicted power consumption value of 2.3 A by the prediction time of 1 hour yielded a power consumption of 2.3 Ah, representing 23% of the rated total capacity of 10 Ah. Subtracting 23% from the remaining capacity baseline of 85% yielded an estimated remaining power of 62%. Since the expected remaining battery power of 62% is greater than the preset safe battery power threshold of 20%, and the jitter count of 12 times is not less than the second preset threshold of 10 times, the battery power supply link is determined and switched to battery power. Before switching, the battery-side bus is pre-charged, gradually increasing the battery-side bus voltage from 0V to 95% of the external power supply voltage of 12V, i.e., 11.4V. Then, the battery output switch and the switching switch are closed, and the external power input is disconnected, achieving a smooth switching. After the switching is completed, the battery output voltage is measured at 11.2V, and the load receiving voltage is measured at 11.0V, with a difference of 0.2V, which is within the preset compensation range, confirming that the power supply status switch is complete.

[0064] The following example illustrates the end-to-end nature of the above technical solution using the terminal's operation during a complete power supply fluctuation cycle. The external power supply rated voltage is set to 12 V, the preset undervoltage threshold to 10.8 V, the fast window length to 10 ms, the slow window length to 500 ms, the first preset threshold to 3 cycles, the second preset threshold to 10 cycles, the hysteresis to 1 cycle, the power reduction level boundary to 6 cycles, and the preset safe charge threshold to 20%. The battery pack consists of three lithium batteries connected in series, with an open-circuit voltage of 11.85 V, a rated total capacity of 10 Ah, and an initial state of charge of 85%. At the start of the cycle, the terminal is normally powered by the external power supply, the supercapacitor is charged to its rated voltage of 12 V, and both the battery output switch and the switching switch remain open.

[0065] First, the external power supply experienced intermittent power drops due to poor connector contact. Within a certain slow window, two abnormal events were recorded, with a jitter count of 2, which was below the first preset threshold of 3, thus indicating intermittent disconnection. During the power drop, the supercapacitor discharged to the load, maintaining the load voltage above 10.8V and sustaining external power supply. During the interval between power drops, the supercapacitor resumed charging. Subsequently, the fluctuations intensified, and five abnormal events were recorded within the slow window. The jitter count of five events was not less than the first preset threshold of 3, indicating that the system entered a level 1 power reduction phase, shutting down the peripheral display module and reducing the average power consumption from 30W to 26W. The fluctuations continued, and eight abnormal events were recorded within the slow window. The jitter count of eight events was not less than the power reduction level boundary of 6 events and was less than the second preset threshold of 10 events, indicating a level 2 power reduction phase. The high-speed data transmission module was then shut down, and the average power consumption dropped to 20W. After a brief period of fluctuation, the slow window statistics showed 3 abnormal events. The jitter count of 3 events caused the power reduction level to revert from level 2 to level 1. However, the 3 events were not lower than the exit threshold of 2 events. The power supply mode remained in the state of reducing terminal power consumption without reverting to maintaining external power supply. This reflects the suppression of repeated switching of power supply mode near the critical number of jitter events by the hysteresis judgment.

[0066] Next, the external power adapter failed, and the bus voltage remained below the preset undervoltage threshold. Slow window statistics, including the drop test period, showed 12 abnormal events (continuous drops covering 12 fast windows, i.e., 120 ms). The jitter count was 12 times, not less than the second preset threshold of 10 times and higher than the exit threshold of 8 times for battery power withdrawal. This was determined to be a genuine fault, and the decision to switch back to battery power was maintained. During the transition period between the decision and the completion of the switch, the supercapacitor provided 20 W of power at reduced power consumption. A power assessment was performed: the communication module was in a locked state, corresponding to a discharge coefficient of 2.0 A. The number of state transitions within the 500 ms time window was 8. A discharge loss compensation value of 0.3 A was obtained from the table, and the predicted power consumption was 2.3 A. Multiplying this by the predicted time of 1 hour yielded a power consumption of 2.3 Ah, accounting for 23% of the rated total capacity of 10 Ah. This was deducted from the remaining capacity baseline of 85%, resulting in an estimated remaining usable power of 62%, which is greater than the preset safe power threshold of 20%. The safe power threshold was passed, and the battery power supply link was confirmed.

[0067] Then, a pre-charge synchronous switching is performed: The pre-charge switch is closed, and the battery charges the 100 μF bus capacitor through a 100 Ω current-limiting resistor for a time constant of 10 ms. The pre-charge target voltage is the smaller of the product of the external power supply voltage (12 V) and a preset ratio of 95%, and the difference between the battery pack terminal voltage (11.85 V) and a preset margin of 0.2 V, i.e., 11.4 V. The battery-side bus voltage rises exponentially, reaching above 11.4 V and approaching the battery terminal voltage after approximately 50 ms, at which point the pre-charge process ends. The battery output switch is closed, connecting the battery to the battery-side bus, with a voltage difference of approximately 0.1 V and a very small inrush current. The switching switch is then closed, maintaining the voltage difference between the battery-side bus and the external power supply bus within 0.6 V, with an inrush current far smaller than when the switching switch is directly closed. The physical layer power supply control module disconnects the external power input, and the load is switched to battery power. No significant voltage drop occurs at the load end during the switching process. After the switchover, the battery output voltage was measured at 11.2 V and the load receiving voltage at 11.0 V. The difference of 0.2 V was within the preset compensation range of 0 V to 0.5 V, confirming that the power supply status was normal. The 11.2 V battery output voltage was the load voltage resulting from the open-circuit voltage of 11.85 V, after the load current caused a voltage drop across the battery's internal resistance and the power supply line. During battery power supply, the supercapacitor was recharged from the bus to its rated voltage of 12 V.

[0068] Finally, the external power supply returns to normal. After the external power supply bus voltage rises back to the rated value of 12V and remains for a preset recovery confirmation time (exemplarily 100 ms), the difference between it and the load bus voltage under battery power does not exceed 0.8V. When the external power supply input branch is closed, this voltage difference is limited by the battery's internal resistance and the external power supply's source impedance, resulting in limited transient current and uninterrupted load power supply. Subsequently, the switching switch and battery output switch are disconnected, and the load current smoothly transfers to the external power supply. The load voltage rises from 11.2V under load to 12V, completing the transition from battery power to external power supply. It should be noted that if the estimated remaining available power read when a real fault is determined is lower than the preset safe power threshold by 20%, for example, only 15%, then the switch to battery power is not executed. Instead, the battery management chip performs limited discharge on the core data storage module within a range not lower than the minimum safe discharge depth to complete the data write-to-disk operation. The terminal enters a low-power standby state, waiting for the external power supply to recover to above the preset undervoltage threshold before resuming normal power supply. The aforementioned complete cycle covers all decision branches, including maintaining external power supply, reducing terminal power consumption, switching to battery power, and reverse fallback, as well as the complete working process with hysteresis determination and safe power gating.

[0069] For example, Figure 4 This is a schematic diagram of the pre-charge synchronization switching. After pre-charging begins, the battery-side bus voltage gradually increases from 0 V with a time constant of 10 milliseconds, exceeding the pre-charge target voltage of 11.4 V before the battery output switch is closed. Subsequently, the load-side received voltage smoothly transitions from 12 V under external power supply to 11.0 V under battery power supply, remaining above the preset undervoltage threshold of 10.8 V throughout the switching process. After the switching stabilizes, the difference between the battery output voltage and the load-side received voltage is 0.2 V, falling within the preset compensation range of 0 V to 0.5 V.

[0070] In summary, this invention discloses a dual-mode power supply adaptive control method based on lithium batteries. This invention distinguishes between intermittent power supply disconnections and actual faults by using a multi-window sliding comparator and jitter counting. It predicts the expected remaining available power through load adaptive ampere-hour integration, and then combines jitter counting and remaining power to perform hysteresis-based hierarchical switching decisions. Finally, it achieves seamless switching of power supply modes through pre-charging synchronous switching.

[0071] The second embodiment of the present invention provides a dual-mode power supply adaptive control system based on a lithium battery, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the dual-mode power supply adaptive control method based on a lithium battery as described above.

[0072] It should be noted that the dual-mode power supply adaptive control system based on lithium battery provided in this embodiment of the invention is used to execute all the process steps of the dual-mode power supply adaptive control method based on lithium battery in the above embodiment. The working principle and beneficial effect of the two are one-to-one, so they will not be described again.

[0073] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A dual-mode power supply adaptive control method based on lithium batteries, characterized in that, include: A continuous voltage sequence of an external power supply is obtained. Anomalies are detected in the continuous voltage sequence using a multi-window sliding comparator. Instantaneous voltage changes in the continuous voltage sequence are captured and the number of abnormal events per unit time is counted to obtain a jitter count. Obtain battery state of charge and communication load data, perform load-adaptive ampere-hour integration on the battery state of charge and the communication load data, and obtain the estimated remaining available power. Based on the jitter count and the estimated remaining available power, a hysteresis-based hierarchical switching decision is made to obtain a power supply switching decision; Based on the power supply switching decision, the battery-side bus is pre-charged, and a power supply switching control command is sent to the physical layer power supply control module through pre-charging synchronous switching to obtain the corresponding power supply status.

2. The dual-mode power supply adaptive control method based on lithium battery according to claim 1, characterized in that, The step of performing anomaly detection on the continuous voltage sequence using a multi-window sliding comparator, capturing instantaneous voltage changes in the continuous voltage sequence, and counting the number of abnormal events per unit time to obtain a jitter count includes: Set up fast and slow windows; The instantaneous voltage changes of the continuous voltage sequence are captured through the fast window; The number of abnormal events is obtained by counting the number of abnormal events per unit time using the slow window. The jitter count is determined by a jitter counter based on the number of abnormal events.

3. The dual-mode power supply adaptive control method based on lithium battery according to claim 1, characterized in that, The step of making a hysteresis-based tiered switching decision based on the jitter count and the estimated remaining available power to obtain a power supply switching decision includes: If the jitter count is less than the first preset threshold, then the external power supply is maintained; If the jitter count is not less than the first preset threshold and less than the second preset threshold, then the terminal power consumption is reduced to maintain the external power supply. If the jitter count is not less than the second preset threshold, then switch to battery power to obtain the power supply switching decision; The entry and exit of the power supply mode are determined using different thresholds, and the corresponding thresholds are selected based on the current power supply status. To switch from maintaining external power supply to reducing terminal power consumption, the jitter count must not be less than the first preset threshold. To switch from reducing terminal power consumption back to maintaining external power supply, the jitter count must be lower than the first exit threshold, which is lower than the first preset threshold. To switch from reducing terminal power consumption to battery power supply, the jitter count must not be less than the second preset threshold. To switch from battery power supply back, the jitter count must be lower than the second exit threshold, which is lower than the second preset threshold.

4. The dual-mode power supply adaptive control method based on lithium battery according to claim 1, characterized in that, The step of performing load-adaptive ampere-hour integration on the battery state of charge and the communication load data to obtain the estimated remaining available power includes: Determine the corresponding discharge coefficient based on the operating status of the communication module; Based on the discharge coefficient, the dynamic discharge rate is calculated using a state-by-state discharge model; wherein the state-by-state discharge model uses the operating state of the communication module as segmented variables, and each operating state corresponds to a constant discharge rate. The estimated remaining available power is calculated by ampere-hour integration based on the dynamic discharge rate.

5. The dual-mode power supply adaptive control method based on lithium battery according to claim 4, characterized in that, The step of calculating the estimated remaining available power capacity by integrating the ampere-hours based on the dynamic discharge rate includes: Based on the communication load data, load fluctuation characteristics are extracted; based on the load fluctuation characteristics and the dynamic discharge rate, a discharge loss compensation value is determined; based on the dynamic discharge rate and the discharge loss compensation value, a power consumption prediction value is calculated. The predicted power consumption is multiplied by a preset prediction time to obtain the power consumption; the proportion of the power consumption to the battery's rated total capacity is subtracted from the remaining capacity baseline to obtain the estimated remaining power.

6. The dual-mode power supply adaptive control method based on lithium battery according to claim 1, characterized in that, The process of pre-charging the battery-side bus involves sending a power supply switching control command to the physical layer power supply control module via pre-charging synchronization to obtain the corresponding power supply status, including: The battery-side bus is pre-charged to obtain the pre-charged battery-side bus voltage. Based on the pre-charged battery-side bus voltage, close the switching switch to obtain the battery power supply link; Disconnect the external power input and determine the corresponding power supply status based on the battery power supply link.

7. The dual-mode power supply adaptive control method based on lithium battery according to claim 3, characterized in that, The step of switching to battery power if the jitter count is not less than the second preset threshold includes: If the estimated remaining available power is greater than the preset safe power threshold, then the battery power supply link is determined, and the system switches to battery power supply based on the battery power supply link. If the estimated remaining available power is not greater than the preset safe power threshold, the system will not switch to battery power, but will gradually reduce the terminal power consumption and perform limited discharge on the core data storage module to complete the core data write-to-disk operation. Then, it will enter a low-power standby state and resume normal power supply after the external power supply recovers to above the preset undervoltage threshold.

8. The dual-mode power supply adaptive control method based on lithium battery according to claim 3, characterized in that, The step of reducing terminal power consumption to maintain external power supply if the jitter count is not less than the first preset threshold and less than the second preset threshold includes: The power reduction level is determined based on the jitter count; Based on the power reduction level, the peripheral function modules of the corresponding power reduction level in the terminal are turned off to obtain the power supply power of the terminal after power reduction. Based on the reduced power consumption of the terminal, the external power supply is maintained.

9. The dual-mode power supply adaptive control method based on lithium battery according to claim 1, characterized in that, The method further includes: During normal power supply, the supercapacitor is charged by an external power source; during voltage drop when the jitter count is less than a first preset threshold, the supercapacitor discharges to the load to maintain the load voltage at a level not lower than a preset undervoltage threshold; during the transition period after a real fault is determined and before switching to battery power supply, the supercapacitor maintains power supply to the load at the terminal power after reducing terminal power consumption.

10. A dual-mode power supply adaptive control system based on a lithium battery, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the dual-mode power supply adaptive control method based on lithium battery as described in any one of claims 1 to 9.

Images