An environment adaptive low power management system and method for an electronic timer

Abstract

The present application relates to the technical field of electronic timer, and especially to an environment adaptive low power consumption management system and method for electronic timer, the system comprising an environment perception array, a physical state evaluator, a dynamic power consumption strategy generator, a closed loop execution and calibration unit and an adaptive compensator; the present application dynamically collects multi-dimensional environment parameters, constructs a coupling model representing the relationship between frequency offset and compensation power consumption, combines the residual power of the device and the cumulative value of timing error, generates joint control parameters through multi-objective optimization, drives hardware execution and real-time monitors operation deviation, and realizes adaptive iterative optimization of strategy through feedback closed loop. The present application effectively solves the pain point of mutual restriction between precision and power consumption of traditional electronic timer, greatly improves the environment adaptive ability, endurance time and whole life cycle operation stability of the device.

Description

Technical Field

[0001] This invention relates to the field of electronic timer technology, specifically to an environment-adaptive low-power management system and method for electronic timers. Background Technology

[0002] Electronic timers, as core components for time measurement and timing control, are widely used in many fields such as industrial measurement and control, consumer electronics, and portable outdoor equipment. Timing accuracy and continuous battery life are the core indicators that determine their product performance and application boundaries.

[0003] Most mainstream electronic timers currently use crystal oscillators as their timing reference. The resonant frequency of these components is easily affected by external factors such as ambient temperature, mechanical vibration, and photothermal effects caused by ambient light, resulting in uncontrollable frequency shifts. This directly leads to accumulated timing errors and affects the accuracy of the equipment. Traditional solutions to address frequency shifts caused by environmental factors often employ fixed-parameter hardware temperature compensation circuits or open-loop, fixed-period compensation strategies, which cannot accurately match the compensation intensity according to dynamic environmental changes. These solutions either maintain full-load compensation operation to ensure timing accuracy, resulting in significant unnecessary power consumption and drastically shortening the device's battery life, or reduce compensation actions to control power consumption, failing to cope with error exceeding limits caused by sudden environmental changes. Ultimately, they cannot overcome the industry bottleneck of the trade-off between timing accuracy and operating power consumption.

[0004] Traditional electronic timers often employ a single-module independent control mode for power consumption management, setting fixed power control strategies only for the display panel or power management unit. This approach fails to achieve coordinated optimization of multiple core power consumption units, such as temperature compensation circuits, frequency division counting units, and display modules, making it difficult to minimize overall system power consumption and severely limiting the potential for improving device endurance. Furthermore, the traditional open-loop management architecture lacks a complete closed-loop feedback and adaptive compensation mechanism, making it unable to adapt to the characteristic drift and hardware aging issues of timing reference components after long-term use. This leads to a continuous decline in timing accuracy and a gradual increase in power consumption after prolonged operation, requiring frequent manual calibration and maintenance, resulting in high operating costs and insufficient long-term stability and reliability.

[0005] In summary, existing technologies have not yet formed an integrated low-power management solution that can perceive multi-dimensional environmental changes in real time, simultaneously achieve accurate prediction of frequency offset, collaborative optimization of multi-target power consumption, closed-loop execution calibration, and adaptive strategy updates. This makes it difficult to meet the core usage requirements of electronic timers for high precision, long battery life, and high reliability in complex application scenarios. Summary of the Invention

[0006] The purpose of this invention is to provide an environment-adaptive low-power management system and method for electronic timers, in order to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] An environment-adaptive low-power management system for an electronic timer includes:

[0009] An environmental sensing array is used to acquire multi-dimensional environmental parameters at a dynamically adjustable sampling frequency. These multi-dimensional environmental parameters include at least ambient temperature, vibration spectrum, and ambient illuminance.

[0010] The physical state evaluator, connected to the environmental sensing array, is configured to construct an environment-error coupling model for the current moment based on multi-dimensional environmental parameters. The environment-error coupling model is used to characterize the functional relationship between the expected frequency offset of the timing reference element of the electronic timer under the current environmental parameters and the temperature compensation power consumption.

[0011] A dynamic power consumption strategy generator, connected to a physical state evaluator, is configured to input the environment-error coupling model, the current remaining power of the electronic timer, and the current accumulated timing error into a multi-objective optimization decision network, and output a set of joint control parameters. The joint control parameters include at least the duty cycle of the temperature compensation circuit of the timing reference element, the power supply voltage level of the frequency division counting unit, and the refresh rate threshold of the display panel.

[0012] The closed-loop execution and calibration unit is connected to the functional modules of the dynamic power consumption strategy generator and the electronic timer, respectively. It is configured to parse the joint control parameters into hardware control commands and send them out for execution, while monitoring the actual frequency offset of the timing reference element and the actual power consumption of the display panel in real time to generate a measured deviation vector.

[0013] An adaptive compensator, connected to the closed-loop execution and calibration unit and the dynamic power consumption strategy generator respectively, is configured to feed back the measured deviation vector to the multi-objective optimization decision network, triggering the multi-objective optimization decision network to perform online parameter optimization in the neighborhood of the joint control parameters with the goal of minimizing the weighted sum of timing error and power consumption per unit time, and update the optimization result to the strategy benchmark of the dynamic power consumption strategy generator.

[0014] An environment-adaptive low-power management method for electronic timers, executed by a system, includes the following steps:

[0015] Step S1: Collect multi-dimensional environmental parameters at a dynamically adjustable sampling frequency using an environmental sensing array. The multi-dimensional environmental parameters include at least ambient temperature, vibration spectrum, and ambient illuminance.

[0016] Step S2: Receive the multi-dimensional environmental parameters generated in step S1 through the physical state evaluator, and construct the environment-error coupling model at the current moment based on the multi-dimensional environmental parameters. The environment-error coupling model is used to characterize the functional relationship between the expected frequency offset of the timing reference element of the electronic timer under the current environmental parameters and the temperature compensation power consumption.

[0017] Step S3: Receive the environment-error coupling model generated in step S2 through the dynamic power consumption strategy generator, and input the environment-error coupling model together with the current remaining power of the electronic timer and the current accumulated timing error into the multi-objective optimization decision network, and output a set of joint control parameters. The joint control parameters include at least: the duty cycle of the temperature compensation circuit of the timing reference element, the power supply voltage level of the frequency division counting unit, and the refresh rate threshold of the display panel.

[0018] Step S4: The closed-loop execution and calibration unit receives the joint control parameters output in step S3, parses the joint control parameters into hardware control instructions and sends them to the functional modules of the electronic timer for execution. At the same time, it monitors the actual frequency offset of the timing reference element and the actual power consumption of the display panel in real time and generates a measured deviation vector.

[0019] Step S5: Receive the measured deviation vector generated in step S4 through the adaptive compensator, and feed the measured deviation vector back to the multi-objective optimization decision network in step S3. Trigger the multi-objective optimization decision network to perform online parameter optimization in the neighborhood of the joint control parameters with the goal of minimizing the weighted sum of timing error and power consumption per unit time. Update the optimization results to the policy benchmark of the dynamic power consumption strategy generator to replace the original joint control parameters in step S3 in the next round of control command issuance.

[0020] As can be seen from the technical solutions provided by the present invention above, the environmentally adaptive low-power management system and method for electronic timers provided by the present invention have the following beneficial effects:

[0021] This invention achieves dynamic acquisition of multi-dimensional environmental parameters such as temperature, vibration, and illuminance through an environmental sensing array. It constructs a coupled model of environmental stress and frequency offset of timing reference elements, which can accurately quantify the impact of environmental changes on timing accuracy, predict frequency offset trends in advance, and adjust compensation strategies accordingly. Whether it is a complex scenario with wide temperature range changes, strong mechanical vibration, or frequent changes in light intensity, the system can respond and adapt quickly, avoiding timing errors caused by sudden environmental changes. This allows the electronic timer to be stably adapted to various complex application scenarios such as industrial sites, outdoor portable devices, and consumer products, greatly expanding the product's applicability.

[0022] This invention achieves coordinated optimization of the entire system's power consumption through the joint regulation of three core power consumption modules: temperature compensation circuit, frequency division counting unit, and display panel. Compared with power consumption optimization of a single module, it can achieve a more significant power reduction effect. Under the same battery capacity and timing accuracy requirements, it can greatly extend the continuous working time of the device, reduce the frequency of battery replacement, and reduce the cost of device use and maintenance. At the same time, the closed-loop execution and calibration mechanism can ensure that the hardware modules always operate within the rated safe range, and the adaptive compensation mechanism can adapt to the performance changes caused by the drift and aging of hardware components, avoid hardware overload operation, and effectively extend the service life of core components and the whole machine.

[0023] This invention forms a complete closed-loop management chain from environmental perception, decision generation, execution monitoring to feedback optimization by combining real-time status monitoring of the closed-loop execution and calibration unit with online parameter optimization of the adaptive compensator. The system can continuously iterate and optimize the control strategy based on the measured deviation of the actual hardware operation. Without manual intervention and on-site debugging, it can autonomously adapt to the effects of changes in device characteristics, application scenario switching and long-term hardware aging, ensuring that the equipment always maintains the optimal operating state throughout its entire life cycle. This avoids the problems of decreased accuracy and increased power consumption after long-term use, and ensures the stability and reliability of the product's long-term operation.

[0024] This invention achieves coordinated control of timing accuracy and power consumption through intelligent optimization of software algorithms, significantly reducing the reliance on high-specification timing reference components and high-precision hardware temperature compensation circuits. Mid-to-low-end hardware platforms can achieve high-performance timing accuracy control and low-power operation through the management system and method of this invention, without the need to add complex hardware compensation circuits and high-cost core components. This effectively simplifies the hardware design architecture, reduces product manufacturing costs and development cycles, and enhances the product's market competitiveness.

[0025] Throughout the entire process of strategy generation and parameter optimization, this invention consistently adheres to multiple constraints, including the hardware's rated operating range, the upper limit of timing accuracy, and the lower limit of battery life. All output control parameters undergo rigorous constraint verification, effectively preventing hardware damage and operational failures caused by abnormal parameters and ensuring the safety of equipment operation. Simultaneously, the system's fully autonomous closed-loop operation mode eliminates the need for manual user setup and debugging, automatically optimizing operating strategies based on usage scenarios and equipment status, significantly improving product usability and user experience. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of an environment-adaptive low-power management system for an electronic timer according to the present invention.

[0027] Figure 2This is a schematic diagram of the steps of an environmentally adaptive low-power management method for an electronic timer according to the present invention. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0029] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific embodiments.

[0030] like Figure 1-2 As shown, this embodiment of the invention provides an environment-adaptive low-power management system for an electronic timer, comprising:

[0031] An environmental sensing array is used to acquire multi-dimensional environmental parameters at a dynamically adjustable sampling frequency. These multi-dimensional environmental parameters include at least ambient temperature, vibration spectrum, and ambient illuminance.

[0032] The physical state evaluator, connected to the environmental sensing array, is configured to construct an environment-error coupling model for the current moment based on multi-dimensional environmental parameters. The environment-error coupling model is used to characterize the functional relationship between the expected frequency offset of the timing reference element of the electronic timer under the current environmental parameters and the temperature compensation power consumption.

[0033] A dynamic power consumption strategy generator, connected to a physical state evaluator, is configured to input the environment-error coupling model, the current remaining power of the electronic timer, and the current accumulated timing error into a multi-objective optimization decision network, and output a set of joint control parameters. The joint control parameters include at least the duty cycle of the temperature compensation circuit of the timing reference element, the power supply voltage level of the frequency division counting unit, and the refresh rate threshold of the display panel.

[0034] The closed-loop execution and calibration unit is connected to the functional modules of the dynamic power consumption strategy generator and the electronic timer, respectively. It is configured to parse the joint control parameters into hardware control commands and send them out for execution, while monitoring the actual frequency offset of the timing reference element and the actual power consumption of the display panel in real time to generate a measured deviation vector.

[0035] An adaptive compensator, connected to the closed-loop execution and calibration unit and the dynamic power consumption strategy generator respectively, is configured to feed back the measured deviation vector to the multi-objective optimization decision network, triggering the multi-objective optimization decision network to perform online parameter optimization in the neighborhood of the joint control parameters with the goal of minimizing the weighted sum of timing error and power consumption per unit time, and update the optimization result to the strategy benchmark of the dynamic power consumption strategy generator.

[0036] In this embodiment, the environmental sensing array is the core sensor for the environmental adaptive low-power management system of the electronic timer to sense changes in the external environment. It completes the accurate acquisition of multi-dimensional environmental parameters through a dynamically adjustable sampling mechanism, providing basic data support for subsequent error assessment and power consumption strategy generation of the system.

[0037] The core function of the environmental sensing array is to acquire multi-dimensional environmental parameters at a dynamically adjustable sampling frequency. These parameters cover at least three core indicators: ambient temperature, vibration spectrum, and ambient illuminance. The array achieves a dynamic balance between environmental sensing accuracy and operating power consumption through a combination of low-power intermittent monitoring and high-precision synchronous acquisition. Simultaneously, it performs standardized preprocessing and encapsulation of the acquired signals, outputting multi-dimensional environmental parameters in a unified format to the physical state evaluator. This serves as the data source foundation for the entire system's environmental self-adaptation capability and is a crucial pre-processing step in generating the system's low-power management strategy. The environmental sensing array specifically includes:

[0038] Low power monitoring unit:

[0039] Coarse-grained environmental monitoring: It acquires coarse-grained change values ​​of environmental parameters according to a preset basic sleep-wake cycle. During non-working periods, it enters a deep sleep state, retaining only the minimum power supply. After waking up, it quickly completes coarse sampling of three types of environmental parameters: temperature, vibration, and illuminance, to obtain basic data that can reflect the overall trend of environmental change, and transmits the coarse-grained change values ​​to the frequency scheduling unit.

[0040] Sleep-wake scheduling: Strictly follow the preset basic sleep-wake cycle to switch working states. After completing a single coarse-grained sampling and data transmission, immediately switch back to deep sleep state to maximize the proportion of sleep time, fundamentally reduce the basic power consumption of the environmental monitoring link, and avoid unnecessary power consumption caused by continuous sampling.

[0041] Frequency scheduling unit:

[0042] Environmental change fluctuation calculation: The system receives coarse-grained change values ​​from the low-power monitoring unit, calculates the environmental state fluctuation rate at the current moment, quantifies the severity of environmental parameter changes, and provides a quantitative basis for dynamic adjustment of the sampling frequency. The formula for calculating the environmental state fluctuation rate is:

[0043] ,in, The environmental state volatility at the current moment. This refers to the number of samples in coarse-grained sampling. For the first Environmental parameter measurements from secondary coarse-grained sampling for Average values ​​of environmental parameters from secondary coarse-grained sampling;

[0044] Dynamic sampling frequency correction: A frequency adjustment coefficient is generated based on the environmental state fluctuation rate. This coefficient is then used to correct the preset base sampling frequency, resulting in a dynamically adjustable sampling frequency. The formula for calculating the frequency adjustment coefficient is as follows:

[0045] ,in, This is the frequency adjustment factor. This is the volatility response coefficient, used to control the sensitivity of the sampling frequency to environmental changes;

[0046] The formula for calculating the dynamically adjustable sampling frequency is:

[0047] ,in, For dynamically adjustable sampling frequency, The preset base sampling frequency;

[0048] When environmental fluctuations are severe, the sampling frequency is increased to ensure data integrity, and when the environment is stable, the sampling frequency is reduced to control operating power consumption, thus achieving dynamic adaptation between sampling accuracy and power consumption.

[0049] Sampling command issuance: The final determined dynamic adjustable sampling frequency is transmitted to the high-precision acquisition unit, triggering the start of the high-precision acquisition process and ensuring that the acquisition action is accurately matched with the environmental change state;

[0050] High-precision acquisition unit:

[0051] Multi-sensor synchronous acquisition: Based on a dynamically adjustable sampling frequency and a unified synchronous clock signal, the temperature sensing component, vibration sensing component, and photosensitive sensing component are driven to acquire data synchronously to obtain ambient temperature data, raw vibration signal, and ambient illuminance data, ensuring the time synchronization of the three types of parameters and eliminating time deviation of multi-source data.

[0052] Signal preprocessing and conversion: The original vibration signal is converted into a time-frequency domain through a fast Fourier transform, and the time-domain vibration signal is converted into a frequency-domain vibration spectrum. This completes the statistical processing of the frequency band energy distribution of the vibration signal, providing quantifiable frequency domain data for subsequent mechanical interference feature extraction.

[0053] Data standardization and encapsulation: Ambient temperature data, vibration spectrum and ambient illuminance data are processed in a unified format and encapsulated into standardized multidimensional environmental parameters. These multidimensional environmental parameters are then stably output to the physical state evaluator, providing standardized basic data for subsequent model building and strategy generation.

[0054] In this embodiment, the physical state evaluator is the core analysis unit for the environmental adaptive low power management system of the electronic timer. It receives the multi-dimensional environmental parameters output by the environmental sensing array, completes the in-depth mining of environmental stress characteristics and the accurate prediction of the frequency offset of the timing reference element, and finally constructs the core model characterizing the relationship between environmental parameters and power consumption error. It is the core analysis link for realizing environmental adaptive low power management.

[0055] The physical state evaluator is connected to the environmental sensing array. Its core function is to construct an environment-error coupling model for the current moment based on multi-dimensional environmental parameters. This model characterizes the functional relationship between the expected frequency offset of the electronic timer's timing reference element under the current environmental parameters and the temperature compensation power consumption. By extracting deep features from the multi-dimensional environmental parameters, the evaluator quantifies the comprehensive stress exerted by the current environment on the timing reference element, accurately predicts the frequency offset trend caused by environmental changes, and quantifies the power consumption cost required to offset the frequency offset, thus building a core bridge between environmental sensing data and power optimization decisions. The environment-error coupling model output by the evaluator is the core constraint for generating subsequent dynamic power consumption strategies and the core foundation for the entire system to achieve a balance between timing accuracy and operating power consumption. The physical state evaluator specifically includes:

[0056] Feature extraction subunit:

[0057] The core function of the feature extraction subunit is to receive multi-dimensional environmental parameters, extract and fuse multi-dimensional environmental features, generate standardized environmental stress feature vectors, and provide high-quality input features for subsequent frequency shift prediction.

[0058] Multi-dimensional environmental feature extraction: Time-domain sliding variance calculation is performed on the environmental temperature data to generate temperature fluctuation feature values, quantifying the long-term trend and intensity of environmental temperature fluctuations; the calculation formula is as follows:

[0059] ,in, These are characteristic values ​​of temperature fluctuation. The length of the sliding window. For the first in the sliding window Ambient temperature data at each moment, This represents the average ambient temperature data within the sliding window.

[0060] The vibration spectrum is statistically analyzed for frequency band energy distribution to generate mechanical interference characteristic values, quantifying the mechanical vibration intensity affecting the timing reference element; the calculation formula is as follows:

[0061] ,in, For mechanical interference characteristic values, The frequency coordinates of the vibration spectrum to For the frequency band sensitive to mechanical vibration of the timing reference element, The amplitude of the vibration spectrum at the corresponding frequency point;

[0062] A step change identification is performed on the ambient illuminance data to generate photothermal effect feature values, quantifying the photothermal effect caused by abrupt changes in ambient illuminance; the calculation formula is as follows:

[0063] ,in, The characteristic value of photothermal influence, This represents the number of illuminance data samples within the current sampling period. For the first Environmental illuminance data at each sampling time, For the first Environmental illuminance data at each sampling time;

[0064] Multi-feature fusion processing: Temperature fluctuation feature values, mechanical disturbance feature values, and photothermal effect feature values ​​are standardized and fused to generate an environmental stress feature vector, which fully characterizes the comprehensive stress of the current environment on the timing reference element; the calculation formula is:

[0065] ,in, This represents the characteristic vector of environmental stress. These are characteristic values ​​of temperature fluctuation. For mechanical interference characteristic values, These are characteristic values ​​related to the effects of photothermal radiation.

[0066] The feature extraction subunit outputs the fused environmental stress feature vector to the frequency drift prediction subunit, providing a standardized input for subsequent frequency shift prediction;

[0067] Frequency drift prediction subunit:

[0068] The frequency drift prediction subunit is connected to the feature extraction subunit. Its core function is to receive the environmental stress feature vector, match the physical characteristics of the timing reference element, and output the expected frequency offset under the current environment, providing core parameters for the subsequent construction of the coupled model.

[0069] Feature matching and trend retrieval: The system receives the environmental stress feature vector output by the feature extraction subunit, performs similarity matching with feature samples in a pre-trained device physical property mapping library, retrieves the set of samples closest to the current environmental stress features, and extracts the resonant frequency variation trend data of the timing reference element under the corresponding sample. The similarity calculation formula is:

[0070] in, The cosine similarity of the feature vectors. This is the current input environmental stress feature vector. For the first device physical property mapping library Feature vectors of each sample;

[0071] Expected frequency offset calculation: Based on the set of highly similar samples obtained through matching, and combined with the corresponding similarity weights, the expected frequency offset of the timing reference element under the current environmental stress is calculated; the calculation formula is:

[0072] ,in, This is the expected frequency offset. The number of highly similar samples obtained through matching. For the first The cosine similarity corresponding to each highly similar sample For the first Frequency offset of timing reference element corresponding to each highly similar sample;

[0073] The frequency drift prediction subunit outputs the calculated expected frequency offset to the coupling relationship construction subunit, providing a core basis for power consumption quantization calculation;

[0074] Coupling relationships construct sub-units:

[0075] The coupling relationship construction sub-unit is connected to the frequency drift prediction sub-unit. Its core function is to quantify the corresponding compensation power consumption based on the expected frequency offset and construct an environment and error coupling model that characterizes the functional relationship between frequency offset and temperature compensation power consumption.

[0076] Quantitative calculation of power consumption compensation: Based on the expected frequency offset output by the frequency drift prediction subunit, the temperature compensation power consumption required to completely offset this frequency offset is calculated, clarifying the quantitative correspondence between precise control requirements and power consumption; the calculation formula is:

[0077] ,in, To compensate for power consumption by temperature, The power consumption factor is determined by the hardware characteristics of the temperature compensation circuit. This represents the expected frequency offset.

[0078] Functional Relationship Fitting and Model Construction: Using the expected frequency offset as the independent variable and the corresponding required temperature compensation power consumption as the dependent variable, polynomial curve fitting is performed to generate a continuously differentiable environment-error coupling model, which fully characterizes the functional relationship between frequency offset and temperature compensation power consumption under the current environment; the calculation formula is:

[0079] ,in, For the output temperature compensation power consumption of the environmental and error coupling model, This refers to the frequency offset of the timing reference element. To determine the order of the fitted polynomial, To obtain the fitted first Coefficients of a polynomial of order 1;

[0080] The coupling relationship construction sub-unit outputs the completed environment and error coupling model to the dynamic power consumption strategy generator, providing core constraints for multi-objective optimization decision-making.

[0081] In this embodiment, the dynamic power consumption strategy generator is the core decision-making hub for the environmental adaptive low-power management system of the electronic timer to achieve a balance between timing accuracy and operating power consumption. Through the fusion analysis of multi-dimensional input information and multi-objective intelligent optimization, it outputs joint control parameters that adapt to the current environmental state and equipment operating conditions, providing accurate decision-making basis for the stable low-power operation of the system in all scenarios.

[0082] The dynamic power consumption strategy generator is primarily responsible for receiving the environment and error coupling model output by the physical state evaluator, synchronously collecting the current remaining power and current accumulated timing error value of the electronic timer, and inputting these three types of core information into the multi-objective optimization decision network to complete intelligent decision-making. The module aims to minimize the weighted sum of timing error and power consumption per unit time, optimizing parameters within the boundaries of hardware specifications and performance constraints. The final output includes joint control parameters for the duty cycle of the temperature compensation circuit of the timing reference element, the power supply voltage level of the frequency division counting unit, and the refresh rate threshold of the display panel. The module can dynamically adjust the priority of optimization objectives based on environmental changes, power status, and error accumulation, maximizing the reduction of device operating power consumption while ensuring timing accuracy meets requirements. It also provides a standardized control benchmark for subsequent closed-loop execution and adaptive compensation, making it the core decision-making link for the entire system to achieve environmentally adaptive low-power management. The dynamic power consumption strategy specifically includes:

[0083] Input encoding subunit:

[0084] Multi-source input information reception: Real-time reception of the environment and error coupling model output by the physical state evaluator, synchronous acquisition of the current remaining power of the electronic timer output by the power management unit, and the current cumulative value of timing error counted by the timing calibration unit, to complete the synchronous reception and timing alignment of the three types of core input information, ensuring the time consistency and data validity of the input information;

[0085] State factor quantization generation: The current remaining battery power is normalized to generate a battery state factor, using the formula:

[0086] ,in, For the state of charge factor, This represents the current remaining battery power of the electronic timer. The rated full charge capacity of the electronic timer battery;

[0087] An error urgency factor is generated by comparing the current accumulated error value with a threshold, using the following formula:

[0088] ,in, As the error urgency factor, This is the current accumulated error value of the electronic timer. The maximum allowable timing error threshold for an electronic timer;

[0089] Decision Feature Vector Encoding: The energy state factor, error urgency factor, and polynomial coefficients of the environment-error coupling model are standardized and encoded into a decision feature vector of uniform dimension, using the formula:

[0090] ,in, For decision feature vectors, to These are the coefficients of the polynomials of various orders in the environment-error coupling model. The order of the polynomial fitted to the environment and error coupling model is determined; after encoding, the decision feature vector is synchronously transmitted to the target weighted sub-unit and the constraint boundary sub-unit.

[0091] Target weighted sub-unit:

[0092] Dynamic matching of weighting coefficients: The decision feature vector output from the input encoding subunit is received. Power consumption optimization weighting coefficients are determined based on the mapping relationship between the power state factor and the preset power weights. Accuracy optimization weighting coefficients are determined based on the mapping relationship between the error urgency factor and the preset error weights. The power consumption optimization weighting coefficients and accuracy optimization weighting coefficients satisfy the constraint that their sum is 1, using the formula:

[0093] ;

[0094] ;

[0095] in, To optimize the weighting coefficients for power consumption, To optimize accuracy, the weighting coefficient is adjusted accordingly. When the power status factor decreases, the power consumption optimization weighting coefficient is increased accordingly, and the module prioritizes reducing operating power consumption. When the error urgency factor increases, the accuracy optimization weighting coefficient is increased accordingly, and the module prioritizes ensuring timing accuracy.

[0096] Loss function weighting scheme construction: A loss function weighting scheme for the multi-objective optimization decision network is constructed using power consumption optimization weighting coefficients and accuracy optimization weighting coefficients. The mathematical expression of the core optimization objective is determined using the following formula:

[0097] ,in, To optimize the loss function of a multi-objective decision network, The power consumption per unit time of the electronic timer. This is the absolute value of the frequency offset of the timing reference element;

[0098] The weight adaptive adjustment mechanism dynamically corrects the weight coefficients by combining the changing trends of the environmental and error coupling model. When the fluctuation rate of environmental parameters exceeds the preset threshold, the precision optimization weight coefficients are appropriately increased to avoid exceeding the time error limit caused by sudden environmental changes. When the remaining power is lower than the preset low power threshold, the power consumption optimization weight coefficients are appropriately increased to ensure the basic battery life of the device. The corrected weight coefficients still maintain the constraint that the sum is 1 to ensure the normalization of the optimization target.

[0099] Constrained boundary sub-units:

[0100] Core parameter constraint boundary calculation: Receive the decision feature vector output from the input encoding subunit, analyze the functional relationship in the environment-error coupling model, and determine the minimum duty cycle boundary of the temperature compensation circuit under the maximum allowable frequency offset limit, using the formula:

[0101] ,in, This represents the minimum duty cycle boundary for the temperature compensation circuit. To compensate for the temperature-compensated power consumption required to offset the maximum permissible frequency offset, The rated power consumption of the temperature compensation circuit when it is running at full duty cycle;

[0102] Hardware specification adaptation constraint generation: Based on the hardware specifications of the electronic timer, determine the set of available power supply voltage levels for the frequency divider counting unit and the allowable refresh rate range of the display panel; the set of available power supply voltage levels must meet the minimum operating voltage requirements of the frequency divider counting unit and the allowable refresh rate range must meet the normal display and minimum visibility requirements of the display panel, and eliminate invalid parameter ranges that exceed the rated operating range of the hardware.

[0103] Parameter search space construction: Integrating the duty cycle boundary of the temperature compensation circuit, the available power supply voltage range of the frequency divider counting unit, and the allowable refresh rate range of the display panel, a complete parameter search space constraint is generated, using the formula:

[0104] ,in, For the parameter search space, The duty cycle of the temperature compensation circuit. This refers to the power supply voltage range of the frequency divider counting unit. This represents the total number of available power supply voltage levels. For the refresh rate of the display panel, and These are the minimum and maximum refresh rates allowed by the display panel, respectively; after construction, the parameter search space constraints are transferred to the parameter decoding subunit;

[0105] Parameter decoding subunit:

[0106] Multi-objective iterative optimization: Connected to a multi-objective optimization decision network, iterative optimization is performed within the parameter search space constraints, with the goal of minimizing the loss function. The iterative process is based on gradient descent, gradually adjusting the parameters to be optimized along the direction of loss function descent until the preset iteration termination condition is reached, generating the optimal parameter combination. The core update formula for iterative optimization is:

[0107] ,in, For the first The combination of parameters to be optimized in the next iteration. For the iterative learning rate, For the loss function with respect to the th The gradient of the parameter combination in the next iteration;

[0108] Optimal parameter combination analysis: The optimal parameter combination generated by iterative optimization is decomposed into dimensions, and the optimal duty cycle of the temperature compensation circuit of the corresponding timing reference element, the optimal power supply voltage level of the frequency division counting unit, and the optimal refresh rate threshold of the display panel are analyzed respectively; the analysis process strictly matches the control protocol of the hardware module to ensure that the output parameters can be directly recognized and called by the hardware execution unit.

[0109] Joint control parameter encapsulation and output: The three types of core control parameters that have been parsed are standardized and encapsulated to generate joint control parameters in a unified format. These parameters are then output synchronously to the closed-loop execution and calibration unit and written into the strategy benchmark storage unit inside the module. This serves as the parameter update benchmark for the subsequent adaptive compensation stage and the initial reference for the next round of decision-making.

[0110] In this embodiment, the closed-loop execution and calibration unit is the core execution hub for the environmental adaptive low-power management system of the electronic timer to realize decision implementation and closed-loop calibration. Through standardized instruction parsing and hardware driving, it converts the joint control parameters generated by the upper layer into executable hardware control actions. At the same time, through real-time monitoring and deviation calculation of the entire link, it provides accurate measured feedback data for the adaptive optimization of the system.

[0111] The closed-loop execution and calibration unit is primarily responsible for receiving the joint control parameters output by the dynamic power consumption strategy generator, completing protocol conversion and instruction parsing, generating register configuration instructions for the corresponding hardware modules, and issuing them for execution according to the timing sequence. This drives the timing reference element, frequency divider counting unit, and display panel to synchronously enter the target working state. Simultaneously, the unit monitors the actual operating data of the hardware modules in real time, calculates the actual frequency offset and actual power consumption deviation, generates a measured deviation vector, and outputs it to the adaptive compensator, achieving full-link closed-loop control from decision execution to state feedback. The unit must ensure the accurate implementation of the upper-level control strategy and provide reliable operating data support for the system's adaptive optimization. It is the core link connecting the system's decision-making layer and the hardware execution layer, and a key link in ensuring the system's closed-loop optimization capability. The closed-loop execution and calibration unit specifically includes:

[0112] Instruction parsing subunit:

[0113] Joint control parameter reception and protocol conversion: Real-time reception of standardized joint control parameters output by the dynamic power consumption strategy generator, completion of parameter integrity verification and timing alignment, ensuring that the received parameters are completely consistent with the content of the upper-layer decision output; hardware protocol conversion of joint control parameters, converting standardized numerical parameters into register configuration formats that can be recognized by the corresponding hardware modules, adapting to the communication protocols and data bit width requirements of different hardware modules, and eliminating protocol barriers between the decision layer and the hardware layer;

[0114] Control parameter decomposition and classification: The joint control parameters that complete the protocol conversion are decomposed into dimensions, and the duty cycle control value of the temperature compensation circuit of the timing reference element, the power supply voltage level selection value of the frequency division counting unit, and the refresh rate threshold setting value of the display panel are analyzed respectively; the three types of control parameters after decomposition are range-checked, and abnormal parameters that exceed the rated operating range of the hardware are eliminated to ensure that the output control value meets the safe operation requirements of the hardware module.

[0115] Register configuration instruction encapsulation: The verified duty cycle control value, power supply voltage level selection value, and refresh rate threshold setting value are encapsulated into register configuration instructions for the corresponding hardware modules. Each configuration instruction includes hardware address information, register offset address, configuration data content, and checksum to ensure that the instruction can be accurately identified and executed by the corresponding hardware module. After encapsulation, the register configuration instructions are synchronously transmitted to the hardware driver subunit.

[0116] Hardware driver subunit:

[0117] Instruction timing planning and distribution: Receive register configuration instructions output by the instruction parsing subunit, plan the instruction issuance order and time interval according to the working timing requirements of each hardware module, and generate a standardized timing scheduling scheme; ensure that the configuration instructions of the temperature compensation circuit, frequency division counting unit and display panel are issued synchronously, realize the synchronous switching of the working state of the three types of hardware modules, and avoid timing errors or display abnormalities caused by asynchronous switching of different module states;

[0118] Register configuration write execution: According to the timing scheduling scheme, the duty cycle control value of the temperature compensation circuit is written to the control register corresponding to the timing reference element, the power supply voltage level selection value is written to the power management register corresponding to the frequency division counting unit, and the refresh rate threshold setting value is written to the display control register corresponding to the display panel. During the writing process, the read and write timing requirements of the hardware module are strictly followed. After each write is completed, a readback verification is performed to ensure that the configuration data in the register is completely consistent with the issued control value.

[0119] Hardware operating status synchronous triggering: After completing the configuration writing and verification of all registers, a synchronous triggering command is sent to drive the timing reference element, frequency division counting unit and display panel to synchronously enter the target operating state, ensuring that the hardware module operates according to the parameters set by the joint control parameters; at the same time, the status signal of hardware triggering completion is synchronously fed back to the synchronous monitoring subunit to start the subsequent real-time monitoring process.

[0120] Synchronous monitoring subunit:

[0121] Real-time monitoring thread startup and high-frequency sampling: Upon receiving the trigger completion signal from the hardware driver subunit, the real-time monitoring thread is immediately started to continuously collect two core operating data at a preset high-frequency sampling interval. The first channel is the real-time output frequency of the timing reference element, which is acquired with high precision through a frequency counting unit. The second channel is the real-time operating current of the display panel, which is continuously acquired through a high-precision current sampling circuit. The setting of the high-frequency sampling interval must satisfy the Nyquist sampling theorem to ensure that the collected data can completely restore the actual operating state of the hardware module.

[0122] Actual frequency offset calculation: Based on the acquired real-time output frequency and the nominal frequency of the timing reference element, the actual frequency offset is calculated using the formula:

[0123] ,in, This represents the actual frequency offset of the timing reference element. The real-time output frequency of the timing reference element. The nominal frequency of the timing reference element;

[0124] Actual power consumption offset calculation: Based on the collected real-time operating current and the power supply voltage of the display panel, the actual operating power consumption of the display panel is calculated. Combined with the expected power consumption value set by the dynamic power consumption strategy generator, the actual power consumption offset is calculated using the formula:

[0125] ;

[0126] ;

[0127] in, This refers to the actual operating power consumption of the display panel. The rated power supply voltage for the display panel, This is the real-time operating current of the display panel. This represents the actual power consumption offset of the display panel. The expected power consumption value for the display panel set for the dynamic power consumption strategy generator;

[0128] Monitoring data preprocessing and caching: The calculated actual frequency offset and actual power consumption deviation are subjected to moving average filtering to remove noise interference during the sampling process and improve the stability and accuracy of the data; the preprocessed monitoring data is stored in the local cache unit and simultaneously transmitted to the deviation calculation subunit to provide a data basis for subsequent deviation calculation;

[0129] Deviation calculation sub-unit:

[0130] Frequency offset deviation calculation: The actual frequency offset output from the receiving synchronization monitoring subunit is obtained, and the expected frequency offset output from the physical state evaluator is acquired synchronously. The difference between the two sets of data is calculated to generate the frequency offset deviation, using the formula:

[0131] ,in, This is the frequency offset deviation. The expected frequency offset output by the physical state evaluator;

[0132] Power consumption execution deviation calculation: Receive the actual power consumption deviation output from the synchronization monitoring subunit, extract the actual operating power consumption and the expected power consumption value set by the dynamic power consumption strategy generator, perform a difference calculation to generate the power consumption execution deviation, using the formula:

[0133] ,in, Performance deviation due to power consumption;

[0134] Measured Deviation Vector Encapsulation and Output: The calculated frequency offset deviation and power consumption execution deviation are standardized and combined into a two-dimensional measured deviation vector, using the formula:

[0135] ,in, The measured deviation vector is encapsulated and then stably output to the adaptive compensator, providing accurate feedback data support for the system's adaptive parameter optimization.

[0136] Furthermore, the synchronous timing-driven technology, based on the hardware synchronous clock theory, solves the synchronization problem of state switching of multiple hardware modules. By using a unified system clock, it provides a time reference for instruction issuance, register writing, and state triggering, and plans the issuance sequence and triggering timing of configuration instructions for each hardware module, ensuring that the temperature compensation circuit, frequency division counting unit, and display panel complete the switching of working states at the same time. This technology can avoid problems such as timing pulse loss, frequency division counting errors, or abnormal display caused by asynchronous state switching of different modules, ensuring the accurate implementation of upper-level control strategies and eliminating additional timing errors and power consumption caused by timing deviations.

[0137] High-precision synchronous sampling technology, based on synchronous data acquisition theory, achieves high-precision synchronous acquisition of frequency and current signals. It simultaneously initiates frequency counting and current sampling processes using the same sampling trigger signal, ensuring complete consistency of timestamps for both data streams and eliminating time deviations between different sampling channels. A high-frequency sampling clock enhances the resolution of frequency counting, while a high-precision differential sampling circuit reduces noise interference in current acquisition, ensuring that the acquired actual operating data accurately reflects the working status of the hardware modules. This technology provides high-fidelity raw data for subsequent deviation calculations, avoiding feedback deviations caused by sampling errors and guaranteeing the accuracy of system closed-loop calibration.

[0138] The closed-loop deviation quantization technology, based on error propagation theory, standardizes and quantifies the difference between the actual operating state of the hardware module and the expected target. By calculating the frequency offset deviation and power consumption execution deviation separately, it decouples the two core deviations in the hardware execution process, corresponding to the execution effects of the timing accuracy target and the power consumption optimization target, respectively. Through vector encapsulation, the two types of deviations are integrated into a unified measured deviation vector, realizing a standardized expression of the multi-objective execution effect, which can be directly identified and processed by the upper-level multi-objective optimization decision network. This technology builds a feedback bridge between the hardware execution layer and the decision optimization layer, and is the core technical foundation for the system to achieve closed-loop adaptive optimization.

[0139] In this embodiment, the adaptive compensator is the core feedback adjustment unit used in the environmental adaptive low power management system of the electronic timer to realize closed-loop self-optimization and continuous performance iteration. It completes the dynamic correction and iterative update of the control strategy through real-time analysis and reverse optimization of hardware execution deviation, so that the system can adapt to the performance fluctuations caused by environmental changes, device characteristic drift and hardware aging.

[0140] The adaptive compensator is primarily responsible for receiving the measured deviation vector output by the closed-loop execution and calibration unit, performing analysis and standardization mapping of the deviation components, and generating a feedback driving vector that can be recognized by the multi-objective optimization decision network. The unit determines the degree of deviation between the actual system operating state and the expected optimization objective based on the feedback data. When the deviation exceeds a preset threshold, it triggers the multi-objective optimization decision network to perform online parameter optimization within the neighborhood of the current joint control parameters. During the optimization process, it strictly adheres to timing accuracy and endurance constraints, retaining only the optimal candidate parameter combinations that meet the requirements. After the optimization effect is verified, the effective parameters are updated to the policy benchmark of the dynamic power consumption strategy generator for the next round of hardware control command issuance. The unit constructs a complete closed-loop link from decision execution to feedback optimization, which is the core guarantee for achieving system environmental adaptability and long-term low-power stable operation. The adaptive compensator specifically includes:

[0141] Bias analysis and eigenmap subunit:

[0142] Actual deviation reception and verification: The measured deviation vector output by the closed-loop execution and calibration unit is received in real time to complete the data integrity and validity verification, and invalid data with sampling abnormalities or timing misalignments are eliminated to ensure that the input data can truly reflect the actual operating status of the hardware module; the measured deviation vector that has passed the verification is decomposed in dimensions to extract the frequency offset deviation component and the power consumption execution deviation component, and the two types of core deviations are decoupled and separated.

[0143] Standardization of Deviation Components: Dimensionless standardization is performed on the two types of deviation components after decomposition to eliminate dimensional differences between different physical quantities, ensuring that the two types of deviations can be comprehensively evaluated in the same dimension. The standardization process uses the following formula:

[0144] ,in, This refers to the standardized frequency offset deviation component. This is the frequency offset deviation. This is the maximum frequency offset allowed by the electronic timer;

[0145] The standardization of power consumption execution deviation is performed using the following formula:

[0146] ,in, The standardized power consumption is used to perform the deviation component. For power consumption execution deviation, The rated power consumption per unit time of the electronic timer;

[0147] Feedback-driven vector generation and output: The standardized frequency offset and power consumption execution offset components are mapped to the input feature space of the multi-objective optimization decision network to generate a standardized feedback-driven vector, using the formula:

[0148] ,in, For feedback driving vector;

[0149] After generation, the feedback driving vector is synchronously transmitted to the deviation judgment and optimization triggering subunit, and the original deviation data is stored in the local storage unit for subsequent statistical analysis and parameter adaptive adjustment.

[0150] Deviation determination and optimization triggering sub-unit:

[0151] Comprehensive Deviation Measurement Calculation: Receive the feedback driving vector output from the deviation analysis and feature mapping subunits, simultaneously acquire the currently effective precision optimization weight coefficients and power optimization weight coefficients from the dynamic power consumption strategy generator, and calculate the comprehensive deviation of the current operating state from the expected optimization target using the formula:

[0152] ,in, The overall deviation from the current operating status. To optimize the weighting coefficients for higher accuracy, Optimize weighting coefficients for power consumption;

[0153] Adaptive adjustment threshold dynamic update: The preset adaptive adjustment threshold is dynamically corrected based on the current environmental fluctuation rate, remaining power, and error urgency. When the environmental fluctuation rate increases, the threshold is appropriately lowered to improve the system response sensitivity. When the remaining power is lower than the low power threshold, the threshold is appropriately raised to reduce unnecessary optimization calculations. When the error urgency increases, the threshold is appropriately lowered to enhance the timing accuracy guarantee capability.

[0154] Optimization Trigger Judgment and Initial Condition Configuration: The calculated comprehensive deviation is compared with the dynamically updated adaptive adjustment threshold. When the comprehensive deviation is lower than the threshold, only the current deviation data is recorded, the optimization process is not triggered, and new measured deviation vectors are continuously monitored. When the comprehensive deviation is higher than the threshold, the online parameter optimization process is immediately triggered, the currently effective joint control parameters are obtained as the initial optimization starting point, and the current environment and error coupling model and remaining power data are retrieved simultaneously as constraints for the optimization process.

[0155] Constrained neighborhood parameter optimization sub-unit

[0156] Local Neighborhood Search Space Construction: Centered on the current joint control parameters and combined with the preset neighborhood search step size, a local neighborhood search space is constructed. Simultaneously, hardware specification constraints, accuracy constraints, and battery life constraints are integrated to define the boundaries of the search space and eliminate invalid parameter ranges that exceed hardware capabilities and performance requirements. The local neighborhood search space is constructed using the following formula:

[0157] ,in, For the local neighborhood search space, This is the current duty cycle of the temperature compensation circuit. The neighborhood search step size is the duty cycle. This represents the minimum duty cycle boundary for the temperature compensation circuit. This refers to the current power supply voltage level of the frequency divider counting unit. This represents the neighborhood search range for the voltage level. This is a set of available power supply voltage ranges. This represents the current refresh rate threshold for the display panel. The neighborhood search step size is the refresh rate. and These are the minimum and maximum refresh rates allowed by the display panel, respectively.

[0158] Joint Iterative Optimization: Within the local neighborhood search space, aiming to minimize the weighted sum of timing error and power consumption per unit time, a multi-parameter joint iterative optimization process is initiated. The gradient descent method is used to continuously update the candidate parameter combinations along the descent direction of the loss function. The iterative update uses the following formula:

[0159] ,in, For the first Candidate parameter combinations for the next iteration. For iterative learning rate in local optimization, For the loss function with respect to the th The gradient of the candidate parameter combination in the next iteration;

[0160] Real-time constraint verification: During the iterative optimization process, a dual-constraint real-time verification is performed on each generated candidate parameter combination. The first type is accuracy constraint, which uses the environment-error coupling model as a hard constraint to verify whether the expected frequency offset under the candidate parameter combination meets the maximum allowable frequency offset limit. The second type is endurance constraint, which uses the current remaining battery power as a boundary condition to verify whether the estimated continuous working time under the candidate parameter combination meets the minimum threshold requirement. The estimated continuous working time is calculated using the following formula:

[0161] ,in, To estimate sustainable working hours, This is the current remaining battery level. This represents the estimated power consumption per unit time under candidate parameter combinations.

[0162] Only candidate parameter combinations that simultaneously satisfy both types of constraints are retained, and all invalid parameters that do not meet the requirements are eliminated to ensure that the optimization results can simultaneously meet the core requirements of accuracy and battery life.

[0163] Optimal candidate parameter output: When the iteration process reaches the preset iteration termination condition, the optimization calculation stops, and the group with the smallest loss function value is selected from all valid candidate parameter combinations as the optimal candidate parameter combination for this optimization, and output to the optimization effect verification and parameter update subunit.

[0164] Optimization effect verification and parameter update subunit:

[0165] Optimization magnitude quantization calculation: The optimal candidate parameter combination output by the constrained neighborhood parameter optimization subunit is received. The loss function values ​​corresponding to the candidate parameter combination and the current joint control parameters are calculated respectively to quantify the optimization magnitude of the candidate parameter combination using the formula:

[0166] ,in, The optimization magnitude of the candidate parameter combination, This represents the loss function value corresponding to the current joint control parameters. The loss function value corresponding to the candidate parameter combination;

[0167] Update threshold determination and instruction encapsulation: The calculated optimization magnitude is compared with the preset parameter update threshold; when the optimization magnitude is less than or equal to the update threshold, it indicates that the optimization effect of the candidate parameter combination is not significant. The currently effective joint control parameters are retained, the strategy update is not executed, the optimization data for this time is recorded, and the deviation monitoring process is returned; when the optimization magnitude is greater than the update threshold, it indicates that the candidate parameter combination can significantly improve system performance. The candidate parameter combination is encapsulated into a standardized parameter update instruction and sent to the dynamic power consumption strategy generator.

[0168] Strategy benchmark update execution: After receiving the parameter update command, the dynamic power consumption strategy generator parses the updated duty cycle of the temperature compensation circuit, the power supply voltage level of the frequency division counting unit, and the refresh rate threshold of the display panel. It writes the updated joint control parameters into the strategy benchmark storage unit to replace the original joint control parameters for the control command issuance of the next round of closed-loop execution and calibration unit. At the same time, the updated parameters are synchronously fed back to the adaptive compensator as the initial benchmark for the subsequent optimization process.

[0169] Furthermore, the deviation-driven negative feedback closed-loop control technology is based on the negative feedback control mechanism in classical automatic control theory. It uses the measured deviation of the hardware execution layer as feedback input to correct the control parameters of the decision layer, forming a complete negative feedback closed-loop control system. By standardizing and mapping the measured deviation into a feedback driving vector, it directly acts on the optimization process of the multi-objective optimization decision network, allowing the system to continuously correct the control strategy according to the actual operating effect. This solves the defects of static open-loop strategies that cannot adapt to dynamic environmental changes, device characteristic drift, hardware aging, etc., and ensures the long-term stability and optimization effect of the system.

[0170] The constrained local neighborhood optimization technique is based on the local search theory in numerical optimization. It performs small-scale directional optimization within the neighborhood of the current optimal parameter. Compared with global optimization, this technique significantly reduces the parameter search space, reduces the amount of iterative computation and operation time, and lowers the power consumption of the optimization process. It is perfectly suited to the limited computing resources of the low-power embedded microprocessor on the electronic timer. At the same time, the accuracy constraints of the environment and error coupling model and the battery life constraints of the remaining power are embedded as hard conditions into the entire optimization process to ensure that all candidate parameters meet the basic performance requirements of the system and avoid problems such as exceeding accuracy limits or insufficient battery life during the optimization process. It balances optimization efficiency, optimization effect and system operation safety.

[0171] The adaptive threshold event-triggered technology is based on event-driven intelligent control theory. It measures the execution effect of the current control strategy by comprehensively measuring deviation. The parameter optimization process is only triggered when the deviation between the actual operating state and the expected target exceeds the adaptive adjustment threshold. This technology avoids unnecessary computational power consumption caused by periodic continuous optimization and significantly reduces the system's own operating losses. At the same time, the adaptive adjustment threshold can be dynamically adjusted according to environmental volatility, remaining power, and error urgency. It achieves the optimal balance between response sensitivity and system power consumption under different operating conditions, ensuring the system's rapid response capability to abnormal states while meeting the core design goal of low-power operation.

[0172] An environmentally adaptive low-power management method for electronic timers is proposed. Based on multi-dimensional dynamic environmental perception, the method centers on modeling the coupling relationship between environment and error, supported by multi-objective intelligent optimization decision-making, and guaranteed by closed-loop execution and adaptive feedback updates. It constructs a complete closed-loop management system from environmental perception to strategy optimization and execution calibration. The method dynamically adjusts the device's operating parameters according to environmental changes, device power status, and accumulated timing errors. While ensuring timing accuracy meets design requirements, it maximizes the reduction of the electronic timer's power consumption and extends the device's continuous battery life. The method is executed by the system and includes the following steps:

[0173] Step S1: Obtain coarse-grained change values ​​of environmental parameters using a preset basic sleep-wake cycle; during non-working periods, the device enters deep sleep mode, retaining only the minimum power supply for operation. After waking up, it quickly completes coarse sampling of three types of environmental parameters: temperature, vibration, and illuminance, to obtain basic data that can reflect the overall trend of environmental changes; after completing a single coarse-grained sampling and data transmission, it immediately switches back to deep sleep mode to maximize the proportion of sleep time and reduce the basic power consumption of the environmental monitoring process;

[0174] It receives coarse-grained change values ​​from coarse-grained environmental monitoring output, calculates the environmental state fluctuation rate at the current moment, and quantifies the degree of drastic change in environmental parameters;

[0175] A frequency adjustment coefficient is generated based on the environmental state fluctuation rate. The preset base sampling frequency is then corrected using the frequency adjustment coefficient to obtain a dynamically adjustable sampling frequency.

[0176] When environmental fluctuations are severe, the sampling frequency is increased to ensure data integrity; when the environment is stable, the sampling frequency is reduced to control operating power consumption, thus achieving dynamic adaptation between sampling accuracy and power consumption. Finally, the determined dynamically adjustable sampling frequency is used as the basis for subsequent high-precision acquisition.

[0177] Based on a dynamically adjustable sampling frequency and a unified synchronous clock signal, the temperature sensing component, vibration sensing component, and photosensor are driven to synchronously acquire data, obtaining ambient temperature data, raw vibration signals, and ambient illuminance data. This ensures the time synchronization of the three types of parameters and eliminates time deviations from multi-source data. The raw vibration signal undergoes time-frequency domain transformation processing, converting the time-domain vibration signal into a frequency-domain vibration spectrum through a fast Fourier transform, completing the statistical processing of the vibration signal's frequency band energy distribution. The ambient temperature data, vibration spectrum, and ambient illuminance data are then processed in a unified format, encapsulated into standardized multi-dimensional environmental parameters, and output to subsequent model building steps.

[0178] Step S2: Receive multi-dimensional environmental parameters, extract and fuse three types of core environmental features respectively; perform time-domain sliding variance calculation on the environmental temperature data to generate temperature fluctuation feature values, and quantify the long-term trend and intensity of environmental temperature fluctuations.

[0179] Perform frequency band energy distribution statistics on the vibration spectrum to generate mechanical interference characteristic values ​​and quantify the mechanical vibration intensity that affects the timing reference element;

[0180] A step change identification is performed on the ambient illuminance data to generate photothermal effect feature values ​​and quantify the photothermal effect caused by sudden changes in ambient illuminance.

[0181] The temperature fluctuation characteristic value, mechanical disturbance characteristic value and photothermal influence characteristic value are standardized and fused to generate an environmental stress characteristic vector, which fully characterizes the comprehensive stress of the current environment on the timing reference element;

[0182] The system receives environmental stress feature vectors, performs similarity matching with feature samples in a pre-trained device physical property mapping library, retrieves the sample set that is closest to the current environmental stress features, and extracts the resonant frequency change trend data of the timing reference element under the corresponding sample.

[0183] Based on the expected frequency offset, the temperature compensation power required to completely offset the frequency offset is calculated, and the quantitative correspondence between precise control requirements and power consumption is clarified.

[0184] Using the expected frequency offset as the independent variable and the corresponding required temperature compensation power consumption as the dependent variable, a polynomial curve fitting is performed to generate a continuous and differentiable environment and error coupling model, which fully characterizes the functional relationship between frequency offset and temperature compensation power consumption under the current environment.

[0185] The completed environment and error coupling model is output to the subsequent joint control parameter generation step;

[0186] Step S3: Receive the environment and error coupling model, the current remaining power and the current accumulated timing error value, normalize the current remaining power to generate a power status factor, and compare the current accumulated timing error value with a threshold to generate an error urgency factor;

[0187] The power status factor, error urgency factor, and polynomial coefficients of the environment and error coupling model are standardized and encoded into a decision feature vector of uniform dimension, providing standardized input for subsequent optimization decisions.

[0188] The power consumption optimization weight coefficient is determined based on the mapping relationship between the power status factor and the preset power weight, and the accuracy optimization weight coefficient is determined based on the mapping relationship between the error urgency factor and the preset error weight.

[0189] When the power state factor decreases, the power consumption optimization weight coefficient increases synchronously, and the method prioritizes reducing operating power consumption; when the error urgency factor increases, the accuracy optimization weight coefficient increases synchronously, and the method prioritizes ensuring timing accuracy; using the power consumption optimization weight coefficient and the accuracy optimization weight coefficient, a loss function of a multi-objective optimization decision network is constructed to determine the core optimization objective;

[0190] The functional relationship in the environmental and error coupling model is analyzed to determine the minimum duty cycle boundary of the temperature compensation circuit under the maximum allowable frequency offset limit.

[0191] Based on the hardware specifications of the electronic timer, the set of available power supply voltage levels for the frequency division counting unit and the allowable refresh rate range of the display panel are determined; the duty cycle boundary of the temperature compensation circuit, the set of available power supply voltage levels for the frequency division counting unit, and the allowable refresh rate range of the display panel are integrated to generate a complete parameter search space constraint.

[0192] Within the parameter search space constraints, iterative optimization is performed with the goal of minimizing the loss function. The iterative process is based on the gradient descent method, and the parameters to be optimized are gradually adjusted along the direction of loss function descent until the preset iteration termination condition is reached, generating the optimal parameter combination.

[0193] The optimal parameter combination generated by iterative optimization is decomposed into dimensions, and the optimal duty cycle of the temperature compensation circuit of the corresponding timing reference element, the optimal power supply voltage level of the frequency division counting unit, and the optimal refresh rate threshold of the display panel are respectively analyzed. The three types of core control parameters are standardized and encapsulated to generate joint control parameters in a unified format, which are synchronously output to the subsequent closed-loop execution steps and written into the strategy reference storage unit as a reference for subsequent adaptive updates.

[0194] Step S4: Receive the joint control parameters, perform hardware protocol conversion on the joint control parameters, convert the standardized numerical parameters into a register configuration format recognizable by the corresponding hardware module, and parse out the duty cycle control value of the temperature compensation circuit, the power supply voltage level selection value of the frequency divider counting unit, and the refresh rate threshold setting value of the display panel; encapsulate the parsed control values ​​into register configuration instructions for the corresponding hardware module, and synchronously write the configuration instructions into the control register of the corresponding hardware module according to the preset timing requirements; after writing, perform readback verification to ensure the accuracy of the configuration data; after verification, send a synchronization trigger command to drive the timing reference element, frequency divider counting unit, and display panel to synchronously enter the target working state;

[0195] After the hardware status is triggered, the real-time monitoring thread is started to synchronously collect the real-time output frequency of the timing reference element and the real-time operating current of the display panel at a preset high-frequency sampling interval; the actual frequency offset is calculated based on the collected real-time output frequency and the nominal frequency of the timing reference element.

[0196] Based on the collected real-time operating current and the rated power supply voltage of the display panel, the actual operating power consumption of the display panel is calculated, and combined with the preset expected power consumption value, the actual power consumption deviation is calculated.

[0197] The difference between the actual frequency offset and the expected frequency offset is calculated to generate the frequency offset deviation;

[0198] The difference between the actual operating power consumption and the expected power consumption value is calculated to generate the power consumption execution deviation.

[0199] The frequency offset deviation and power consumption execution deviation are standardized and combined into a two-dimensional measured deviation vector.

[0200] The generated measured deviation vector is output to the subsequent adaptive feedback and policy update steps;

[0201] Step S5: Receive the measured deviation vector, decompose the measured deviation vector into dimensions, extract the frequency offset deviation component and the power consumption deviation component, perform dimensionless standardization on the two types of deviation components to eliminate dimensional differences;

[0202] The standardized bias components are mapped to the input feature space of the multi-objective optimization decision network to generate standardized feedback driving vectors.

[0203] Receive feedback driving vector, synchronously obtain the currently effective precision optimization weight coefficient and power consumption optimization weight coefficient, and calculate the comprehensive deviation of the current running state from the expected optimization target;

[0204] The preset adaptive adjustment threshold is dynamically corrected by combining the current environmental fluctuation rate, remaining power, and error urgency. The overall deviation is compared with the dynamically updated adaptive adjustment threshold. When the overall deviation is lower than the threshold, only the current deviation data is recorded, the optimization process is not triggered, and the process returns to the initial environmental data acquisition step to start the next management cycle. When the overall deviation is higher than the threshold, the online parameter optimization process is immediately triggered to obtain the currently effective joint control parameters as the initial optimization starting point.

[0205] Centered on the current joint control parameters, and combined with the preset neighborhood search step size, a local neighborhood search space is constructed. At the same time, hardware specification constraints, accuracy constraints and battery life constraints are integrated to complete the boundary limitation of the search space.

[0206] Within the local neighborhood search space, a multi-parameter joint iterative optimization process is initiated with the goal of minimizing the loss function. During the iteration process, the accuracy constraint and endurance constraint are verified in real time for each candidate parameter combination, and only candidate parameter combinations that satisfy both types of constraints are retained. When the iteration process reaches the preset termination condition, the optimization calculation is stopped, and the optimal candidate parameter combination with the minimum loss function value is selected.

[0207] Calculate the optimization magnitude of the optimal candidate parameter combination relative to the current joint control parameters;

[0208] The optimization magnitude is compared with the preset parameter update threshold. When the optimization magnitude is less than or equal to the update threshold, the currently effective joint control parameters are retained, the strategy update is not performed, the initial environment acquisition steps are returned, and the next management cycle is started. When the optimization magnitude is greater than the update threshold, the candidate parameter combination is encapsulated into a standardized parameter update instruction and sent to the strategy baseline storage unit.

[0209] After receiving the parameter update instruction, the updated core control parameters are parsed out, and the updated joint control parameters are written into the strategy benchmark storage unit to replace the original joint control parameters for the issuance of hardware control instructions in the next management cycle. After completing the strategy benchmark update, the system returns to the initial environment acquisition step and starts the next complete management cycle, realizing the continuous closed-loop iteration and adaptive optimization of the method.

[0210] The five core steps of this method are sequentially linked to form a complete low-power management cycle. Simultaneously, the feedback update mechanism in the fifth step forms a closed-loop link with the initial environmental acquisition step, enabling continuous iterative optimization of the management strategy. The method's operation is triggered in three modes: the first is periodic timed triggering, which initiates the complete management process according to a preset management cycle to achieve regular strategy updates; the second is environmental mutation triggering, which immediately initiates the complete management process to quickly respond to environmental changes when coarse-grained environmental monitoring detects a significant change in environmental parameters; and the third is deviation exceeding a threshold triggering, which immediately initiates online optimization and strategy update processes to quickly correct execution deviations when the overall deviation of the measured deviation vector exceeds the adaptive adjustment threshold. These three triggering modes work together to ensure both low power consumption and rapid response to changes in operating conditions. During operation, each closed-loop iteration optimizes the control strategy, allowing the management strategy to continuously adapt to the current environmental state, hardware characteristics, and operating conditions, achieving long-term stable low-power, high-precision operation.

[0211] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An environment-adaptive low-power management system for electronic timers, characterized in that: include: An environmental sensing array is used to acquire multi-dimensional environmental parameters at a dynamically adjustable sampling frequency. These multi-dimensional environmental parameters include at least ambient temperature, vibration spectrum, and ambient illuminance. The physical state evaluator, connected to the environmental sensing array, is configured to construct an environment-error coupling model for the current moment based on multi-dimensional environmental parameters. The environment-error coupling model is used to characterize the functional relationship between the expected frequency offset of the timing reference element of the electronic timer under the current environmental parameters and the temperature compensation power consumption. A dynamic power consumption strategy generator, connected to a physical state evaluator, is configured to input the environment-error coupling model, the current remaining power of the electronic timer, and the current accumulated timing error into a multi-objective optimization decision network, and output a set of joint control parameters. The joint control parameters include at least the duty cycle of the temperature compensation circuit of the timing reference element, the power supply voltage level of the frequency division counting unit, and the refresh rate threshold of the display panel. The closed-loop execution and calibration unit is connected to the functional modules of the dynamic power consumption strategy generator and the electronic timer, respectively. It is configured to parse the joint control parameters into hardware control commands and send them out for execution, while monitoring the actual frequency offset of the timing reference element and the actual power consumption of the display panel in real time to generate a measured deviation vector. An adaptive compensator, connected to the closed-loop execution and calibration unit and the dynamic power consumption strategy generator respectively, is configured to feed back the measured deviation vector to the multi-objective optimization decision network, triggering the multi-objective optimization decision network to perform online parameter optimization in the neighborhood of the joint control parameters with the goal of minimizing the weighted sum of timing error and power consumption per unit time, and update the optimization result to the strategy benchmark of the dynamic power consumption strategy generator.

2. The environmentally adaptive low-power management system for electronic timers according to claim 1, characterized in that: The environmental sensing array specifically includes a low-power monitoring unit, a frequency scheduling unit, and a high-precision acquisition unit. The low-power monitoring unit is configured to acquire coarse-grained change values ​​of environmental parameters with a preset basic sleep-wake cycle, and transmit the coarse-grained change values ​​to the frequency scheduling unit. The frequency scheduling unit is configured to receive coarse-grained change values, calculate the environmental state fluctuation rate at the current moment, generate a frequency adjustment coefficient based on the environmental state fluctuation rate, and use the frequency adjustment coefficient to correct the preset basic sampling frequency to obtain a dynamically adjustable sampling frequency. The high-precision acquisition unit is configured to drive the temperature sensing component, vibration sensing component and photosensitive sensing component to perform synchronous data acquisition according to the dynamically adjustable sampling frequency, so as to obtain ambient temperature data, raw vibration signal and ambient illuminance data. The high-precision acquisition unit is also configured to perform time-frequency domain transformation processing on the original vibration signal to generate a vibration spectrum, and encapsulate the ambient temperature data, vibration spectrum and ambient illuminance data into multi-dimensional environmental parameters. The environmental sensing array outputs the multidimensional environmental parameters to the physical state evaluator.

3. The environmentally adaptive low-power management system for electronic timers according to claim 1, characterized in that: The physical state evaluator specifically includes a feature extraction subunit, a frequency drift prediction subunit, and a coupling relationship construction subunit; The feature extraction subunit is configured to receive multidimensional environmental parameters, perform time-domain sliding variance calculation on environmental temperature data to generate temperature fluctuation feature values, perform frequency band energy distribution statistics on vibration spectrum to generate mechanical interference feature values, and perform step change identification on environmental illuminance data to generate photothermal influence feature values, and fuse temperature fluctuation feature values, mechanical interference feature values ​​and photothermal influence feature values ​​into an environmental stress feature vector. The frequency drift prediction subunit is connected to the feature extraction subunit and is configured to receive the environmental stress feature vector, input it into the pre-trained device physical characteristic mapping library, match the resonant frequency change trend of the timing reference element under the current environmental stress feature vector, and output the expected frequency offset. The coupling relationship construction subunit is connected to the frequency drift prediction subunit. It is configured to calculate the heating or digital correction energy consumption required to offset the expected frequency offset based on the expected frequency offset, obtain the temperature compensation power consumption, and use the expected frequency offset as the independent variable and the temperature compensation power consumption as the dependent variable to perform curve fitting to generate an environment-error coupling model to characterize the functional relationship between the two.

4. The environmentally adaptive low-power management system for electronic timers according to claim 1, characterized in that: The dynamic power consumption strategy generator specifically includes an input encoding subunit, a target weighting subunit, a constraint boundary subunit, and a parameter decoding subunit; The input encoding subunit is configured to receive the environment-error coupling model, the current remaining power and the current accumulated timing error value, normalize the current remaining power to generate a power status factor, perform threshold comparison on the current accumulated timing error value to generate an error urgency factor, encode the power status factor, the error urgency factor and the model parameters of the environment-error coupling model into a decision feature vector, and transmit the decision feature vector to the target weighting subunit and the constraint boundary subunit. The target weighting subunit is configured to receive decision feature vectors, determine power consumption optimization weight coefficients based on the power status factor and the preset power-weight mapping relationship, determine accuracy optimization weight coefficients based on the error urgency factor and the preset error-weight mapping relationship, construct a loss function weight allocation scheme for a multi-objective optimization decision network using the power consumption optimization weight coefficients and the accuracy optimization weight coefficients, and transmit the loss function weight allocation scheme to the parameter decoding subunit. The constraint boundary subunit is configured to receive the decision feature vector, parse the functional relationship in the environment-error coupling model, determine the minimum duty cycle boundary of the temperature compensation circuit under the maximum allowable frequency offset limit, and determine the available power supply voltage range of the frequency division counting unit and the allowable refresh rate range of the display panel in combination with the hardware specifications of the electronic timer, generate parameter search space constraints, and transmit the parameter search space constraints to the parameter decoding subunit. The parameter decoding subunit is connected to a multi-objective optimization decision network and is configured to perform iterative optimization within the parameter search space constraints, with the goal of minimizing the weighted sum defined by the loss function weight allocation scheme, to generate the optimal parameter combination. The optimal parameter combination is then parsed into the duty cycle of the temperature compensation circuit of the timing reference element, the power supply voltage level of the frequency division counting unit, and the refresh rate threshold of the display panel, and encapsulated as joint control parameters and output.

5. The environmentally adaptive low-power management system for electronic timers according to claim 1, characterized in that: The closed-loop execution and calibration unit specifically includes an instruction parsing subunit, a hardware driver subunit, a synchronization monitoring subunit, and a deviation calculation subunit; The instruction parsing subunit is configured to receive the joint control parameters output by the dynamic power consumption strategy generator, perform protocol conversion on the joint control parameters, and parse out the duty cycle control value of the temperature compensation circuit of the timing reference element, the power supply voltage level selection value of the frequency division counting unit, and the refresh rate threshold setting value of the display panel, respectively. The duty cycle control value, the power supply voltage level selection value, and the refresh rate threshold setting value are then encapsulated into register configuration instructions for the corresponding hardware modules. The hardware driver subunit is connected to the instruction parsing subunit and is configured to receive register configuration instructions. According to the preset timing requirements, the duty cycle control value of the temperature compensation circuit is written into the control register corresponding to the timing reference element, the power supply voltage level selection value is written into the power management register corresponding to the frequency divider counting unit, and the refresh rate threshold setting value is written into the display control register corresponding to the display panel. This drives the timing reference element, the frequency divider counting unit, and the display panel to enter the target working state synchronously. The synchronous monitoring subunit is connected to the hardware driver subunit and is configured to start a real-time monitoring thread after the hardware driver subunit performs a write operation. The thread continuously collects the real-time output frequency of the timing reference element and the real-time operating current of the display panel at a preset high-frequency sampling interval. The actual frequency offset is calculated based on the real-time output frequency and the nominal frequency of the timing reference element, and the actual power consumption deviation is calculated based on the real-time operating current and the theoretical power consumption value of the display panel. The deviation calculation subunit is connected to the synchronization monitoring subunit and the adaptive compensator, respectively. It is configured to receive the actual frequency offset and the actual power consumption deviation, perform difference calculation between the actual frequency offset and the expected frequency offset output by the physical state evaluator to generate a frequency offset deviation, perform difference calculation between the actual power consumption deviation and the expected power consumption value set in the dynamic power consumption strategy generator to generate a power consumption execution deviation, combine the frequency offset deviation and the power consumption execution deviation into a measured deviation vector, and transmit the measured deviation vector to the adaptive compensator.

6. The environmentally adaptive low-power management system for electronic timers according to claim 1, characterized in that: The adaptive compensator feeds back the measured deviation vector to the multi-objective optimization decision network, triggering the network to perform online parameter optimization within the neighborhood of the jointly controlled parameters, with the objective of minimizing the weighted sum of timing error and power consumption per unit time. The optimization results are then updated in the policy benchmark of the dynamic power consumption policy generator. Specifically, this includes: The adaptive compensator receives the measured deviation vector output by the closed-loop execution and calibration unit, analyzes the measured deviation vector, extracts the frequency offset deviation component and the power consumption execution deviation component, and maps the frequency offset deviation component and the power consumption execution deviation component to the input feature space of the multi-objective optimization decision network to form a feedback driving vector. The multi-objective optimization decision network, based on the feedback driving vector, first determines the degree of deviation between the actual operating state of the electronic timer under the current joint control parameters and the expected optimization objective. When the degree of deviation exceeds the preset adaptive adjustment threshold, the current joint control parameters are used as the initial optimization starting point. Within the preset neighborhood centered on the initial optimization starting point, the network performs joint iterative optimization on the duty cycle of the temperature compensation circuit, the power supply voltage level of the frequency division counting unit, and the refresh rate threshold of the display panel, with the goal of minimizing the weighted sum of timing error and power consumption per unit time, to generate candidate parameter combinations. In the joint iterative optimization process, the multi-objective optimization decision network uses the environment-error coupling model output by the physical state evaluator as a constraint condition to verify in real time whether the expected frequency offset under the candidate parameter combination meets the maximum allowable frequency offset limit of the timing reference element. It also uses the current remaining power as a boundary condition to verify whether the estimated total power consumption under the candidate parameter combination exceeds the sustainable working time threshold supported by the current remaining power. Only candidate parameter combinations that simultaneously meet the constraints of the maximum allowable frequency offset limit and the sustainable working time threshold are retained. The adaptive compensator compares the candidate parameter combination finally output by the multi-objective optimization decision network with the current joint control parameters. When the optimization magnitude of the candidate parameter combination relative to the current joint control parameters in the weighted sum of timing error and power consumption per unit time is greater than the preset update threshold, the candidate parameter combination is encapsulated into a parameter update instruction and sent to the dynamic power consumption strategy generator. The dynamic power consumption strategy generator receives parameter update instructions, parses the updated duty cycle of the temperature compensation circuit, the power supply voltage level of the frequency division counting unit, and the refresh rate threshold of the display panel, and writes the updated joint control parameters into the strategy reference of the dynamic power consumption strategy generator to replace the original joint control parameters for the control instructions to be issued by the next round of closed-loop execution and calibration unit.

7. An environment-adaptive low-power management method for an electronic timer, said method being executed by the system according to any one of claims 1-6, characterized in that: The method includes the following steps: Step S1: Collect multi-dimensional environmental parameters at a dynamically adjustable sampling frequency using an environmental sensing array. The multi-dimensional environmental parameters include at least ambient temperature, vibration spectrum, and ambient illuminance. Step S2: Receive the multi-dimensional environmental parameters generated in step S1 through the physical state evaluator, and construct the environment-error coupling model at the current moment based on the multi-dimensional environmental parameters. The environment-error coupling model is used to characterize the functional relationship between the expected frequency offset of the timing reference element of the electronic timer under the current environmental parameters and the temperature compensation power consumption. Step S3: Receive the environment-error coupling model generated in step S2 through the dynamic power consumption strategy generator, and input the environment-error coupling model together with the current remaining power of the electronic timer and the current accumulated timing error into the multi-objective optimization decision network, and output a set of joint control parameters. The joint control parameters include at least: the duty cycle of the temperature compensation circuit of the timing reference element, the power supply voltage level of the frequency division counting unit, and the refresh rate threshold of the display panel. Step S4: The closed-loop execution and calibration unit receives the joint control parameters output in step S3, parses the joint control parameters into hardware control instructions and sends them to the functional modules of the electronic timer for execution. At the same time, it monitors the actual frequency offset of the timing reference element and the actual power consumption of the display panel in real time and generates a measured deviation vector. Step S5: Receive the measured deviation vector generated in step S4 through the adaptive compensator, and feed the measured deviation vector back to the multi-objective optimization decision network in step S3. Trigger the multi-objective optimization decision network to perform online parameter optimization in the neighborhood of the joint control parameters with the goal of minimizing the weighted sum of timing error and power consumption per unit time. Update the optimization results to the policy benchmark of the dynamic power consumption strategy generator to replace the original joint control parameters in step S3 in the next round of control command issuance.

Images