A method and system for operation guarantee of a low-power-consumption cold-resistant collection terminal

Abstract

The application provides a low-power consumption cold-resistant collection terminal operation guarantee method and system, which is applied to the embedded system power management field in industrial big data; the method obtains the terminal battery end voltage, load current and environment temperature signal through a sensor, determines the battery charge state and equivalent internal resistance based on a battery model and a Kalman filtering algorithm, calculates an instability index, generates a power consumption distribution strategy and controls a power management unit to adjust power supply; through the above scheme, the application can improve the power supply stability of the collection terminal in an extremely cold environment and prolong the equipment endurance time.

Description

Technical Field

[0001] This invention relates to the field of embedded system technology in industrial big data, and in particular to a method and system for ensuring the operation of a low-power, cold-resistant data acquisition terminal. Background Technology

[0002] In scientific and industrial fields such as polar expeditions, high-latitude meteorological observations, and cold-region resource exploration, data acquisition terminals are essential devices for obtaining firsthand data. These devices typically need to operate stably for extended periods in unattended, extremely cold environments.

[0003] The challenge of existing technologies lies in the impact of extremely cold environments on the energy systems of devices, particularly lithium batteries. At low temperatures, the rate of electrochemical reactions within the battery decreases, leading to a reduction in usable capacity, meaning that only a portion of the nominal charge may be discharged. Simultaneously, increased internal resistance causes a voltage drop at the battery's terminals during high-current discharge, making it impossible to provide a stable operating voltage for the module, thus causing system restarts or functional failures.

[0004] Current data acquisition terminals generally employ static or simple power management strategies. For example, they might set a fixed low-voltage threshold to shut down the device, or disable some secondary functions when the battery is low. These strategies cannot accurately detect the deterioration of battery health due to low temperatures, nor can they dynamically allocate remaining power based on task priority. As a result, the terminal may unexpectedly crash due to insufficient instantaneous power even when it still has spare power; or, when the battery is about to run out, the data acquisition and storage functions may be prematurely interrupted due to indiscriminate power consumption strategies, leading to data loss and task failure. Therefore, there is an urgent need for an operational assurance technology that can accurately sense the battery status in extremely cold environments and intelligently and adaptively allocate power consumption to solve the problem that existing technologies cannot guarantee continuous operation in harsh environments.

[0005] It should be noted that the information disclosed in the background section above is only used to enhance the understanding of the background of the present invention, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] In view of this, the present invention provides a method and system for ensuring the operation of a low-power, cold-resistant data acquisition terminal, aiming to solve the technical problems of unstable power supply and operation interruption caused by battery performance degradation and lack of effective power management strategies in the existing technology of data acquisition terminals in extremely cold environments. The method improves the operational reliability of data acquisition terminals in industrial big data in extremely cold environments by acquiring physical signals of terminal operation status in real time, accurately estimating battery status based on temperature-related battery models, calculating the instability index by fusing state of charge and equivalent internal resistance, generating a power allocation strategy based on the instability index and task power consumption requirements table, and adjusting the power supply of non-functional modules by a programmable power management unit.

[0007] This invention provides a method for ensuring the operation of a low-power, cold-resistant data acquisition terminal, comprising: The terminal's battery voltage, load current, and ambient temperature signals are acquired in real time through sensors. The processor is based on a battery model associated with the ambient temperature signal and uses a Kalman filter algorithm to process the battery terminal voltage and load current signals to determine the battery's state of charge and equivalent internal resistance. The processor integrates the state of charge and equivalent internal resistance to calculate an instability index that characterizes the stability of the power supply. The processor generates a power allocation strategy based on the instability index and a preset task power requirement table that defines the priority and power requirements of different functional modules. The processor translates the power allocation strategy into control instructions and sends them to the programmable power management unit, which then physically adjusts the power supply to at least one non-core functional module within the terminal.

[0008] In some optional embodiments, the step of generating a power allocation strategy specifically involves: using a genetic algorithm for iterative optimization to generate a power allocation strategy that aims to minimize the total power consumption of the terminal while ensuring the operation of core functional modules.

[0009] In some optional embodiments, the fitness function employed by the genetic algorithm includes a penalty term; when the power allocated to any core functional module is lower than its minimum survivability power defined in the task power requirement table, the penalty term is activated to reduce the fitness value of the power allocation strategy.

[0010] In some optional embodiments, the constraints for iterative optimization include: the total power consumption of the terminal does not exceed the current maximum available output power calculated in real time by the battery model.

[0011] In some optional embodiments, the Kalman filter algorithm is the extended Kalman filter algorithm, and the battery model is a second-order RC equivalent circuit model.

[0012] In some alternative embodiments, the step of calculating the instability index includes: calculating the residual energy decay rate based on the state of charge, calculating the output capability degradation rate based on the equivalent internal resistance, and performing a weighted summation of the residual energy decay rate and the output capability degradation rate.

[0013] In some alternative embodiments, physical regulation specifically involves reducing the supply voltage of non-core functional modules or performing power gating to interrupt their power supply.

[0014] In some optional embodiments, the task power consumption requirement table defines at least the priority level, normal operating power consumption, and minimum survival power consumption for each functional module.

[0015] In some optional embodiments, the method further includes: When the instability index exceeds the preset emergency survival threshold for the first time, the core survival mode is triggered and entered. After entering the core survival mode, the data acquisition module and non-volatile storage module are driven first to perform a final data snapshot acquisition and forced storage operation; After confirming that the final data snapshot has been successfully stored, activate the wireless communication module to put it into the preset beacon communication mode.

[0016] In some alternative embodiments, the steps of performing the final data snapshot acquisition and forced storage operation include: temporarily suspending the data transmission function of the wireless communication module and driving the data acquisition module and non-volatile storage module with the highest priority and maximum available power.

[0017] In some alternative embodiments, in beacon communication mode, the wireless communication module periodically sends heartbeat packets containing only the device ID, the last data snapshot timestamp, and a status code at time intervals that are dynamically adjusted based on the remaining battery power.

[0018] In some optional embodiments, the method further includes: The data communication link quality parameters of the terminal's wireless communication module are monitored in parallel. When communication link quality parameters show persistent abnormal deterioration, the model credibility-related parameters in the Kalman filter algorithm are dynamically corrected.

[0019] In some optional embodiments, the step of determining that persistent abnormal degradation has occurred includes: Assuming the signal strength of the external network where the terminal is located is normal, determine whether the communication link quality parameters, such as data retransmission rate or bit error rate, continuously exceed the preset threshold.

[0020] In some optional embodiments, dynamic correction specifically refers to: The process noise covariance matrix Q value in the Kalman filter algorithm is automatically increased to reduce dependence on the battery model and enhance the tracking of the current battery terminal voltage and load current signals.

[0021] In some alternative embodiments, non-core functional modules include: a display module, a GPS positioning module, or an auxiliary sensor module.

[0022] This invention provides an operation support system for a low-power, cold-resistant data acquisition terminal, comprising: The signal acquisition module is configured to acquire the terminal's battery voltage, load current, and ambient temperature signals in real time. The state determination module is configured to determine the battery's state of charge and equivalent internal resistance based on a battery model associated with an ambient temperature signal and by using a Kalman filter algorithm to process the battery terminal voltage and load current signals. The index calculation module is configured to combine the state of charge and equivalent internal resistance to calculate the instability index, which characterizes the stability of the power supply. The strategy generation module is configured to generate a power allocation strategy based on the instability index and a preset task power requirement table that defines the priority and power requirements for different functional modules. The regulation control module is configured to translate power allocation strategies into control commands and send them to the programmable power management unit for physical regulation of the power supply to at least one non-core functional module within the terminal.

[0023] In some optional embodiments, the policy generation module is specifically configured to: Genetic algorithms are used for iterative optimization to generate power allocation strategies.

[0024] In some optional embodiments, the fitness function employed by the genetic algorithm includes a penalty term; when the power allocated to any core functional module is lower than its minimum survivability power defined in the task power requirement table, the penalty term is activated to reduce the fitness value of the power allocation strategy.

[0025] In some optional embodiments, the system further includes a monitoring module for monitoring data communication link quality parameters of the wireless communication module; and the status determination module is further configured to: When communication link quality parameters show a continuous abnormal deterioration after eliminating external network signal interference, the model credibility-related parameters in the Kalman filter algorithm are automatically and dynamically corrected.

[0026] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit the invention.

[0027] The operation guarantee method and system for a low-power, cold-resistant data acquisition terminal of the present invention have the following beneficial effects: This method, through precise state estimation and dynamic power consumption control, effectively avoids unexpected downtime caused by instantaneous power shortages, ensuring stable operation of the terminal throughout its entire lifecycle. Global power consumption optimization allocates every unit of electrical energy to functionality, extending the effective operating time of the terminal by more than 50% with the same battery capacity. This method ensures that the most valuable data is completely preserved before device failure and that there is an opportunity to send a distress signal, fundamentally preventing data loss. By treating temperature as a variable, the terminal can adapt to changes in the external environment and its own state, exhibiting high autonomy and robustness. Attached Figure Description

[0028] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings.

[0029] Figure 1 This is a flowchart of an operation guarantee method for a low-power, cold-resistant data acquisition terminal according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the operation support system of a low-power cold-resistant acquisition terminal according to an embodiment of the present invention. Detailed Implementation

[0030] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the invention will be more comprehensive and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0031] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0032] The flowchart shown in the attached diagram is merely an illustrative example and does not necessarily include all steps. For example, some steps may be broken down, while others may be combined or partially combined. Therefore, the actual execution order may change depending on the specific circumstances.

[0033] Data acquisition terminals operate in extremely cold environments, where battery performance is significantly affected by temperature, manifesting as capacity degradation and increased internal resistance. The battery's state of charge (SOC) and internal resistance directly impact power supply stability and the terminal's endurance. Accurately assessing battery status and managing power consumption are crucial for ensuring stable terminal operation. The Kalman filter algorithm, by establishing an equivalent circuit model of the battery and integrating multi-dimensional information such as voltage, current, and temperature, achieves accurate estimation of the battery's state. By fusing the estimated SOC and internal resistance, the instability of the power supply system can be quantitatively characterized. Based on the quantified instability indicators and combined with task priorities, the power consumption allocation of each module can be dynamically adjusted to ensure task operation. Through this dynamic power management, power supply instability caused by battery performance degradation can be effectively avoided, thereby extending the terminal's operating time.

[0034] like Figure 1 As shown in the figure, this embodiment of the invention provides a method for ensuring the operation of a low-power, cold-resistant data acquisition terminal. The method includes the following steps: Step S100: Acquire the battery terminal voltage, load current, and ambient temperature signals of the terminal in real time using sensors. In a specific example, a voltage sensor, a current sensor, and a temperature sensor located inside the acquisition terminal work together at a set frequency (e.g., 1Hz) to acquire the voltage value across the battery terminals, the current value flowing through the battery, and the temperature data of the battery or its surrounding environment in real time. These sensors can be discrete components or integrated into an all-in-one sensor chip. The sensors convert the acquired analog signals into digital signals and transmit the data to the processor via a bus interface (e.g., I2C or SPI). In some other alternative implementations, the ambient temperature signal can also be acquired using a temperature sensor deployed outside the acquisition terminal, or by sampling at different locations using multiple temperature sensors to obtain more comprehensive temperature information.

[0035] Step S200: The processor, based on a battery model associated with the ambient temperature signal, uses a Kalman filter algorithm to process the battery terminal voltage and load current signals to determine the battery's state of charge (SOC) and equivalent internal resistance. Specifically, the processor has a battery state estimation module that stores a mathematical model describing the battery's electrochemical characteristics. This model includes multiple parameters characterizing the battery's internal resistance, capacity, etc. These parameters are associated with the ambient temperature signal obtained in step S100; each parameter is expressed as a function of temperature and is determined through experimental calibration or table lookup. The Kalman filter algorithm uses this battery model, combined with the battery terminal voltage and load current signals obtained in step S100, to recursively estimate the battery's SOC and equivalent internal resistance. In some alternative implementations, other types of state estimation algorithms, such as particle filtering and unscented Kalman filtering, can be used instead of the Kalman filter algorithm.

[0036] Step S300: The processor fuses the state of charge (SOC) and equivalent internal resistance (EMR) to calculate an instability index characterizing power supply stability. For example, in the index calculation module, the SOC and EMR estimated in step S200 are used as inputs and fused using a weighted average or nonlinear function to calculate a value characterizing the current instability of the power supply system, i.e., the instability index. The higher the index, the more unstable the power supply system is, and the higher the risk of unexpected shutdown. In some other optional implementations, factors such as battery temperature and discharge rate can be incorporated into the calculation of the instability index to more comprehensively assess power supply stability.

[0037] Step S400: The processor generates a power allocation strategy based on the instability index and a preset task power requirement table that defines the priorities and power consumption requirements for different functional modules. For example, the processor internally stores a task power requirement table, which records in detail the priority of each functional module in the acquisition terminal, the power consumption requirements in normal working mode, and the minimum power consumption required to maintain basic operation. The power management module in the processor queries this table based on the instability index calculated in step S300 and determines the power allocation scheme for each functional module according to preset rules or algorithms. In some other optional implementations, the task power requirement table can be stored in external memory or dynamically obtained from a cloud server via wireless communication to adapt to different application scenarios and requirements.

[0038] Step S500: The processor converts the power allocation strategy into control instructions and sends them to the programmable power management unit (PMU), which then physically adjusts the power supply to at least one non-functional module within the terminal. The processor converts the power allocation strategy generated in step S400 into specific control instructions, such as setting voltage and current limits. These instructions are sent to the PMU via a preset communication interface (e.g., I2C, SPI). Based on the received instructions, the PMU adjusts the power supply voltage, current, or on / off state of the corresponding functional module, thereby achieving precise control over the terminal's power consumption. In other optional implementations, multiple PMUs can control different functional modules separately, or a PMU with more channels and more flexible configuration options can be used to achieve more refined power management.

[0039] Through the above steps, the sensor collects battery and environmental information in real time. The processor uses a battery model and Kalman filter algorithm to accurately estimate the battery state and dynamically generates a power allocation strategy based on the instability index and task power consumption requirements. Finally, the programmable power management unit physically adjusts the power consumption of non-functional modules. These technical features work together to sense changes in battery state in real time and intelligently adjust the terminal's power allocation, thereby extending the terminal's battery life in harsh environments such as extreme cold and ensuring stable operation. This solves the technical problems of existing static power management strategies being unable to cope with low-temperature environments, easily leading to unexpected device downtime and data loss.

[0040] Through the above-described scheme, this embodiment can dynamically adjust battery model parameters according to ambient temperature, improving the accuracy of battery state estimation; it can integrate multi-dimensional information such as state of charge and equivalent internal resistance to more accurately assess the stability of the power supply system; and it can dynamically generate power allocation strategies based on the instability index and task power consumption requirements, achieving intelligent management of terminal power consumption. Compared with existing technologies, this embodiment can effectively extend the battery life of the acquisition terminal in harsh environments such as extreme cold and improve its operational reliability, thereby avoiding unexpected shutdowns and data loss due to insufficient power.

[0041] In one specific implementation, the steps for generating a power allocation strategy are as follows: A genetic algorithm is used for iterative optimization to generate a power allocation strategy that aims to minimize the total power consumption of the terminal while ensuring the operation of functional modules. First, for each controllable functional module of the terminal, a set of power consumption levels is predefined, for example, it can be set as a set of {"Off", "Low Power Mode", "Normal Mode", "High Performance Mode"}. Specifically, each individual in the genetic algorithm represents a power allocation scheme, i.e., a vector containing the power consumption levels of all controllable modules. Then, the genetic algorithm evaluates and selects individuals in the population based on a fitness function, guiding the population to evolve towards a better direction. The fitness function is designed to quantify the merits of each power allocation scheme, with the goal of minimizing the total power consumption of the terminal while meeting the minimum power consumption requirements of the functional modules. Next, the algorithm performs genetic operations such as selection, crossover, and mutation to generate new individuals. For the selection operation, methods such as roulette wheel selection and tournament selection can be used. For the crossover operation, methods such as single-point crossover and multi-point crossover can be used to generate new power allocation schemes. The mutation operation randomly alters certain genes of an individual with a certain probability, introducing new possibilities. Through the above iterative process, the genetic algorithm continuously optimizes the power allocation scheme, ultimately generating a power allocation strategy that minimizes the total power consumption of the terminal while ensuring the operation of functional modules. In other optional implementations, the genetic algorithm can also be replaced by other global optimization algorithms such as particle swarm optimization or simulated annealing to find the optimal power allocation strategy. These algorithms can also search within the solution space to find the optimal solution that meets preset conditions.

[0042] Through the above scheme, this embodiment can effectively perform global optimization in the solution space of the power allocation scheme using a genetic algorithm. Under the premise of ensuring the stable operation of functional modules, it can reduce the total power consumption of the terminal as much as possible, thereby maximizing the working time of the terminal in extremely cold environments.

[0043] In one specific implementation, the fitness function employed by the genetic algorithm includes a penalty term; when the power allocated to any core functional module is lower than its minimum survivability power defined in the task power requirement table, the penalty term is activated to reduce the fitness value of the power allocation strategy. Specifically, the fitness function takes the following form:

[0044] In the above formula, the parameters are as follows: Fitness represents the fitness value of an individual power allocation strategy; the larger the value, the better the solution. N is the total number of controllable functional modules within the terminal. Let be the power consumption of the i-th functional module. This is the total power consumption weighting factor, used to adjust the weight of total power consumption in fitness evaluation. Core represents the set of indexes for core functional modules. Penalty item for the core functional module with index j. The minimum survivability power consumption of the core functional module with index j, which is predefined by the task power consumption requirement table. A preset maximum penalty constant in C, for example, set to... .

[0045] Penalty items The calculation follows the following conditional rules: The power consumption allocated to the core functional module with index j Below its minimum survivability power consumption When this occurs, the penalty term is activated, and its calculation method is as follows:

[0046] In other cases, the value of the penalty term is:

[0047] The fitness function is designed to ensure that any power allocation scheme that causes the core functional module's power consumption to fall below its survival threshold will have its fitness value approach zero due to a large penalty term, thus being naturally eliminated during the "survival of the fittest" selection process of the genetic algorithm. In some other optional implementations, to further enhance the penalty effect, the penalty term can also be in an exponential form, for example... Or, the penalty constant. The value of is not fixed, but dynamically adjusted according to the magnitude of the instability index PII. The higher the PII value, the more stable the instability index. The value of is also increased accordingly, thus the penalty for violating the module power consumption constraint is greater.

[0048] Through the above scheme, this embodiment ensures that the genetic algorithm always prioritizes the stable operation of core functions when optimizing power consumption, even if it means sacrificing some overall power consumption performance, thereby ensuring the reliable operation of the terminal under energy shortage conditions.

[0049] In one specific implementation, the constraints of the iterative optimization include: the total power consumption of the terminal does not exceed the current maximum available output power calculated in real time by the battery model. First, before each iteration of the genetic algorithm, the processor obtains the current state of charge of the battery from the state determination module. and equivalent internal resistance The estimated values. Then, using these values, according to the formula... Calculate the battery's current maximum available output power. .in, It is based on and ambient temperature The open-circuit voltage is obtained by looking up a table. Next, when generating a new individual, the genetic algorithm checks the total power consumption of that individual. Does it exceed If the limit is exceeded, the individual will be discarded and a new individual will be generated until the constraints are met. Specifically, a maximum number of attempts can be set. If an individual that meets the constraints cannot be generated after reaching the maximum number of attempts, the power consumption requirement of the module will be reduced to ensure that the module can at least work, thereby guaranteeing the basic functionality of the system.

[0050] By employing the above approach, this embodiment can avoid the power allocation strategy exceeding the actual power supply capacity of the battery, thereby preventing system crashes and further improving system reliability.

[0051] In one specific implementation, the Kalman filtering algorithm used in the state determination module is specifically the Extended Kalman Filter (EKF) algorithm, and the battery model adopts a second-order RC equivalent circuit model to more accurately capture the polarization effect of the battery.

[0052] Specifically, the state-space expression of the battery model upon which the EKF algorithm is based can be written as:

[0053] This is the state equation, where: State vector Defined as:

[0054] The state transition matrix A is defined as follows:

[0055] The input matrix B is defined as follows:

[0056] The observation equation for this model is:

[0057] In the above formula, the meaning of each parameter is as follows: k: index of the discrete time step. : The state of charge at time k. Polarization voltage on the two parallel RC branches at time k. : Sampling time interval. The time constant of the first RC branch. : The time constant of the second RC branch, . Coulomb efficiency. : Rated capacity of the battery. : Load current at time k-1. The process noise at time k-1 is assumed to be Gaussian white noise. The measurement noise at time k is assumed to be Gaussian white noise. : Battery terminal voltage measured at time k. The open-circuit voltage of a battery, which indicates its state of charge. It is a nonlinear function of temperature T. The circuit parameters of the model, including ohmic internal resistance, polarization resistance, and polarization capacitance.

[0058] In this embodiment, the circuit parameters of the model and open-circuit voltage function All parameters were pre-calibrated experimentally and their functional relationship with temperature T was established and stored in a lookup table. Before each EKF iteration, the processor looks up and interpolates the corresponding model parameters based on the currently measured temperature T to improve modeling accuracy at different temperatures. Due to the observation equations... With state Due to the nonlinear relationship, the Extended Kalman Filter (EKF) algorithm is employed. This algorithm locally linearizes the nonlinear function by calculating the Jacobian matrix of the observation equation with respect to the state variables at each time step, thereby applying the standard Kalman filter framework to iteratively perform prediction and update steps, dynamically tracking changes in the battery state. In some alternative implementations, the second-order RC equivalent circuit model can be replaced with other equivalent circuit models, such as the first-order RC model or the Thevenin model; the EKF algorithm can also be replaced with other nonlinear filtering algorithms, such as the Unscented Kalman Filter (UKF) or the Particle Filter (PF).

[0059] Through the above scheme, this embodiment can accurately simulate the dynamic characteristics of the battery, improve the accuracy of battery state estimation, and thus provide a more reliable data foundation for subsequent power consumption optimization and management.

[0060] In one specific implementation, the steps for calculating the instability index include: first, based on the state of charge... Calculate the remaining energy decay rate The attenuation rate is obtained using the following formula:

[0061] Specifically, the lower the state of charge, the higher the rate of remaining energy decay, indicating that the battery has less remaining usable capacity and the system's power supply risk is higher.

[0062] Then, based on the equivalent internal resistance Calculate the output capability degradation rate The degradation rate is obtained using the following formula:

[0063] in, This is the reference internal resistance of the battery in its brand-new state. This reference internal resistance can be calibrated at the battery factory and stored in the terminal's non-volatile memory. Specifically, the higher the equivalent internal resistance, the higher the rate of output capability degradation, indicating a weaker battery load-carrying capacity and a higher risk to the system's power supply.

[0064] Next, the remaining energy decay rate was analyzed. and output capability degradation rate By performing a weighted summation, the instability index is obtained. :

[0065] in, and These are preset weighting coefficients used to adjust the contributions of remaining energy decay rate and output capability degradation rate to the instability index. For example, in some implementations, the weighting can be adjusted according to the battery type and application scenario. Set to 0.6, Setting it to 0.4 emphasizes the impact of state of charge on power supply stability. In other implementations, if the application scenario has high instantaneous power requirements, it can be set to... Set it to a higher value, such as 0.7, to give more weight to the impact of internal resistance on power supply stability.

[0066] In some other alternative implementations, the calculation of the remaining energy decay rate and the output capability degradation rate can employ a non-linear transformation. For example, the Sigmoid function can be used to... and Normalization is performed before calculating the attenuation rate and degradation rate.

[0067] Through the above approach, this embodiment can more comprehensively assess the power supply risk of the battery, thereby providing a more accurate basis for generating subsequent power allocation strategies. Compared with existing technologies, this embodiment, by integrating information from both the state of charge and internal resistance dimensions, can more accurately quantify the battery's health status, avoiding the risk of misjudgment caused by relying solely on a single voltage threshold.

[0068] In one specific implementation, physical regulation involves the following steps: First, identifying non-functional modules requiring power consumption adjustment, such as the display screen, GPS module, and environmental sensors. Specifically, for the display screen module, the regulation control module sends a command to the display driver chip via the I2C bus to reduce the backlight brightness level, adjusting it from the maximum brightness of 255 / 255 to 128 / 255 or even lower. Then, for the GPS module, if its priority in the task power consumption requirement table is low, or its instability index (PII) is high, the regulation control module controls the GPS module's power enable pin via GPIO to directly shut down its power supply, achieving power gating. Next, for the environmental sensor, if its sampling frequency is higher than the minimum requirement, the regulation control module writes a new sampling frequency value to the sensor control register via the SPI bus, for example, reducing the sampling frequency from 1Hz to 0.1Hz, thereby reducing its power consumption. In other alternative implementations, for the display screen module, power consumption can be reduced by adjusting its operating mode (e.g., switching from color mode to grayscale mode). For the GPS module, instead of directly shutting down the power supply, its data update frequency is reduced or some satellite signal reception channels are selectively disabled. For environmental sensors, power consumption can be adjusted by reducing their operating voltage.

[0069] Through the above scheme, this embodiment can use different physical adjustment methods to finely control the power consumption of non-functional modules according to the specific characteristics of functional modules and system status, thereby minimizing the total power consumption of the system and extending the overall battery life of the terminal while ensuring the operation of functions.

[0070] In one specific implementation, the task power consumption requirement table is a two-dimensional array stored in the processor's Flash memory. Each row of the array corresponds to a controllable functional module, including but not limited to: wireless communication modules, data acquisition modules, storage modules, display modules, GPS modules, and various sensor modules. Each column of the array defines the power consumption parameters of that module.

[0071] Specifically, the power consumption requirements table for this task shall include at least the following: Priority level: An integer value, such as from 1 to 5, where 1 represents the highest priority and 5 represents the lowest priority. Modules with higher priority will be prioritized in power allocation. Data acquisition, storage, and communication modules are typically set to the highest priority of 1 or 2.

[0072] Normal operating power consumption: Typical power consumption value under normal module operating conditions, in milliwatts (mW).

[0073] Minimum survivability power consumption: The minimum power consumption required to maintain the module's most basic functions, measured in milliwatts (mW). For example, for a data acquisition module, this value might correspond to the power consumption required to keep the ADC in standby mode. For a communication module, this value might correspond to the minimum power required to maintain the ability to transmit beacon signals.

[0074] First, during the generation of the power allocation strategy, the fitness function of the genetic algorithm forces the minimum survival power requirements of high-priority modules to be met, ensuring continuous operation even under extreme energy constraints. Specifically, the fitness function includes a penalty term for each module. If the power allocated to a module is lower than its minimum survival power, this penalty term is activated, significantly reducing the fitness of the entire power allocation strategy, thus eliminating the module during the iteration process of the genetic algorithm.

[0075] Then, for non-module components, the task power consumption requirement table provides a range for power consumption adjustment. For example, a display module can have multiple power consumption levels corresponding to different brightness levels. The strategy generation module can select an appropriate power consumption level based on the current energy situation and system load. When the instability index is high, the strategy generation module will tend to select a lower power consumption level to save energy.

[0076] Next, the processor, based on the definitions in the task power consumption requirement table, translates the generated optimal power allocation strategy into specific control instructions, which are then sent to the programmable power management unit (PMU) via a standard bus (such as IC / SPI). The PMU then precisely adjusts the power supply voltage or on / off state of each functional module according to the instructions.

[0077] In other alternative implementations, the task power consumption requirement table may include more information, such as the module's startup time, shutdown time, and power consumption curves under different operating modes. Priority levels can be represented using enumeration types, such as "Critical," "High," "Medium," "Low," and "Optional." Minimum survivability power consumption can be represented using duty cycles, meaning that the module can operate intermittently within a certain time period, rather than having to continuously maintain the minimum survivability power consumption state.

[0078] Through the above solution, this embodiment can further improve the precision and flexibility of energy allocation, and maximize the effective working time of the terminal while ensuring continuous operation of functions.

[0079] In one specific implementation, the method further includes: When the instability index calculated by the index calculation module reaches or exceeds the preset emergency survival threshold, the system's operational support process will enter survival mode. First, the system interrupts its current normal operation, suspending all non-essential tasks. Specifically, the processor controls the power management unit to temporarily stop power supply to most peripherals, retaining power only for the minimum hardware units required to maintain data processing and storage functions. Then, the system drives the data acquisition module and non-volatile storage module to enter working mode, performing a rescue storage operation. This operation writes the final environmental data collected by the sensors, system operation logs, and other status information into the non-volatile memory as a data snapshot, ensuring that this data is preserved even in the event of a complete system power failure. Next, after confirming that the data snapshot has been successfully stored, the system activates the wireless communication module and puts it into the preset beacon communication mode, preparing to send a distress signal to the outside world.

[0080] In some alternative implementations, the aforementioned emergency survival threshold is not a fixed value, but can be dynamically adjusted according to the specific environmental conditions of the terminal. For example, in areas with poor signal, the threshold can be appropriately lowered to enter survival mode earlier and ensure data security. Furthermore, the storage location of the final data snapshot can also be configurable; it can be either internal non-volatile memory or external removable storage media to improve data security.

[0081] Through the above solution, this embodiment can ensure data security to the greatest extent possible and increase the possibility of equipment rescue when the data acquisition terminal encounters extremely harsh working conditions, even if the system is about to crash.

[0082] In one specific implementation, when the instability index (PII) calculated by the index calculation module first exceeds the emergency survival threshold, the regulation control module suspends any data transmission tasks being performed by the wireless communication module. Specifically, this is achieved by writing an instruction to the control register of the communication module, forcing it to stop data transmission and clearing the transmission buffer.

[0083] Then, the control module sends a command to the programmable power management unit (PMU) to power off all modules except the main control unit, data acquisition modules (such as the ADS1256 high-precision ADC), and non-volatile memory modules (such as the W25Q series SPI Flash). Next, the PMU supplies all remaining maximum available power to the data acquisition module and non-volatile memory modules. At this time, the task scheduler in the main control unit is configured to enter the highest priority mode, stopping all background tasks and adjusting the CPU clock frequency to the preset highest frequency (e.g., 168MHz) to ensure that the data acquisition module and non-volatile memory modules operate at maximum speed.

[0084] In other alternative implementations, suspending the data transmission task of the wireless communication module can also be achieved by modifying its transmit power control register to force its transmit power to 0, or by disabling its transmit interrupt. Driving the data acquisition module and non-volatile memory module can be achieved not only by increasing the CPU clock frequency, but also by optimizing the data transmission protocol, such as changing from SPI mode 0 to mode 3, or adjusting the DMA transfer burst size to improve the data transmission rate.

[0085] Through the above solution, this embodiment can ensure that when entering survival mode, the data acquisition module and non-volatile storage module are given priority to obtain the maximum power resources, so as to complete the final data snapshot acquisition and storage as quickly as possible and reduce the risk of data loss due to power depletion.

[0086] In one specific implementation, upon entering beacon communication mode, the processor first reads a battery-time mapping table from a pre-defined Flash storage area. This table records the correspondence between the remaining battery power and the beacon transmission interval. For example, when the remaining battery power is greater than 20%, the beacon transmission interval is 30 minutes; when the remaining battery power is between 10% and 20%, the transmission interval is extended to 1 hour; and when the remaining battery power is less than 10%, it is further extended to 2 hours.

[0087] Then, the processor calls the power estimation module (which can reuse the state determination module in Embodiment 1, or use a simplified open-circuit voltage method) to estimate the current remaining power of the battery. .

[0088] Next, the processor consults the power-time mapping table, according to... The value determines the sleep time for this beacon transmission. .

[0089] Specifically, the processor is configured with a low-power timer (such as LPTIM in the STM32L series microcontrollers) to enable it to... An interrupt is generated after a certain time. After the timer starts, the processor immediately enters a deep sleep mode to minimize power consumption.

[0090] When a timer interrupt occurs, the processor is woken up and performs the following operations: Preparing the beacon data packet: Read the device ID and the timestamp of the last data snapshot from the Flash memory, and combine them with a predefined "SOS" status code to form a very simple data packet. The length of this data packet should be as short as possible, for example, no more than 20 bytes, to reduce the transmission time and power consumption of the wireless module.

[0091] Configure the wireless module: Configure the operating parameters of the wireless module (e.g., LoRa SX1278), including transmission frequency, modulation method, and transmission power. To further reduce power consumption, the lowest possible transmission power and the narrowest possible bandwidth can be used.

[0092] Sending the beacon: The prepared data packet is sent out via the wireless module. After transmission is complete, the power to the wireless module is immediately turned off, and the processor is put back into deep sleep mode, waiting for the next timer interrupt.

[0093] In some alternative implementations, the power-time map may not be stored in local flash memory, but rather dynamically updated from a cloud server via OTA (Over-The-Air). Furthermore, the method for determining the beacon transmission interval may not rely on table lookups, but can be directly calculated using a preset function, for example... Where K is a constant. In some alternative implementations, the status code may include more granular status information in addition to "SOS", such as battery health and current device orientation, to help search and rescue personnel assess the situation more accurately.

[0094] Through the above solution, this embodiment can dynamically adjust the beacon transmission frequency according to the remaining battery power, thereby ensuring that a distress signal can be sent at any time while maximizing the overall battery life of the device.

[0095] In one specific implementation, the method further includes parallel monitoring of the data communication link quality of the terminal's wireless communication module. Specifically, a new independent link quality monitoring module is added to the system. This module runs in parallel with the state determination module, without affecting the basic processes of state estimation and power consumption control.

[0096] First, the monitoring module is configured to periodically collect various link quality parameters of the wireless communication module. These parameters may include, but are not limited to, packet transmission delay, packet loss rate, and Received Signal Strength Indicator (RSSI). The collection period can be flexibly adjusted according to the actual application scenario, for example, set to 1 second or 5 seconds. The monitoring module can obtain these parameters through various methods, such as reading the communication module's internal registers, parsing statistical information from the communication protocol stack, or utilizing the API interface provided by the communication module.

[0097] The monitoring module then analyzes and evaluates the collected link quality parameters according to preset rules. For example, a sliding window can be set to calculate the average packet loss rate over a period of time (such as 10 seconds or 30 seconds). If the average packet loss rate exceeds a preset threshold, the link quality is considered to have deteriorated. Optionally, more complex statistical analysis methods can be introduced, such as using a Kalman filter to smooth the link quality parameters to reduce the impact of noise.

[0098] When the monitoring module detects a persistent and abnormal degradation in link quality parameters, it triggers a correction action. This correction action operates on the state determination module, aiming to dynamically adjust the model confidence-related parameters in the Kalman filter algorithm. Various dynamic adjustment strategies can be employed, such as directly adjusting the Q-value of the process noise covariance matrix or the R-value of the measurement noise covariance matrix of the Kalman filter, or adjusting certain elements in the state transition matrix or observation matrix. The specific adjustment values ​​can be adaptively adjusted according to the degree of link quality degradation.

[0099] In other alternative implementations, the monitoring module can also use machine learning algorithms to predict trends in link quality and adjust the model reliability of the Kalman filter algorithm in advance based on the prediction results, thereby achieving a more proactive system adaptability. For example, time series prediction models (such as ARIMA models or LSTM neural networks) can be used to model historical link quality data and predict link quality over a future period.

[0100] Through the above scheme, this embodiment can dynamically correct the model confidence parameters in the Kalman filter algorithm based on real-time monitoring of wireless communication link quality, thereby improving the accuracy and robustness of battery state estimation and ultimately improving the operational reliability of the terminal.

[0101] In one specific implementation, the steps for determining persistent abnormal degradation include: first, assessing the external network signal strength. Specifically, the wireless communication module periodically (e.g., every 5 seconds) reads its own RSSI value and provides it to the monitoring module. The monitoring module maintains a sliding window (e.g., 60 seconds in length) and records the most recent 12 RSSI readings. Then, it calculates the average RSSI value within this window. If this average value is higher than a preset signal strength threshold (e.g., -90 dBm, which can be adjusted according to the actual communication technology and deployment environment), the monitoring module considers the current external network signal strength to be at a normal level, ruling out communication quality degradation caused by external network factors. Only then does the monitoring module further assess whether communication link quality parameters such as data retransmission rate or bit error rate exceed preset thresholds.

[0102] In other alternative implementations, in addition to using RSSI values, other network-side parameters, such as Reference Signal Received Power (RSRP) and Reference Signal Received Quality (RSRQ), can be considered to evaluate external network signal strength. Furthermore, different statistical methods can be employed to evaluate RSSI values, such as using the median or weighted average, to improve the accuracy and robustness of the evaluation.

[0103] Through the above solution, this embodiment can more accurately determine the cause of communication quality degradation, avoid unnecessary battery model corrections triggered by external network environment problems, and thus improve the accuracy and reliability of system adaptive adjustment.

[0104] In one specific implementation, the dynamic correction process is carried out as follows: First, when the monitoring module detects a persistent abnormal degradation in the communication link quality, it reads the current value of the process noise covariance matrix Q, which is pre-stored in the processor's Flash memory. Specifically, this matrix Q is a 3x3 diagonal matrix, with its three diagonal elements corresponding to the state vectors. The noise variance.

[0105] Then, the monitoring module will take the first diagonal element of the Q matrix corresponding to the SOC state (denoted as ). The value of ) is read out and multiplied by a magnification factor. To obtain a new Value, represented as In this embodiment, The value is 10.

[0106] Next, the monitoring module will... The value is written back to the position of the first diagonal element of the process noise covariance matrix Q, completing the update of the Q matrix. This operation is achieved by directly modifying the corresponding address of the Q matrix stored in the processor memory.

[0107] Specifically, the aforementioned memory address can be obtained by consulting the processor's memory mapping table. In some other alternative implementations, the amplification factor... It can also adaptively adjust based on the degree of communication link quality degradation. For example, the higher the data retransmission rate exceeds the threshold, the more... The larger the value of Q, the faster the model can be adjusted. Furthermore, the structure of the Q matrix is ​​not limited to a diagonal matrix; more complex covariance structures can also be used, depending on the specific application scenario.

[0108] Through the above scheme, this embodiment can reduce the dependence of the state estimator on the preset battery model by dynamically adjusting the process noise covariance matrix Q value in the Kalman filter algorithm, so that it can rely more on the current real-time battery terminal voltage and load current signal, thereby improving the system's adaptive ability to sudden changes in battery state or model inaccuracy.

[0109] In one specific implementation, the non-core functional modules include: a display screen module for displaying the status of the data acquisition terminal and receiving user instructions, a GPS positioning module for providing geographical location information, and auxiliary sensor modules such as an ambient light sensor and a humidity sensor for assisting data acquisition.

[0110] The display module uses a low-power OLED screen that enters sleep mode when not needed (e.g., when there is no user interaction or alarm information). The processor controls the OLED screen's driver chip via the I2C bus to control screen wake-up, content refresh, and sleep. Specifically, the processor dynamically adjusts the screen brightness based on the Instability Index (PII) value; for example, when the PII is higher than 0.7, the screen brightness is reduced to 50%, and when it is higher than 0.9, the screen is forcibly turned off to save power.

[0111] The GPS positioning module employs a sleep-wake mechanism, waking up only when location information needs to be uploaded or time synchronization is required. Upon waking, the GPS module attempts to locate itself. If it fails to locate within a preset time (e.g., 30 seconds), it immediately enters sleep mode to avoid the extra power consumption associated with prolonged satellite signal searches. In other optional implementations, the GPS module's positioning frequency can be dynamically adjusted based on task priority. For example, during urgent data uploads, the GPS module's wake-up frequency is increased to ensure the accuracy of location information; while in standby mode, the GPS module is completely shut down to minimize power consumption.

[0112] The sampling frequency of auxiliary sensor modules (such as ambient light and humidity sensors) can also be dynamically adjusted according to task requirements and the instability index. For example, when the instability index is high, the sensor sampling frequency can be reduced, or sampling can be performed only once at a specific moment (such as before starting a data acquisition task). Specifically, the processor controls the power supply switch of the auxiliary sensors via GPIO and reads sensor data using SPI or I2C interfaces. In some alternative implementations, the data from the auxiliary sensors can serve as supplementary input for the instability index calculation; for example, when ambient humidity is too high, it may cause battery performance to degrade, thereby further increasing the weight of the instability index.

[0113] Through the above solution, this embodiment can further refine the management of power consumption of non-functional modules, ensuring that in extremely cold environments, the limited power of the acquisition terminal is prioritized for supplying functional modules, thereby extending the effective working time of the terminal to a greater extent.

[0114] This invention provides an operational support system for a low-power, cold-resistant data acquisition terminal, such as... Figure 2 As shown, the system includes: a signal acquisition module M100, a state determination module M200, an index calculation module M300, a strategy generation module M400, and an adjustment and control module M500.

[0115] The signal acquisition module M100 is configured to acquire the battery terminal voltage, load current, and ambient temperature signals of the terminal in real time. In one embodiment, the signal acquisition module M100 includes an INA219 current / voltage monitoring chip and a DS18B20 temperature sensor. The INA219 chip obtains the battery terminal voltage and load current by measuring the voltage across the battery and the current flowing through the battery. The DS18B20 temperature sensor measures the temperature of the battery surface as the ambient temperature signal. The sensors periodically sample at a frequency of 1Hz and transmit the acquired data to the status determination module M200. In other alternative embodiments, other types of voltage / current sensors and temperature sensors can be used, such as Hall effect current sensors, thermistors, etc., as long as they can provide accurate voltage, current, and temperature measurements.

[0116] The state determination module M200 is configured to determine the battery's state of charge (SOC) and equivalent internal resistance based on a battery model associated with the ambient temperature signal and using a Kalman filter algorithm to process the battery terminal voltage and load current signals. In this embodiment, the algorithm of the state determination module M200 is an extended Kalman filter (EKF). The battery model on which the EKF algorithm relies is a second-order RC equivalent circuit model, in which parameters such as resistance and capacitance are not fixed values, but pre-calibrated, temperature-dependent functions. The state determination module M200 receives the battery terminal voltage, load current, and ambient temperature signals from the signal acquisition module M100, and, based on the current ambient temperature, looks up the corresponding battery model parameters at a pre-stored lookup table. Then, it substitutes these parameters into the EKF algorithm to estimate the battery's SOC and equivalent internal resistance. In other optional implementations, other types of battery models, such as first-order RC models, Thevenin models, etc., and other types of filtering algorithms, such as unscented Kalman filtering (UKF), particle filtering, etc., can be used.

[0117] The index calculation module M300 is configured to integrate the state of charge (SOC) and equivalent internal resistance to calculate an instability index characterizing power supply stability. In one implementation, the index calculation module M300 uses a weighted average method to integrate the SOC and equivalent internal resistance. First, the index calculation module M300 calculates the remaining energy decay rate, i.e., (1 - SOC), where SOC is the SOC estimated by the aforementioned state determination module M200. Then, it calculates the output capability degradation rate, i.e. ,in It is the equivalent internal resistance estimated by the state determination module M200. This is the baseline internal resistance of the battery in its brand-new state. Ultimately, the instability index... Where w1 and w2 are preset weighting coefficients. In other optional implementations, other fusion algorithms can be used, such as neural networks, fuzzy logic, etc., as long as they can comprehensively consider the impact of state of charge and equivalent internal resistance on power supply stability.

[0118] The strategy generation module M400 is configured to generate a power allocation strategy based on the instability index and a preset task power requirement table that defines the priority and power consumption requirements for different functional modules. In this embodiment, the task power requirement table is stored in the processor's Flash memory, which defines information such as the priority level, normal operating power consumption, and minimum survival power consumption of each functional module within the terminal. The strategy generation module M400 receives the instability index PII output by the index calculation module M300 and determines the current system risk level based on the PII value. Then, the strategy generation module M400 generates a power allocation strategy with reference to the task power requirement table. This strategy specifies how much power should be allocated to each functional module under the current risk level. For example, under a low risk level, all functional modules can operate normally; under a high risk level, some non-modules need to be shut down to ensure the operation of the modules. In other optional implementations, other types of priority scheduling algorithms can be used, such as EDF (Earliest Deadline First) and RM (Rate Monotonic).

[0119] The regulation control module M500 is configured to translate power allocation strategies into control commands and send them to the programmable power management unit (PMU) for physical regulation of the power supply to at least one non-functional module within the terminal. In one embodiment, the PMU is a TI TPS65086x series chip. The regulation control module M500 communicates with the PMU via an I2C bus, translating the power allocation strategy into configuration commands for the PMU's internal registers. For example, the power supply voltage of a non-functional module can be reduced by adjusting the output voltage of the PMU's DC-DC converter, thereby reducing its power consumption. Alternatively, the power supply to a non-functional module can be directly disconnected via a load switch integrated into the PMU. In other alternative implementations, other types of power management chips can be used, such as Analog Devices' LTC series or Maxim Integrated's MAX series.

[0120] The signal acquisition module M100, state determination module M200, exponent calculation module M300, strategy generation module M400, and regulation and control module M500 work together in a coordinated manner. The signal acquisition module M100 provides raw data to the state determination module M200. After processing the raw data, the state determination module M200 outputs the battery's state of charge and equivalent internal resistance. The exponent calculation module M300 combines the state of charge and equivalent internal resistance to obtain an instability index characterizing power supply stability. The strategy generation module M400 generates a power allocation strategy based on the instability index and the task power consumption requirement table. The regulation and control module M500 translates the power allocation strategy into control commands, controlling the programmable power management unit to adjust the power supply to the functional modules within the terminal, ultimately ensuring the operation of the data acquisition terminal. This cooperative relationship can solve the problem of unstable power supply caused by battery performance degradation in existing data acquisition terminals in extremely cold environments.

[0121] Through the above solution, this embodiment can dynamically adjust the power consumption allocation strategy according to the actual state of the battery and the priority of the task, thereby ensuring the stable operation of the functional modules and extending the effective working time of the terminal.

[0122] In one specific implementation, the strategy generation module M400 is configured to generate a power allocation strategy through iterative optimization using a genetic algorithm. This genetic algorithm runs internally to the processor and interacts with the task power requirement table and battery status information. Specifically, the strategy generation module M400 includes the following sub-modules: The encoding module is configured to discretize the power consumption level of each controllable module within the terminal into a finite number of levels, and encode the power consumption levels of all modules into an individual in a genetic algorithm. Each individual can be represented as a vector, with each element of the vector corresponding to the power consumption level of a module. For example, if the terminal has 5 controllable modules, and each module has 10 power consumption levels, then an individual can be represented as [2, 5, 8, 1, 9], representing the power consumption levels currently assigned to these 5 modules.

[0123] The fitness evaluation module is configured to calculate the fitness value for each individual based on the current battery state (including state of charge (SOC) and equivalent internal resistance) and the task power consumption requirement table. The fitness value reflects the merits of the power allocation scheme represented by the individual.

[0124] The selection module is configured to select a subset of individuals from the population to enter the next generation based on their fitness values. Individuals with higher fitness values ​​have a greater probability of being selected. Selection can be performed using methods such as roulette wheel selection or tournament selection.

[0125] The crossover module is configured to perform crossover operations on selected individuals to generate new individuals. Crossover involves exchanging partial genes between two individuals to create new combinations. Crossover operations can be performed using methods such as single-point crossover, multi-point crossover, and uniform crossover.

[0126] The mutation module is configured to perform mutation operations on individuals to increase population diversity. A mutation operation involves randomly changing the value of one or more genes within an individual. Mutation operations can take the form of site mutation, reverse mutation, and other methods.

[0127] During the iterative process of the genetic algorithm, the policy generation module M400 continuously performs selection, crossover, and mutation operations, ultimately finding an individual with the highest fitness. The power allocation scheme represented by this individual is the current optimal power allocation strategy. This strategy is then converted into control instructions and sent to the programmable power management unit for physical adjustment. In other optional implementations, the encoding method can use real number encoding or symbolic encoding; the selection operation can use sorting selection, elite selection, etc.; the crossover operation can use arithmetic crossover, heuristic crossover, etc.; and the mutation operation can use Gaussian mutation, non-uniform mutation, etc.

[0128] Through the above scheme, this embodiment can utilize the global optimization capability of the genetic algorithm to reduce the total power consumption of the terminal as much as possible while meeting the minimum power consumption requirements of the functional modules, thereby maximizing the effective working time of the terminal while ensuring stable operation of the functions.

[0129] In one specific implementation, the fitness function used by the genetic algorithm is designed to include a penalty term that constrains the power consumption allocation of the functional modules. Specifically, the task power consumption requirement table predefines a minimum survival power consumption threshold for each functional module. ,in This represents the i-th functional module. The design idea of ​​the fitness function is: if the genetic algorithm generates any power allocation scheme during the iteration process, such that the actual power consumption allocated to a certain functional module is... Below its corresponding minimum survival power threshold If this happens, the fitness value of the power allocation scheme will be significantly reduced, thus eliminating it in the subsequent genetic selection process.

[0130] The mathematical expression for the fitness function is as follows:

[0131]

[0132] in, It is the power consumption of the i-th module. It is the total number of modules. It is the total power consumption weight, used to balance the impact between total power consumption and the penalty term. Represents a collection of functional modules. This is a penalty term for module j, and its calculation method is as follows:

[0133] in It is a pre-defined, extremely large constant, for example... This is used to ensure that any power allocation scheme that violates the minimum survivability power constraint is quickly eliminated. If the actual power consumption of module j... Below its minimum survivability power consumption Then the penalty item will be activated, and its value will be... Multiplying by the difference between the two will result in Significantly increased, thus making this power allocation scheme... The value decreases significantly. Conversely, if the actual power consumption of module j is greater than or equal to its minimum survivability power consumption, the penalty term is 0, which is incorrect. The value has an impact.

[0134] In other alternative implementations, the penalty term can be calculated using different functional forms, such as quadratic or exponential functions, as long as it ensures that the fitness value is significantly penalized when the module's power consumption falls below the minimum survivability power consumption. Furthermore, the weighting coefficients of the penalty term... It can also be adjusted according to the actual application scenario to balance the strictness of the constraints and the convergence speed of the genetic algorithm.

[0135] Through the above scheme, this embodiment can ensure that the genetic algorithm always takes the minimum survival requirement of the functional module as a hard constraint when optimizing power allocation, thereby avoiding the failure of functions due to excessive reduction of power consumption, and improving the overall stability and reliability of the system.

[0136] In one specific implementation, the system further includes a separate monitoring module. This module interacts with the wireless communication module (e.g., an NB-IoT communication module based on the Quectel EG915U module) via a standard API interface (e.g., the ioctl system call or the interface provided by LWIP), and is configured to passively monitor and analyze the operating status of the wireless communication module. This monitoring module includes the following sub-modules: Data Link Quality Monitoring Submodule: This submodule continuously acquires data link quality parameters from the wireless communication module. In one implementation, these parameters include: Data retransmission rate: This parameter represents the proportion of data packets that need to be retransmitted within a certain time window (e.g., 10 minutes). This value directly reflects the reliability of the current data link.

[0137] Bit Error Rate (BER): The current estimated bit error rate can be obtained by reading the internal registers of the communication module.

[0138] Network Environment Quality Assessment Submodule: This submodule queries the wireless communication module using AT commands (e.g., AT+CSQ) to obtain the current network signal strength indicator (RSSI) and signal-to-noise ratio (SNR). These parameters reflect the quality of the external wireless network environment in which the terminal is located.

[0139] Adaptive Correction Trigger Submodule: This submodule performs a comprehensive analysis based on the data provided by the two submodules mentioned above to determine whether to trigger a correction to the state determination module M200. Its logic is as follows: First, it checks whether the RSSI value provided by the network environment quality assessment submodule is lower than a preset threshold (e.g., -95dBm). If it is lower than this threshold, the poor communication quality is considered to be caused by a poor external network environment, and no correction operation is performed. Conversely, if the RSSI value is higher than this threshold, it further checks whether the data retransmission rate provided by the data link quality monitoring submodule is higher than a preset threshold (e.g., 15%). If it is higher than this threshold, the poor communication quality is considered to be caused by unstable battery power, and a correction operation is triggered.

[0140] In one implementation, the correction operation is triggered by the adaptive correction triggering submodule sending a correction command to the state determination module M200 via an inter-process communication mechanism (e.g., a message queue). This command includes a parameter indicating the correction magnitude. Upon receiving the command, the state determination module M200 adjusts the process noise covariance matrix Q in the extended Kalman filter algorithm based on this parameter. More specifically, it increases the values ​​of the Q matrix elements related to the SOC state by a preset percentage (e.g., 20%).

[0141] In other alternative implementations, the data link quality monitoring submodule can also monitor other parameters such as packet loss rate and latency jitter to more comprehensively assess the quality of the communication link; the network environment quality assessment submodule can also use the location information obtained by the GNSS module, combined with pre-stored base station geographical location information, to calculate the distance between the terminal and the base station, as an auxiliary basis for judging the quality of the external network environment; in addition, the adaptive correction triggering submodule can use more complex algorithms such as fuzzy logic or neural networks to comprehensively evaluate various input parameters, thereby more accurately determining whether a correction operation needs to be triggered.

[0142] Through the above scheme, this embodiment can monitor the performance of upper-layer applications (wireless communication) and reverse-correct the state estimation algorithm of the lower layer, thereby achieving adaptive perception of unforeseen degradation risks.

[0143] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such modifications and substitutions should be considered within the scope of protection of the present invention.

Claims

1. A method for ensuring the operation of a low-power, cold-resistant data acquisition terminal, characterized in that, include: The terminal's battery voltage, load current, and ambient temperature signals are acquired in real time through sensors. The processor uses a battery model associated with the ambient temperature signal and a Kalman filter algorithm to process the battery terminal voltage and load current signals to determine the battery's state of charge and equivalent internal resistance. The processor integrates the state of charge and the equivalent internal resistance to calculate an instability index that characterizes the stability of the power supply. The processor generates a power allocation strategy based on the instability index and a preset task power requirement table that defines the priority and power requirements of different functional modules. The processor converts the power allocation strategy into control instructions and sends them to the programmable power management unit, which then physically adjusts the power supply to at least one non-core functional module within the terminal.

2. The method according to claim 1, characterized in that, The specific steps for generating the power allocation strategy are as follows: using a genetic algorithm for iterative optimization to generate the power allocation strategy with the goal of minimizing the total power consumption of the terminal and ensuring the operation of the core functional modules.

3. The method according to claim 2, characterized in that, The fitness function used in the genetic algorithm includes a penalty term; when the power consumption allocated to any core functional module is lower than the minimum survival power consumption defined in the task power consumption requirement table, the penalty term is activated to reduce the fitness value of the power allocation strategy. The constraints of the iterative optimization include: the total power consumption of the terminal does not exceed the current maximum available output power calculated in real time by the battery model.

4. The method according to claim 1, characterized in that, The Kalman filter algorithm is the extended Kalman filter algorithm, and the battery model is a second-order RC equivalent circuit model.

5. The method according to claim 1, characterized in that, The step of calculating the instability index includes: calculating the remaining energy decay rate based on the state of charge, calculating the output capability degradation rate based on the equivalent internal resistance, and performing a weighted summation of the remaining energy decay rate and the output capability degradation rate.

6. The method according to claim 1, characterized in that, The physical adjustment specifically involves reducing the power supply voltage of the non-core functional modules or performing power gating to interrupt their power supply.

7. The method according to claim 1, characterized in that, The task power consumption requirement table defines at least the priority level, normal operating power consumption, and minimum survival power consumption for each functional module.

8. The method according to claim 1, characterized in that, The method further includes: When the instability index exceeds the preset emergency survival threshold for the first time, the core survival mode is triggered and entered. After entering the core survival mode, the data acquisition module and non-volatile storage module are driven first to perform a final data snapshot acquisition and forced storage operation; After confirming that the final data snapshot has been successfully stored, the wireless communication module is activated to enter the preset beacon communication mode. The steps of performing the final data snapshot acquisition and forced storage operation include: temporarily suspending the data transmission function of the wireless communication module, and driving the data acquisition module and the non-volatile storage module with the highest priority and maximum available power; In the beacon communication mode, the wireless communication module periodically sends heartbeat packets containing only the device ID, the last data snapshot timestamp, and the status code at time intervals that are dynamically adjusted according to the remaining battery power.

9. The method according to claim 1, characterized in that, The method further includes: The data communication link quality parameters of the terminal's wireless communication module are monitored in parallel. When the communication link quality parameters show a continuous abnormal deterioration, the model credibility-related parameters in the Kalman filter algorithm are dynamically corrected. The steps for determining the occurrence of persistent abnormal degradation include: Assuming the external network signal strength of the terminal is normal, determine the communication link quality parameters; The dynamic correction specifically refers to: The process noise covariance matrix Q value in the Kalman filter algorithm is automatically increased to reduce the dependence on the battery model and enhance the tracking of the current battery terminal voltage and load current signals.

10. An operation support system for a low-power, cold-resistant data acquisition terminal, characterized in that, include: The signal acquisition module is configured to acquire the battery terminal voltage, load current and ambient temperature signals of the terminal in real time. A state determination module is configured to determine the state of charge and equivalent internal resistance of the battery based on a battery model associated with the ambient temperature signal and by processing the battery terminal voltage and load current signals using a Kalman filter algorithm. An index calculation module is configured to integrate the state of charge and the equivalent internal resistance to calculate an instability index characterizing the stability of the power supply. The strategy generation module is configured to generate a power allocation strategy based on the instability index and a preset task power requirement table that defines the priority and power requirements for different functional modules. The adjustment control module is configured to convert the power consumption allocation strategy into control commands and send them to the programmable power management unit for physical adjustment of the power supply to at least one non-core functional module within the terminal.

Images