Railway container tracking terminal multi-source wake-up power supply control system

Abstract

The application discloses a railway container tracking terminal multi-source wake-up power supply control system, which comprises a main control module and a power supply module, an event collection module and a task execution module which are electrically connected with the main control module respectively; the main control module is used for confirming whether the wake-up signal collected by the event collection module is a valid wake-up event, if yes, controlling the terminal to enter a full activation state, otherwise, keeping the original state; in the full activation state, exchanging state characteristic data with adjacent terminals through a short-distance communication unit, when the state characteristic data meets a preset same-column consistency condition, establishing a same-column transport cluster; determining the cluster role of the terminal in the same-column transport cluster based on the state characteristic data, executing a cooperative task through task proxy and information sharing among cluster members, and combining battery state information collected by a battery state detection module to determine a task execution level, and performing on-demand power supply control on the task execution module of each terminal itself. The application significantly reduces the group power consumption of the container terminal.

Description

Technical Field

[0001] This invention relates to the field of container terminal management technology, and in particular to a multi-source wake-up power supply control system for railway container tracking terminals. Background Technology

[0002] With the development of railway container transport, multimodal transport (road-rail-water), and cross-border freight train services, railway containers face higher demands for location tracking, status monitoring, anomaly identification, and long-term endurance in loading, unloading, traction, marshalling, long-distance railway operation, station stops, yard storage, and inventory inspection. Tracking terminals are typically installed on the outside or near the container, relying on primary batteries or batteries with limited capacity for extended periods, lacking stable external power supply. Therefore, it is necessary to minimize average power consumption while ensuring responsiveness to critical events.

[0003] Existing IoT tracking terminals typically employ fixed-period wake-up, fixed-period positioning, and fixed-period upload modes, or simple motion wake-up, RTC wake-up, module-on-demand power-up, and communication failure buffering and retransmission strategies. While these solutions can reduce standby power consumption to some extent, they still have shortcomings in railway container scenarios: First, multiple container tracking terminals often exist on the same train during railway transportation. Each terminal exhibits highly consistent start-up, vibration, and position change characteristics. If each terminal independently searches for satellites, networks, and uploads data, it will generate a large number of repetitive, high-power-consumption operations. Second, railway containers are prone to GNSS satellite search timeouts and cellular registration failures in environments with metal containers, train obstructions, station stacking, and weak networks. Long-term retries by a single terminal will result in significant power waste. Third, relying solely on the acceleration signal of a single terminal makes it difficult to distinguish between train braking, track impacts, partial loading / unloading, and abnormal impacts in a single container, easily leading to false alarms or missed alarms.

[0004] The applicant proposed solutions to the above problems. Summary of the Invention

[0005] The foreign language meanings involved in this application are: BLE: Bluetooth Low Energy; ESD: Electrostatic Discharge; FRAM: Ferroelectric Random Access Memory; Flash: Flash memory, a type of data storage device; GNSS: Global Navigation Satellite System; GPS: Global Positioning System; GPIO: General-purpose Input / Output; I2C: Inter-Integrated Circuit, a communication bus; Ioff: Power-off protection; LoRa: Long Range Radio, long-range wireless communication; LTE Cat.1: Performance level for IoT wireless communication devices; MCU: Microcontroller Unit; MEMS: Micro-Electro-Mechanical System; NB-IoT: Narrowband Internet of Things; OCV: Open Circuit Voltage, battery open circuit voltage; RFID: Radio Frequency Identification; RMS: Root Mean Square; RTC: Real-Time Clock; SOC: State of Charge, the state of battery charge; SPI: Serial Peripheral Interface; Sub-GHz: Wireless communication technology operating at frequencies below 1 GHz; STOP: STOP mode (MCU low power mode); STANDBY: STANDBY mode (MCU low power mode); UWB: Ultra-Wideband; UART: Universal Asynchronous Receiver / Transmitter, a serial communication protocol.

[0006] To address the technical problems existing in the background art, this invention proposes a multi-source wake-up power supply control system for railway container tracking terminals.

[0007] The multi-source wake-up power supply control system for railway container tracking terminals proposed in this invention includes a main control module and power supply module, event acquisition module, positioning module, communication module, and sensor acquisition module electrically connected thereto; the communication module includes a long-range communication unit and a short-range communication unit; the main control module is used for: Confirm whether the wake-up signal collected by the event acquisition module is a valid wake-up event. If it is, control the terminal to enter the fully activated state; otherwise, maintain the original state. In the fully active state, status feature data is exchanged with neighboring terminals through the near-field communication unit. When the status feature data meets the preset co-column consistency condition, a co-column transport cluster is established. Based on the state feature data, the terminal's role within the same transport cluster is determined. Collaborative tasks are executed through task delegation and information sharing among cluster members. The task execution level is determined in conjunction with battery status information, and power supply control is applied to the task execution modules of each terminal as needed. After the task is completed, control the terminal to enter a shallow sleep state or a deep sleep state.

[0008] Furthermore, the event acquisition module includes an acceleration sensing unit, an RTC timing unit, and an RFID triggering unit. When the acceleration sensing unit outputs an acceleration wake-up signal, the main control module enters the wake-up confirmation stage. If the acceleration change in the acceleration wake-up signal exceeds a preset acceleration change threshold and the duration exceeds a preset duration, a valid wake-up event is confirmed; otherwise, it is determined to be an invalid triggering event. When the RTC timing unit outputs a long-period RTC timing signal, or outputs a short-period RTC timing signal while the terminal is in a shallow sleep state, the main control module confirms a valid wake-up event. When the RFID triggering unit detects a valid external stimulus or a valid RFID wake-up command, the main control module confirms a valid wake-up event.

[0009] Furthermore, after the main control module completes the task, if it is currently in the same transport cluster and the acceleration wake-up signal determines that the conditions for continuous railway movement are met and there are no tasks to be processed, the control terminal enters a shallow sleep state; if, within the preset observation window, the acceleration wake-up signal determines that the conditions for continuous railway movement are not met and there are no tasks to be processed, the control terminal enters a deep sleep state.

[0010] Furthermore, for terminals participating in the establishment of the same transport cluster, the status characteristic data includes at least terminal identifier, timestamp, battery status information, positioning quality, communication quality, cache summary, abnormal event data, alarm event data, most recent location reference, RFID station anchor point, and acceleration characteristic summary; the acceleration characteristic summary includes at least one or more of the following: start-stop time, root mean square value of vibration, main frequency components, impact peak value, duration, and direction change.

[0011] Furthermore, within a preset short-range communication window after confirming a valid wake-up event, the co-column consistency condition includes short-range communication link quality higher than a preset link quality threshold, and at least two of the following conditions: The acceleration start-stop time difference of multiple terminals within the same time window is less than a preset time difference threshold, and the correlation coefficient of acceleration feature sequences within the same time window is greater than a preset correlation threshold; The difference in the root mean square values ​​of vibration of multiple terminals is less than the preset root mean square difference threshold and the difference in the main frequency components is less than the preset frequency difference threshold. The distance between the nearest location references of multiple terminals is less than a preset distance threshold; Multiple terminals correspond to the same RFID station anchor point, the same station area identifier, or the same railway line corridor.

[0012] Furthermore, the cluster head is generated based on the status feature data of each terminal, and the cluster head organizes the status feature data to form an aggregated data frame containing a common location field and a member difference field, and uploads it to the background platform through the long-distance communication unit; after receiving the upload confirmation information from the cluster head, the cluster member marks the corresponding cached data as uploaded by proxy and keeps its own long-distance communication unit powered off; and when the cluster head obtains a valid positioning result, the cluster member uses the positioning result as the position reference for this round and keeps its own positioning module powered off.

[0013] Furthermore, if any cluster member in the same transport cluster detects abnormal event data based on a wake-up signal, it sends the abnormal event data to the cluster head. The cluster head then requests other cluster members to return an acceleration feature summary within the corresponding time window through the near-field communication unit. If the acceleration features of multiple cluster members are synchronized in time and their amplitude distribution satisfies the whole-column disturbance feature, the cluster head lowers the upload priority. If only the abnormal terminal itself or some nearby terminals detect an impact feature exceeding the impact threshold, the cluster head raises the upload priority.

[0014] Furthermore, the main control module implements on-demand control of the task execution module through the power supply control module. The power supply control module includes an independent load switch connected between the power supply module and the task execution module, and an isolation current limiting unit set on the signal line of the task execution module. The independent load switch is used to provide on-demand power to the task execution module, and the isolation current limiting unit is used to switch the signal line connected to the task execution module to a high impedance state or a pull-down holding state when the power is off.

[0015] Furthermore, the main control module calculates the State of Charge (SOC) based on the battery status information, and determines the task execution level based on the comparison results of the SOC with the preset first threshold, second threshold, and third threshold, wherein the first threshold > the second threshold > the third threshold; When SOC ≥ the first threshold, the terminal takes priority as the cluster head, the task execution level is the standard level, and the task execution module is started as needed. When the second threshold ≤ SOC < the first threshold, the task execution level is reduced, which slightly restricts at least one of the following: the usage time, usage frequency, and workload of the task execution module; When the third threshold ≤ SOC < the second threshold, the terminal is not a cluster head, the task execution level is restricted, and the usage time, frequency of use, and workload of the task execution module are severely restricted. When SOC < the third threshold, the terminal exits the same transport cluster, the task execution level is set to the protection level, and only the task execution modules required for abnormal event detection, location acquisition and alarm reporting are enabled.

[0016] Furthermore, the main control module estimates the SOC by combining ampere-hour integration and open-circuit voltage calibration. When the power module is a primary battery, the main control module pauses open-circuit voltage calibration or sets the open-circuit voltage calibration weight to be lower than the ampere-hour integration weight when it detects that the battery open-circuit voltage is in a flat discharge plateau region. When it detects that the battery open-circuit voltage has entered the inflection point range at the end of the discharge period, it increases the open-circuit voltage calibration weight.

[0017] This invention utilizes the shared trajectory, start / stop, vibration, and station operation characteristics of terminals in railway container train transport. Based on the consistency condition of the same train, it determines whether to establish a transport cluster. A cluster head is elected and roles within the cluster are assigned through factors such as State of Charge (SOC). Cluster members exchange data via low-power short-range communication units. The cluster head acts as an agent for location and data upload tasks, avoiding repetitive satellite search, location, and upload operations by individual terminals on the train, significantly reducing the power consumption of the entire terminal group. Furthermore, the cluster head can combine data from multiple cluster members to comprehensively assess a single abnormal terminal, improving the accuracy of anomaly detection.

[0018] Furthermore, this application links SOC hierarchical management, on-demand power supply control, terminal sleep strategies, and cluster status. Terminals within the same transport cluster enter a shallow sleep state after completing their tasks, quickly communicating with the cluster head via short-cycle RTC wake-up to maintain cluster relationships. Terminals that lose cluster connections or do not meet the conditions for continuous operation enter a deep sleep state, re-attempting cluster establishment after being woken up via long-cycle RTC wake-up. Through this low duty cycle collaborative mechanism, terminals maintain intra-cluster collaborative relationships while remaining in sleep mode most of the time, achieving an optimal balance between power consumption and response speed. Attached Figure Description

[0019] Figure 1 This is a block diagram of the overall structure of the system of the present invention; Figure 2 This is a schematic diagram of the task execution process of the present invention. Detailed Implementation

[0020] The technical solution of the present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0021] like Figure 1 As shown, the multi-source wake-up power supply control system for railway container tracking terminals proposed in this application is suitable for long-term location tracking and status monitoring of railway containers in application scenarios such as cross-border transportation, multimodal transport, station turnover, yard static storage, and inventory management.

[0022] The system includes a main control module and, electrically connected to it, a power supply module, a battery status detection module, an event acquisition module, a positioning module, a communication module, a sensor acquisition module, a power supply control module, and a storage module. The main control module can use a microcontroller with STOP, STANDBY, or other low-power modes; the power supply module preferably uses a primary battery or a low-self-discharge battery pack; the battery status detection module collects battery status information such as voltage, current, and temperature; the positioning module includes a GNSS satellite positioning unit and a base station-assisted positioning unit; the communication module includes a long-range communication unit (cellular communication unit) and a short-range communication unit, where the short-range communication unit can use BLE, Sub-GHz, LoRa, UWB, or other low-power short-range communication methods; the long-range communication unit can use 4G, 5G, NB-IoT, LTE Cat.1, or public or private network communication methods suitable for railway container tracking services; the storage module can use Flash, FRAM, or other low-power non-volatile memory.

[0023] The event acquisition module includes an accelerometer unit, an RTC timing unit, and an RFID triggering unit. The accelerometer unit is preferably a low-power triaxial MEMS device with interrupt output capability. The RTC timing unit can be an independent low-power real-time clock chip or a low-power RTC peripheral integrated into the main control module. The RFID triggering unit is used for low-power wake-up in scenarios such as station inventory, near-field inspection, on-site confirmation, or manual maintenance. The sensor acquisition module and the event acquisition module are set up separately. The event acquisition module is used for normally-on low-power monitoring, while the sensor acquisition module is used for task-oriented data acquisition such as temperature, humidity, shock, and vibration. The sensor acquisition module, along with the positioning module and communication module, serves as a task execution module, powered on as needed by the power supply control module.

[0024] Example 1: The wake-up and state switching of a single terminal will be explained in detail below.

[0025] In this embodiment, the terminal has a fully active state, a shallow sleep state, and a deep sleep state.

[0026] Specifically, the fully active state is used to perform at least one of the following tasks: event determination, status acquisition, location acquisition, network registration, data packaging, and data upload. The shallow sleep state is used to retain the main control module's operating context, RTC timing capabilities, and low-power wake-up paths for listening to wake-up events. The deep sleep state is used to shut down the positioning module, communication module, and high-power sensor acquisition module, retaining only the low-power wake-up paths for listening to wake-up events, namely the RTC timing wake-up path, the RFID trigger path, and the acceleration trigger event listening path.

[0027] Specifically, to balance low-power monitoring and task acquisition, this application designs the event acquisition module and the sensor acquisition module separately. The event acquisition module is used to implement the low-power monitoring path related to wake-up event triggering, while the sensor acquisition module is used to implement environmental monitoring tasks. Specifically, the accelerometer unit, RTC timing unit, and RFID trigger unit used for motion monitoring constitute a normally open low-power event acquisition path; the sensor acquisition unit used for task-oriented data acquisition such as temperature, humidity, shock, and vibration can be powered on as needed by the main control module through the power supply control module. This allows for continuous monitoring of necessary events in deep sleep mode while avoiding the additional power consumption caused by all sensors being powered on for extended periods.

[0028] Specifically, the main control module confirms whether the wake-up signal collected by the event acquisition module is a valid wake-up event. If so, the control terminal enters the fully active state; otherwise, it remains in the original sleep state.

[0029] Specifically, when the acceleration sensing unit outputs an acceleration wake-up signal, the main control module enters the wake-up confirmation stage. If it is determined that the acceleration change in the acceleration wake-up signal exceeds the preset acceleration change threshold and the duration exceeds the preset continuous determination time, then a valid wake-up event is confirmed; otherwise, it is determined to be an invalid triggering event. When the RTC timing unit outputs a long-cycle RTC timing signal, or when the terminal outputs a short-cycle RTC timing signal while in a shallow sleep state, the main control module confirms that a valid wake-up event has been formed. When the RFID triggering unit detects a valid external stimulus or a valid RFID wake-up command, the main control module confirms that a valid wake-up event has been formed.

[0030] Upon initial power-up, the main control module performs initialization, including power stability detection, clock configuration, storage module read / write testing, bus self-test, acceleration interrupt configuration, RTC cycle loading, RFID interface detection, initial battery SOC setting, and availability detection of the positioning and communication modules. If the main control module reads the installation / debugging mode flag, intensive sampling configuration parameters, or remotely issued debugging commands, it can control the terminal to enter a shallow sleep state after initialization to quickly respond to short-cycle detection tasks. If the above configuration is not read and the acceleration wake-up signal does not meet the conditions for continuous railway movement (i.e., the terminal is in a stationary scenario), it controls the terminal to enter a deep sleep state.

[0031] In this embodiment, the terminal is only woken up from deep sleep when the RTC timing unit outputs a long-cycle RTC timing signal. The short-cycle RTC is mainly used for rapid detection in shallow sleep or intermittent positioning in railway operation scenarios. It is not used as a regular wake-up source in deep sleep to avoid being frequently woken up by short-cycle timing during long periods of inactivity, which would result in wasted power consumption.

[0032] In some examples, when the main control module detects an acceleration wake-up signal, it enters the wake-up confirmation phase, subsequently reading the acceleration change, duration of change, direction information, and necessary timestamp information. If a valid wake-up event is determined, the control terminal enters a fully active state; if it is only a short-term disturbance or occasional vibration, it is determined to be an invalid trigger event, a slight disturbance flag is recorded, and the terminal returns to its original shallow or deep sleep state.

[0033] In some examples, when the main control module detects that the wake-up signal is a long-cycle RTC timing signal, the main control module confirms that a valid wake-up event has been formed and controls the terminal to enter the fully active state from the deep sleep state or shallow sleep state to perform periodic inspection tasks; when the main control module detects that the wake-up signal is a short-cycle RTC timing signal and the current terminal is in a shallow sleep state, the main control module confirms that a valid wake-up event has been formed and controls the terminal to enter the fully active state from the shallow sleep state to perform simplified detection, short-cycle positioning or status update tasks.

[0034] RFID triggering units are used to meet the application requirements of station inventory, manual inspection, and near-field identification. The RFID triggering unit can use near-field induction, detecting the radio frequency excitation field generated by external inventory equipment and then generating a wake-up interrupt through detection, comparison, or preamble identification; or it can use low-power active reception, generating a wake-up signal upon receiving a valid wake-up command. When the RFID triggering unit detects a valid external excitation or a valid wake-up command, the main control module confirms the formation of a valid wake-up event, and the control terminal enters a fully active state to execute tasks such as site confirmation, inventory confirmation, near-field management, location acquisition, or status reporting. Furthermore, when high-priority buffered data reaches the retransmission condition and the communication quality meets preset conditions, it can also serve as a communication task trigger condition, and the main control module executes buffer retransmission.

[0035] After completing the task, the main control module reassesses the operating scenario, SOC, communication status, cached task status, and the next task cycle. If the terminal is marked as a railway operating scenario based on the acceleration wake-up signal during this task cycle, and the next data acquisition, positioning, or upload cycle has not yet arrived, and there are no high-priority tasks requiring immediate processing, the terminal enters a shallow sleep state. If, within the preset observation window, the acceleration wake-up signal indicates that the conditions for continuous railway movement are not met, and there are no high-priority cached tasks, the terminal enters a deep sleep state. Here, the motion scenario flag is used for the sleep fallback strategy after the task is completed and is not used as a trigger for repeatedly entering the fully active state under normal continuous vibration conditions.

[0036] Specifically, in the fully active state, the main control module controls the power supply to the task execution module on demand through the power supply control module. Specifically, the power supply control module includes independent load switches connected between the power module and each sub-unit of the positioning module, communication module, and sensing acquisition module, as well as isolation current limiting units installed on the signal lines of the positioning module, communication module, and sensing acquisition module. Existing technologies only disconnect the power supply branch when the task execution module is powered off, without synchronously processing signal lines such as UART, I2C, SPI, or GPIO. This easily leads to backflow leakage paths through the signal lines, affecting low-power performance. In this application, the independent load switches are used to control the switching power supply to the task execution module. The isolation current limiting units are at least one of the following: a bus switch connected in series on the UART, I2C, SPI, or GPIO signal lines; a level converter with high impedance or Ioff function; and a digital isolator. These units are used to switch the signal lines connected to the task execution module to a high-impedance state or a pull-down holding state when power is off, thereby reducing leakage current through the signal lines. After the task is completed, the main control module first configures the output pin of the task execution module to high input impedance, then cuts off the power supply branch of the task execution module through an independent load switch, and cuts off or limits the leakage path of the signal line through the signal line isolation current limiting unit. This operation sequence can simultaneously block the power supply branch and the signal line backflow path, preventing the task execution module from being reverse-powered through the ESD diode, bus pull-up or control line after power failure.

[0037] Furthermore, unlike the traditional approach where all modules power on simultaneously upon system wake-up, this embodiment uses a power supply control module to start and stop task execution modules on demand, based on current task requirements, rather than powering on all modules. For example, if the current task is only short-cycle status detection, the main control module only powers on the sensor acquisition module via the power supply control module; if uploading is required, the cellular communication unit is activated to perform network search and registration; if the network is unavailable, communication stops and data is cached after a limited number of attempts. As another example, if the effective wake-up event is an acceleration-triggered event, a complete trajectory tracking task is typically triggered, controlling the positioning module to power on; if the effective wake-up event is a long-cycle timed event, periodic status checks and necessary location reporting are prioritized, controlling the sensor acquisition module and positioning module to power on; if the wake-up event is an RFID external trigger event, inventory confirmation, arrival confirmation, or near-field management related tasks are prioritized, and a location acquisition or status reporting can be triggered selectively. This preserves the engineering rationality of the task sequence while avoiding an absolute order that does not allow skipping any module.

[0038] To further reduce power consumption, in the fully active state, the main control module reads real-time battery status information such as voltage, current and temperature collected by the battery status detection module, estimates the SOC by combining ampere-hour integration and open-circuit voltage calibration, determines the task execution level, and then controls the corresponding task execution module to power on to complete the above tasks.

[0039] Specifically, the main control module determines the task execution level based on the comparison results between the SOC and the preset first threshold, second threshold, and third threshold, where the first threshold > the second threshold > the third threshold; When SOC ≥ the first threshold, the task execution level is the standard level, the task execution module starts as needed, and performs normal location, upload and appropriate data supplementation. When the second threshold ≤ SOC < the first threshold, the task execution level is reduced, which slightly restricts at least one of the following: the usage time, usage frequency, and workload of the task execution module, such as shortening the satellite search time, reducing the number of sensor samplings, and limiting the amount of supplementary data transmission. When the third threshold ≤ SOC < the second threshold, the task execution level is restricted, severely limiting the usage time, frequency, and workload of the task execution module. Priority is given to retaining anomaly detection, low-frequency location acquisition, and necessary communication, and the task content uploaded is limited to anomaly event data and alarm event data, i.e., only high-priority content is uploaded, reducing the upload of unnecessary content; When SOC < the third threshold, the task execution level is protected, and only the task execution modules required for abnormal event detection, location acquisition and alarm reporting are enabled, that is, only the low-power wake-up path for listening to wake-up events is retained.

[0040] In some examples, the first threshold, second threshold, and third threshold can be set to 60%, 30%, and 15%, respectively. These values ​​are only a set of non-limiting examples, and can be reset according to the battery system and mission environment in actual products.

[0041] Specifically, during the ampere-hour integration process, the main control module comprehensively considers the sampling current, the effective capacity after temperature correction, the self-discharge compensation, and the equivalent static loss current during the dormancy period. When the terminal is in a deep dormancy state, the duration reaches the preset static stability holding time, and the battery current is stably lower than the preset static stability current threshold, the ampere-hour integration result is corrected according to the pre-calibrated OCV-SOC-temperature correspondence to obtain the calibrated SOC estimate.

[0042] Specifically, the main control module calculates the ampere-hour integral result according to the following formula:

[0043] in, This represents the state of charge at the previous moment. To match the current battery temperature The relevant discharge utilization coefficient, The sampling current at the current moment, To match the current battery temperature The relevant equivalent self-discharge current, This represents the equivalent static loss current during the sleep period in the current integration cycle. The integration time interval, Current battery temperature The effective capacity below.

[0044] Furthermore, it can be obtained using any one or more of the following methods. and : Based on the self-discharge rate curve and temperature characteristic table provided by the battery manufacturer, the self-discharge rate is converted into the equivalent current per unit time; or, An empirical mapping table is established through long-term static testing of the prototype; or... When the terminal enters deep sleep and remains stable, the static loss current is calibrated based on low-speed, high-resolution current sampling.

[0045] By introducing the above compensation terms, the cumulative deviation caused by self-discharge, background noise, and static losses during long-term operation can be reduced.

[0046] Furthermore, the main control module estimates the battery SOC using a combination of ampere-hour integration and open-circuit voltage calibration. However, open-circuit voltage calibration is not performed under arbitrary conditions, but only after the static stability condition is met. Specifically, when the terminal is in deep sleep mode and continuously meets the preset static stability maintenance duration and preset static stability current threshold conditions, the terminal is in an approximately open-circuit equilibrium state, at which point the battery open-circuit voltage is read. And according to the pre-defined The calibration function is obtained from the three-dimensional correspondence of temperature or the interpolation function. After calculating the open-circuit voltage calibration value through the calibration function, the main control module adjusts the calibration value according to the preset fusion weights. Make corrections to avoid a single correction causing Dramatic change.

[0047] Specifically, the power module preferably uses a primary lithium battery. When the power module is a primary battery, it has a battery system with a relatively flat discharge platform, and the main control module does not use a full-range equal-weighted discharge system. Calibration strategy. When the battery open-circuit voltage is detected to be in a flat discharge plateau region, the main control module suspends open-circuit voltage calibration or sets the open-circuit voltage calibration weight to be lower than the ampere-hour integration weight, i.e. The estimated values ​​are primarily based on the ampere-hour integral results with temperature correction and static loss compensation. When the battery open-circuit voltage is detected to enter the inflection point range at the end of discharge, the open-circuit voltage calibration weight is increased to enhance the sensitivity of the end-of-discharge capacity estimation. In a set of non-limiting examples, the flat discharge plateau region... The correction weight can be set to 0 to 0.25, for the inflection point range at the end of the discharge. The calibration weight can be set to 0.4 to 0.8; the above weights can be adjusted according to the battery system, calibration data and business risk appetite.

[0048] Example 2, as Figure 2 As shown, this application elaborates on the coordinated power supply control of multiple terminals in the scenario of railway container train transportation.

[0049] The terminal does not continuously scan or broadcast, but only enters the fully activated state after the main control module confirms that a valid wake-up event has been formed. Within the preset short-range communication window after the fully activated state, it exchanges status feature data with the neighboring terminal through the short-range communication unit. When the status feature data meets the preset co-column consistency conditions, it establishes or maintains the co-column transport cluster.

[0050] The terminal can proactively open a near-field communication window to broadcast status feature data during the fully active state when the working cycle is woken up, or it can exchange status feature data when all terminals enter the fully active state due to an acceleration trigger event.

[0051] Specifically, for terminals participating in the same transport cluster, the status characteristic data includes at least the terminal identifier, timestamp, battery status information, positioning quality, communication quality, cache summary, abnormal event data, alarm event data, most recent location reference, RFID station anchor point, and acceleration characteristic summary; the acceleration characteristic summary includes at least one or more of the following: start-stop time, root mean square value of vibration, main frequency components, peak impact, duration, and direction change.

[0052] Specifically, the co-column consistency conditions include short-range communication link quality exceeding a preset link quality threshold, and at least two of the following conditions: If the acceleration start-stop time difference of multiple terminals within the same time window is less than a preset time difference threshold, and the correlation coefficient of the acceleration feature sequence within the same time window is greater than a preset correlation threshold, it can be judged that they constitute strong co-movement consistency. The difference in the root mean square (RMS) values ​​of vibration of multiple terminals is less than a preset RMS difference threshold and the difference in the main frequency components is less than a preset frequency difference threshold, so as to determine the consistency of railway motion vibration. If the distance between the nearest location references of multiple terminals is less than a preset distance threshold, or if multiple terminals correspond to the same RFID station anchor point, the same station area identifier, or the same railway line corridor, the consistency of the location and station can be determined.

[0053] After the same-column transport cluster is established, each terminal determines its intra-cluster role based on its status feature data.

[0054] Specifically, each terminal performs a score calculation based on the exchanged state characteristic data to elect a cluster head. The scoring factors include, but are not limited to, SOC, satellite positioning quality, cellular communication quality, cache task priority, the status of the most recent successful cellular upload, the number of times the cluster head has undertaken tasks recently, the remaining energy budget for this round, and whether it is in the protection block task execution level.

[0055] Cluster head election can be triggered when a co-location cluster is established, during periodic maintenance, or when the cluster head no longer meets the requirements. Specifically, to avoid cluster heads undertaking high-power tasks for extended periods, the system employs a cluster head rotation mechanism. When the cluster head's SOC falls below the cluster head retention threshold, its positioning quality falls below the positioning quality threshold, the number of cellular communication failures exceeds the threshold, its cache workload becomes too heavy, or a cluster head departure event is detected, the terminals within the cluster recalculate the cluster head score and elect a new cluster head.

[0056] Terminals with a SOC ≥ the first threshold and high satellite positioning and cellular communication quality are given priority as cluster heads and can participate in cluster head election, shared positioning, and aggregated uploads. Terminals with a second threshold ≤ SOC < the first threshold can serve as cluster heads and are given priority as ordinary cluster members. Terminals with a third threshold ≤ SOC < the second threshold are not considered cluster heads. Terminals with a SOC < the third threshold are removed from the same transport cluster to maintain low-power operation as much as possible. The main control module determines the task execution level based on the comparison results of SOC with the first, second, and third thresholds, and uses this task execution level to limit intra-cluster roles, positioning duration, communication retries, and buffer retransmission quantities.

[0057] For situations where trains are long and short-range communication is not possible, the terminal in this application uses a short-range communication unit to exchange status feature data, which can form a group of transport clusters within a range of two or three carriages, thus meeting the needs of train container tracking.

[0058] After the roles within the cluster are determined, cluster members perform collaborative tasks through task agents and information sharing.

[0059] Specifically, after a cluster head is elected, the cluster head broadcasts its intra-cluster role to the cluster members. It then organizes the state characteristic data previously sent by the cluster members, forming an aggregated data frame containing a common location field and member-specific fields. This aggregated data frame is then uploaded to the backend platform via a long-range communication unit. The common location field includes the cluster members' shared location, timestamp, railway operation scenario identifier, and cluster number. The member-specific fields include each terminal's terminal ID, SOC, anomaly flag, cache digest, and RFID status. After a successful upload, the cluster head sends an upload confirmation message to the cluster members. Upon receiving this confirmation, the cluster members mark the corresponding cached data as uploaded on their behalf, keep their own long-range communication units powered off, and lower the upload priority of that cached data. This eliminates the need for each cluster member to establish its own cellular connection for uploading, significantly reducing its own communication power consumption.

[0060] The cluster head performs satellite search and positioning. After obtaining valid satellite positioning results, base station-assisted positioning results, or RFID site anchor point information, it broadcasts the position reference, timestamp, positioning quality, and positioning method to the cluster members. When a cluster member determines that the position reference meets the time validity and positioning quality requirements, it uses the result as the position reference for this round and controls its own positioning module to remain powered off or only perform short-term verification, thereby significantly reducing its own satellite search power consumption.

[0061] During the execution of the aforementioned collaborative tasks, short-range collaborative communication between cluster members increases power consumption slightly, but its working time is short, data volume is small, and it is subject to SOC limitations. When shared positioning and aggregated upload are successful, cluster members can reduce GNSS satellite search time, reduce cellular registration times, postpone ordinary buffer retransmission, and reduce the amount of duplicate data uploads. Therefore, replacing the repeated high-power satellite search, network search, and upload by multiple terminals with a small amount of short-range summary communication creates a synergistic effect between clustering relationships, on-demand power supply, and SOC energy budget, thereby reducing total energy consumption from the perspective of group terminals.

[0062] After obtaining the location and upload results, cluster members, based on the task execution level determined by their own SOC, complete other tasks such as sensor data acquisition and control the power supply to the task execution modules as needed. Data completed under normal circumstances can be cached and has low priority, waiting for the next aggregation upload. If there is an abnormal event, it is sent to the cluster head, which decides whether to upload it urgently.

[0063] For example, if a cluster member detects abnormal event data, such as an impact event or abnormal movement event, based on a wake-up signal, it sends the abnormal event data to the cluster head. The cluster head then requests other cluster members to return acceleration feature summaries for the corresponding time window via a proximity communication unit. If the acceleration features of multiple cluster members are synchronized in time and their amplitude distribution meets the characteristics of a train-wide disturbance (such as train braking or track impact), the cluster head classifies the event as a train operation disturbance event and lowers its upload priority to reduce unnecessary cellular uploads. If only the abnormal terminal itself or some nearby terminals detect impact features exceeding the impact threshold (e.g., acceleration of 2g to 8g), the cluster head classifies it as a single-container abnormal impact event or a local loading / unloading event and generates a high-priority alarm frame, increasing its upload priority. This collaborative data judgment improves the reliability of abnormal event judgment within the same train transport cluster.

[0064] In weak network scenarios, if the cluster head successfully registers its cellular data, it will be prioritized for aggregation and uploading. If the cluster head fails to upload, cluster members will not immediately activate their long-distance communication units. Instead, they will decide whether to upload independently based on the SOC, the priority of cached data, and their own tasks. Regular periodic data, such as sensor data, is cached first, while abnormal impacts, abnormal departures from the cluster, RFID arrival confirmations, and the latest location data are uploaded or retained first. The storage module uses a circular queue to manage the cache. When space is insufficient, it prioritizes retaining abnormal alarm data, RFID management confirmation data, and the latest location data, overwriting the oldest regular periodic data.

[0065] Specifically, when a cluster member loses short-range communication with a terminal within the cluster during the effective period of the same transport cluster, and its acceleration characteristics, position reference, or RFID status is no longer consistent with the same transport cluster, the main control module of the abnormal terminal generates a cluster departure event. If the cluster departure event corresponds to a valid RFID arrival confirmation, a background unloading task, or a normal operation instruction, a normal unloading or station operation event is recorded; if it does not correspond to the above information, an abnormal cluster departure alarm is generated. For abnormal cluster departure alarms, the terminal can attempt independent positioning and cellular upload once according to the task execution level determined by the SOC, following the processing mode for a single terminal in Example 1; if communication fails, it is cached with the highest priority.

[0066] After completing tasks such as uploading or caching, the main control module of each terminal re-checks the current motion state, current SOC, communication state, cache task status, and the next scheduled task arrangement. If the terminal is currently in the same transport cluster and detects that the train is in continuous motion based on the wake-up signal and there are no pending tasks, it enters a shallow sleep state and shuts down the short-range communication unit to reduce power consumption. Cluster members, in the shallow sleep state, are woken up by short-cycle RTC timers to re-enter the active state, re-establish communication with the cluster head within a preset short-range communication window, obtain updated position references, and upload confirmation information. Through this low duty cycle coordination mechanism, the terminal maintains intra-cluster coordination while remaining in sleep mode most of the time, achieving a balance between power consumption and response speed.

[0067] If, within the preset observation window, the acceleration wake-up signal indicates that the conditions for continuous railway movement are not met and there are no tasks pending, the control terminal enters a deep sleep state. Before entering deep sleep, the main control module shuts down the positioning module, communication module, and sensor acquisition module, retaining only the low-power wake-up path for listening to wake-up events. When the terminal enters a fully active state via a long-cycle RTC wake-up signal, if the SOC is greater than the third threshold, it exchanges status feature data with neighboring terminals through the near-field communication unit to attempt to establish a co-transport cluster. If successful, the cluster head handles positioning and uploading; if unsuccessful, the relevant tasks are completed as in Example 1 for a single terminal. If the current SOC is not greater than the third threshold, the terminal only performs the minimum tasks such as abnormal event detection, location acquisition, and alarm reporting, and then returns to deep sleep after completion.

[0068] This application leverages the shared trajectory, start / stop, vibration, and station operation characteristics of terminals in railway container train transport. Based on the consistency condition of the same train, it determines whether to establish a transport cluster. A cluster head is elected and roles within the cluster are assigned through factors such as State of Charge (SOC). Cluster members exchange data via low-power short-range communication units. The cluster head acts as an agent for location and data upload tasks, avoiding repetitive satellite search, location, and upload operations by individual terminals on the train, significantly reducing the power consumption of the entire terminal group. Furthermore, the cluster head can combine data from multiple cluster members to comprehensively assess a single abnormal terminal, improving the accuracy of anomaly detection.

[0069] Furthermore, this application links SOC hierarchical management, on-demand power supply control, terminal sleep strategies, and cluster status. Terminals within the same transport cluster enter a shallow sleep state after completing their tasks, quickly communicating with the cluster head via short-cycle RTC wake-up to maintain cluster relationships. Terminals that lose cluster connections or do not meet the conditions for continuous operation enter a deep sleep state, re-attempting cluster establishment after being woken up via long-cycle RTC wake-up. Through this low duty cycle collaborative mechanism, terminals maintain intra-cluster collaborative relationships while remaining in sleep mode most of the time, achieving an optimal balance between power consumption and response speed.

[0070] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A multi-source wake-up power supply control system for railway container tracking terminals, comprising a main control module and power supply module, event acquisition module, positioning module, communication module, and sensor acquisition module electrically connected thereto; characterized in that, The communication module includes a long-range communication unit and a short-range communication unit; the main control module is used for: Confirm whether the wake-up signal collected by the event acquisition module is a valid wake-up event. If it is, control the terminal to enter the fully activated state; otherwise, maintain the original state. In the fully active state, status feature data is exchanged with neighboring terminals through the near-field communication unit. When the status feature data meets the preset co-column consistency condition, a co-column transport cluster is established. Based on the state feature data, the terminal's role within the same transport cluster is determined. Collaborative tasks are executed through task delegation and information sharing among cluster members. The task execution level is determined in conjunction with battery status information, and power supply control is applied to the task execution modules of each terminal as needed. After the task is completed, control the terminal to enter a shallow sleep state or a deep sleep state.

2. The multi-source wake-up power supply control system for railway container tracking terminals according to claim 1, characterized in that, The event acquisition module includes an acceleration sensing unit, an RTC timing unit, and an RFID triggering unit. When the acceleration sensing unit outputs an acceleration wake-up signal, the main control module enters the wake-up confirmation stage. If the acceleration change in the acceleration wake-up signal exceeds a preset acceleration change threshold and the duration exceeds a preset duration, a valid wake-up event is confirmed; otherwise, it is determined to be an invalid triggering event. When the RTC timing unit outputs a long-period RTC timing signal, or outputs a short-period RTC timing signal while the terminal is in a shallow sleep state, the main control module confirms a valid wake-up event. When the RFID triggering unit detects a valid external stimulus or a valid RFID wake-up command, the main control module confirms a valid wake-up event.

3. The multi-source wake-up power supply control system for railway container tracking terminals according to claim 2, characterized in that, After the main control module completes its task, if it is currently in the same transport cluster and determines, based on the acceleration wake-up signal, that the conditions for continuous railway movement are met and there are no tasks to be processed, then the control terminal enters a shallow sleep state. If, within the preset observation window, the acceleration wake-up signal indicates that the conditions for continuous railway movement are not met and there are currently no tasks to be processed, the control terminal enters a deep sleep state.

4. The multi-source wake-up power supply control system for railway container tracking terminals according to claim 3, characterized in that, For terminals participating in the establishment of a co-transport cluster, the status characteristic data includes at least terminal identifier, timestamp, battery status information, positioning quality, communication quality, cache summary, abnormal event data, alarm event data, most recent location reference, RFID station anchor point, and acceleration characteristic summary; the acceleration characteristic summary includes at least one or more of the following: start-stop time, root mean square value of vibration, main frequency components, impact peak value, duration, and direction change.

5. The multi-source wake-up power supply control system for railway container tracking terminals according to claim 4, characterized in that, Within a preset short-range communication window after confirming a valid wake-up event, the co-column consistency condition includes a short-range communication link quality higher than a preset link quality threshold, and at least two of the following conditions: The acceleration start-stop time difference of multiple terminals within the same time window is less than a preset time difference threshold, and the correlation coefficient of acceleration feature sequences within the same time window is greater than a preset correlation threshold; The difference between the root mean square values ​​of vibration of multiple terminals is less than the preset root mean square difference threshold and the difference between the main frequency components is less than the preset frequency difference threshold. The distance between the nearest location references of multiple terminals is less than a preset distance threshold; Multiple terminals correspond to the same RFID station anchor point, the same station area identifier, or the same railway line corridor.

6. The multi-source wake-up power supply control system for railway container tracking terminals according to claim 4, characterized in that, The cluster head is generated based on the status feature data scores of each terminal, and the cluster head organizes the status feature data to form an aggregated data frame containing a common location field and a member difference field, and uploads it to the backend platform through the long-distance communication unit; after receiving the upload confirmation information from the cluster head, the cluster members mark the corresponding cached data as uploaded by proxy and keep their own long-distance communication units powered off; and when the cluster head obtains a valid positioning result, the cluster members use the positioning result as the position reference for this round and keep their own positioning modules powered off.

7. The multi-source wake-up power supply control system for railway container tracking terminals according to claim 6, characterized in that, If any cluster member in the same transport cluster detects abnormal event data based on a wake-up signal, it sends the abnormal event data to the cluster head. The cluster head then requests other cluster members to return an acceleration feature summary within the corresponding time window through the near-field communication unit. If the acceleration features of multiple cluster members are synchronized in time and their amplitude distribution satisfies the whole-column perturbation feature, the cluster head lowers the upload priority. If only the abnormal terminal itself or some nearby terminals detect an impact feature exceeding the impact threshold, the cluster head raises the upload priority.

8. The multi-source wake-up power supply control system for railway container tracking terminals according to claim 1, characterized in that, The main control module implements on-demand control of the task execution module through the power supply control module. The power supply control module includes an independent load switch connected between the power supply module and the task execution module, and an isolation current limiting unit set on the signal line of the task execution module. The independent load switch is used to implement on-demand power supply to the task execution module, and the isolation current limiting unit is used to switch the signal line connected to the task execution module to a high impedance state or a pull-down holding state when the power is off.

9. The multi-source wake-up power supply control system for railway container tracking terminals according to claim 1, characterized in that, The main control module calculates the State of Charge (SOC) based on the battery status information, and determines the task execution level based on the comparison results of the SOC with the preset first threshold, second threshold, and third threshold, wherein the first threshold > the second threshold > the third threshold. When SOC ≥ the first threshold, the terminal takes priority as the cluster head, the task execution level is the standard level, and the task execution module is started as needed. When the second threshold ≤ SOC < the first threshold, the task execution level is reduced, which slightly restricts at least one of the following: the usage time, usage frequency, and workload of the task execution module; When the third threshold ≤ SOC < the second threshold, the terminal is not a cluster head, the task execution level is restricted, and the usage time, frequency of use, and workload of the task execution module are severely restricted. When SOC < the third threshold, the terminal exits the same transport cluster, the task execution level is set to the protection level, and only the task execution modules required for abnormal event detection, location acquisition and alarm reporting are enabled.

10. The multi-source wake-up power supply control system for railway container tracking terminals according to claim 9, characterized in that, The main control module estimates the state of charge (SOC) by combining ampere-hour integration and open-circuit voltage calibration. When the power module is a primary battery, the main control module pauses open-circuit voltage calibration or sets the open-circuit voltage calibration weight to be lower than the ampere-hour integration weight when it detects that the battery open-circuit voltage is in a flat discharge plateau region. When it detects that the battery open-circuit voltage has entered the inflection point range at the end of the discharge period, it increases the open-circuit voltage calibration weight.

Images