Dynamic power consumption adjustment method and device based on load prediction and hierarchical dormancy, equipment and storage medium

Abstract

The invention discloses a dynamic power consumption adjusting method, device and equipment based on load prediction and hierarchical dormancy and a storage medium, and relates to the technical field of dynamic environment monitoring systems, and the method comprises the steps: obtaining processor occupation information, sensor connection quantity information, data collection frequency information, communication state information and to-be-processed data quantity information; inputting a predefined load prediction model based on the processor occupation information, the sensor connection quantity information, the data acquisition frequency information, the communication state information and the to-be-processed data quantity information to predict a load change trend in predefined time, and determining a load change trend prediction result, generating corresponding dormancy control strategy information and power consumption adjustment strategy information based on the load change trend prediction result; and controlling the system to operate normally based on the sleep control strategy information and the power consumption adjustment strategy information. The power consumption of the system is dynamically adjusted through load prediction and hierarchical dormancy, and the energy efficiency and endurance are remarkably improved.

Description

Dynamic power consumption regulation method, device, equipment and storage medium based on load prediction and hierarchical sleep mode. Technical Field

[0001] This application relates to the field of environmental monitoring system technology, and in particular to a dynamic power consumption regulation method, device, equipment and storage medium based on load prediction and hierarchical sleep. Background Technology

[0002] The Environmental Utility Unit (FSU) is a critical monitoring device used in scenarios such as communication base stations, unattended equipment rooms, and outdoor equipment cabinets. It is responsible for real-time collection, processing, and alarm reporting of environmental parameters, power parameters, and equipment operating status. Since the FSU needs to run continuously around the clock, its power consumption level directly affects the equipment's endurance, operational stability, and overall maintenance costs. In environments that rely on battery power, it is necessary to minimize energy consumption and improve system energy efficiency and reliability while ensuring real-time monitoring and alarm functions.

[0003] Currently, existing practices employ fixed operating modes or simple system-wide hibernation strategies, passively responding to real-time load changes. However, these practices lack the ability to predict load trends and often use a one-size-fits-all approach to hibernation control, uniformly implementing hibernation or wake-up for all functional modules. This fails to dynamically adjust power consumption based on load variations. For example, high power consumption under light loads leads to energy waste, while insufficient performance under heavy loads. Furthermore, the lack of load prediction capabilities results in lagging power consumption adjustment and untimely system response. Simultaneously, the failure to differentiate the importance of functional modules can easily lead to delays in critical functions or data loss, impacting system reliability. Therefore, how to accurately and effectively balance system load for dynamic power consumption adjustment has become an urgent problem to be solved.

[0004] The above content is only used to help understand the technical solution of this application and does not represent an admission that the above content is prior art. Summary of the Invention

[0005] The main objective of this application is to provide a dynamic power consumption regulation method, apparatus, device, and storage medium based on load prediction and hibernation, aiming to solve the technical problem of how to accurately and effectively balance system load for dynamic power consumption regulation.

[0006] To achieve the above objectives, this application proposes a dynamic power consumption adjustment method based on load prediction and hierarchical sleep mode. The method includes: acquiring processor occupancy information, sensor connection information, data acquisition frequency information, communication status information, and data volume information to be processed; inputting the processor occupancy information, sensor connection information, data acquisition frequency information, communication status information, and data volume information to a predefined load prediction model to predict the load change trend within a predefined time period, and determining the load change trend prediction result; generating corresponding sleep control strategy information and power consumption adjustment strategy information based on the load change trend prediction result; and controlling the system to operate normally based on the sleep control strategy information and the power consumption adjustment strategy information.

[0007] In one embodiment, the step of predicting the load change trend within a predefined time period based on the processor occupancy information, sensor connection information, data acquisition frequency information, communication status information, and the amount of data to be processed, and determining the load change trend prediction result, includes: performing data cleaning based on the processor occupancy information, sensor connection information, data acquisition frequency information, communication status information, and the amount of data to be processed to determine cleaning operation load parameters; performing normalization processing based on the cleaning operation load parameters to determine normalized operation load parameters; determining corresponding load change trend feature information and periodic feature information based on the normalized operation load parameters; and inputting the load change trend feature information and periodic feature information into the predefined load prediction model to predict the load change trend within a predefined time period to obtain the load change trend prediction result.

[0008] In one embodiment, the step of predicting the load change trend within a predefined time period based on the load change trend feature information and the periodic feature information to obtain a load change trend prediction result includes: training the predefined load prediction model according to historical load information corresponding to the load change trend feature information and the periodic feature information to determine a target load prediction model; adjusting the model parameters in the target load prediction model according to real-time load information corresponding to the load change trend feature information and the periodic feature information to obtain a time-series prediction model; and inputting the time-series prediction model based on the load change trend feature information and the periodic feature information to predict the load change trend within a predefined time period to obtain a load change trend prediction result.

[0009] In one embodiment, the step of generating corresponding sleep control strategy information and power consumption adjustment strategy information based on the load change trend prediction result includes: performing hierarchical sleep on multiple predefined functional modules based on the load change trend prediction result to determine sleep control strategy information; and adjusting the power consumption of the main control chip's operating frequency and power supply status based on the load change trend prediction result and the sleep control strategy information to determine power consumption adjustment strategy information.

[0010] In one embodiment, the step of performing hibernation on multiple predefined functional modules based on the load change trend prediction results and determining hibernation control strategy information includes: determining the load level within a predefined time period based on the load change trend prediction results; and performing hibernation on multiple predefined functional modules based on the load level to obtain hibernation control strategy information.

[0011] In one embodiment, the step of adjusting the power consumption of the main control chip based on the load change trend prediction result and the sleep control strategy information, and determining the power consumption adjustment strategy information, includes: adjusting the operating frequency and power supply status of the main control chip based on the load change trend prediction result, the sleep control strategy information, and the predefined load change trend standard result, and determining the operating frequency information and power supply status information; and obtaining the power consumption adjustment strategy information based on the operating frequency information and the power supply status information.

[0012] In one embodiment, after the step of controlling the system to operate normally based on the sleep control strategy information and the power consumption adjustment strategy information, the method further includes: monitoring the corresponding actual load changes based on the sleep control strategy information and the power consumption adjustment strategy information to determine actual load change monitoring information; when the actual load change monitoring information exceeds a preset threshold or a predefined external wake-up command is received, waking up multiple predefined functional modules from the sleep state in stages according to a predefined control strategy to determine wake-up control strategy information; adjusting the operating frequency and power supply status of the main control chip based on the wake-up control strategy information and the power consumption adjustment strategy information to determine target power consumption adjustment strategy information; and controlling the system to operate normally based on the wake-up control strategy information and the target power consumption adjustment strategy information.

[0013] Furthermore, to achieve the above objectives, this application also proposes a dynamic power consumption adjustment device based on load prediction and hierarchical sleep mode. The dynamic power consumption adjustment device based on load prediction and hierarchical sleep mode includes: an acquisition module for acquiring processor occupancy information, sensor connection information, data acquisition frequency information, communication status information, and data volume information to be processed; a processing module for inputting a predefined load prediction model based on the processor occupancy information, sensor connection information, data acquisition frequency information, communication status information, and data volume information to be processed to predict the load change trend within a predefined time period, and determine the load change trend prediction result; the processing module is also used to generate corresponding sleep control strategy information and power consumption adjustment strategy information based on the load change trend prediction result; and an execution module for controlling the system to operate normally based on the sleep control strategy information and the power consumption adjustment strategy information.

[0014] Furthermore, to achieve the above objectives, this application also proposes a dynamic power consumption regulation device based on load prediction and hierarchical sleep mode. The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. The computer program is configured to implement the steps of the dynamic power consumption regulation method based on load prediction and hierarchical sleep mode as described above.

[0015] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the steps of the dynamic power consumption adjustment method based on load prediction and hierarchical sleep described above.

[0016] One or more technical solutions proposed in this application have at least the following technical effects: The dynamic power consumption adjustment method based on load prediction and hierarchical sleep proposed in this embodiment acquires processor occupancy information, sensor connection information, data acquisition frequency information, communication status information, and data volume information to be processed; based on the processor occupancy information, sensor connection information, data acquisition frequency information, communication status information, and data volume information to be processed, a predefined load prediction model is input to predict the load change trend within a predefined time period, and the load change trend prediction result is determined; based on the load change trend prediction result, corresponding sleep control strategy information and power consumption adjustment strategy information are generated; based on the sleep control strategy information and the power consumption adjustment strategy information, the system operates normally. This application collects multi-dimensional load information such as processor usage, sensor connectivity, data acquisition frequency, communication status, and the amount of data to be processed. It then uses a time-series prediction model to predict future load change trends and dynamically generates corresponding hierarchical sleep control strategies and power consumption adjustment strategies based on the prediction results. This enables the adjustment of the system's operating status and, while ensuring the continuous operation of core functions, allows for refined management of each module. This effectively reduces energy waste under light loads and avoids insufficient response under heavy loads, thereby improving the overall system's operational stability and energy efficiency, and extending the device's battery life in energy storage power supply scenarios. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 is a flowchart illustrating the dynamic power consumption adjustment method based on load prediction and hierarchical sleep according to Embodiment 1 of this application; Figure 2 is a flowchart illustrating the dynamic power consumption adjustment method based on load prediction and hierarchical sleep according to Embodiment 2 of this application; Figure 3 is a simplified flowchart illustrating the dynamic power consumption adjustment method based on load prediction and hierarchical sleep according to the embodiments of this application; Figure 4 is a schematic diagram illustrating the module structure of the dynamic power consumption adjustment device based on load prediction and hierarchical sleep according to the embodiments of this application; Figure 5 is a schematic diagram illustrating the device structure of the hardware operating environment involved in the dynamic power consumption adjustment method based on load prediction and hierarchical sleep according to the embodiments of this application.

[0020] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0021] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0022] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0023] The main solution of this application embodiment is as follows: Acquire processor occupancy information, sensor connection information, data acquisition frequency information, communication status information, and unprocessed data volume information; based on the processor occupancy information, sensor connection information, data acquisition frequency information, communication status information, and unprocessed data volume information, input a predefined load prediction model to predict the load change trend within a predefined time period, and determine the load change trend prediction result; generate corresponding sleep control strategy information and power consumption adjustment strategy information based on the load change trend prediction result; and control the system to operate normally based on the sleep control strategy information and the power consumption adjustment strategy information.

[0024] In this embodiment, for ease of description, the following description will focus on identifying a dynamic power consumption adjustment device based on load prediction and hierarchical sleep as the execution subject.

[0025] Because existing technologies lack the ability to predict load trends, a one-size-fits-all approach is often used for sleep control, applying uniform sleep or wake-up to all functional modules. This fails to dynamically adjust power consumption based on load changes. For example, high power consumption during light loads leads to energy waste, while insufficient performance under heavy loads results in delayed power consumption adjustment and untimely system response. Furthermore, the lack of load prediction capabilities leads to the failure to differentiate the importance of functional modules, which can easily cause delays in critical functions or data loss, affecting system reliability.

[0026] This application provides a solution for a dynamic power consumption regulation method based on load prediction and hierarchical sleep, which can be applied to a dynamic power consumption regulation system based on load prediction and hierarchical sleep. The system includes a cooperating load data acquisition module, a load prediction module, a dynamic power consumption regulation module, a hierarchical sleep control module, and a wake-up and recovery module. In specific implementation, the load data acquisition module collects real-time operating load parameters of the FSU, including processor utilization, number of sensor connections, data acquisition frequency, communication link status, and length of the pending data queue. This data, after preprocessing, is input into the load prediction module. Based on historical load sequences and current real-time data, the load prediction module uses a time-series prediction model to predict the load change trend in the next time period and outputs the predicted load level and trend. The dynamic power consumption adjustment module receives the prediction result and, combined with the current actual load state, generates a corresponding dynamic power consumption adjustment strategy, such as adjusting the operating frequency and voltage of the main control chip. At the same time, the hierarchical sleep control module performs differentiated sleep control on each functional module inside the FSU according to the predicted load level. That is, the core communication and alarm modules are kept running or in shallow sleep, while non-critical data acquisition modules are in deep sleep, thereby achieving fine-grained power consumption management. Subsequently, the wake-up and recovery module continuously monitors the system load and external events. Once it detects an increase in load or that a specific wake-up condition is met, it gradually wakes up the corresponding modules according to a preset sequence, so that the system returns to the operating state that matches the current load. Through the collaborative work of multiple modules, the FSU can achieve adaptive power consumption adjustment under various load scenarios, significantly reducing overall energy consumption while ensuring the real-time performance of key functions and system stability.

[0027] As can be seen from the above embodiments, this application collects multi-dimensional load information such as processor usage, sensor connection quantity, data acquisition frequency, communication status, and amount of data to be processed, and uses a time-series prediction model to predict future load change trends. Based on the prediction results, it dynamically generates corresponding hierarchical sleep control strategies and power consumption adjustment strategies to adjust the system's operating status. Under the premise of ensuring the continuous operation of core functions, it performs refined management of each module, effectively reducing energy waste under light loads and avoiding insufficient response under heavy loads. This improves the overall system's operational stability and energy efficiency, and extends the device's battery life in energy storage power supply scenarios.

[0028] Based on this, this application provides a dynamic power consumption adjustment method based on load prediction and hierarchical sleep. Referring to Figure 1, Figure 1 is a flowchart of the first embodiment of the dynamic power consumption adjustment method based on load prediction and hierarchical sleep of this application.

[0029] In this embodiment, the dynamic power consumption adjustment method based on load prediction and hierarchical sleep includes steps S10 to S40: Step S10, acquiring processor occupancy information, sensor connection information, data acquisition frequency information, communication status information, and data volume information to be processed; it should be noted that the processor occupancy information is the proportion of computing resources consumed by the FSU main control chip in executing tasks per unit time, expressed as a percentage, representing the busy level of the system's core processing capability, wherein FSU (Field Supervision) The FSU (Functional Unit) is a key monitoring device used in scenarios such as communication base stations, unattended equipment rooms, and outdoor equipment cabinets. The sensor connection information is the total number of external sensors that have established a valid communication connection with the FSU through wired or wireless interfaces and are in data acquisition mode, such as temperature and humidity, voltage, and door magnetic sensors. The data acquisition frequency information is the time interval or frequency setting value for the FSU to periodically read or trigger data acquisition from different types of sensors according to preset strategies or instructions. The communication status information is the link status of the FSU and the upper-level monitoring center or adjacent nodes for data communication, including information such as connection establishment, data transmission rate, signal strength, and network latency. The data volume information is the length of the data queue or the size of the buffer occupied by the data that has been collected but has not yet been processed or reported.

[0030] In a specific embodiment, as an optional implementation, the system operating load parameters, including processor usage information, sensor connection information, data acquisition frequency information, communication status information, and pending data volume information, can be collected through the monitoring module inside the FSU. The collected load data is transmitted to the power consumption adjustment control unit and cached or stored. For example, processor usage information can be obtained in real time through system calls or performance monitoring interfaces of the embedded operating system; sensor connection information can be counted by polling the device management table or listening to connection heartbeat packets; data acquisition frequency information can be read from the configuration table of the task scheduler; communication status information can be obtained by querying the network interface status; and pending data volume information can be calculated by monitoring the read / write pointer position of the circular buffer or message queue.

[0031] In a specific embodiment, as another optional implementation, processor occupancy information, sensor connection information, data acquisition frequency information, communication status information, and data volume information to be processed can be uniformly collected and encapsulated into data packets by a dedicated monitoring agent module (Agent) within the FSU, and sent to the dynamic power consumption adjustment system for processing at regular intervals. For example, the monitoring agent collects the above information at a fixed period, such as collecting it every second, and writes it to shared memory or sends it to the power consumption adjustment control unit through the internal bus. In order to reduce the additional load brought by the acquisition overhead itself, the acquisition of some parameters can also adopt a change triggering or differential reporting mechanism. For example, the updated information is only reported when the number of sensor connections changes or the communication link status changes. For continuously changing quantities such as processor occupancy, the sliding average value can be collected instead of the instantaneous value.

[0032] In a specific embodiment, if certain parameters are temporarily unavailable, such as a specific model of sensor that does not support reporting sensor connection information, a preset default value or the most recent valid value can be used as a substitute, and this should be noted in the system log to ensure the continuous operation of the load prediction process.

[0033] Step S20: Based on the processor occupancy information, the sensor connection information, the data acquisition frequency information, the communication status information, and the amount of data to be processed, a predefined load prediction model is input to predict the load change trend within a predefined time period, and the load change trend prediction result is determined. It should be noted that the load change trend prediction result is a quantitative output of the overall load status and change pattern of the FSU within a future time period.

[0034] Understandably, the load change trend within a predefined time period is the dynamic change pattern and regularity of the FSU's integrated operating load within a future time window set by the system, such as the next 10 minutes. It is an overall description of the load evolution process over a period of time, including the starting point of the load level, the rate of change, the fluctuation period, and the possible extreme points. The purpose of analyzing the load change trend within a predefined time period is to achieve forward-looking power management, enabling the system to prepare in advance and smoothly switch states, avoiding response delays or performance bottlenecks caused by sudden load changes.

[0035] In a specific embodiment, as an optional implementation, data cleaning is performed based on the processor occupancy information, the sensor connection information, the data acquisition frequency information, the communication status information, and the amount of data to be processed to determine the cleaning operation load parameters; normalization processing is performed based on the cleaning operation load parameters to determine the normalized operation load parameters; corresponding load change trend characteristics and periodic characteristics are determined based on the normalized operation load parameters; the load change trend characteristics and periodic characteristics are input into a predefined load prediction model to predict the load change trend within a predefined time period, obtaining the load change trend prediction result. That is, before inputting the collected multi-dimensional load data into the time series prediction model, it is necessary to perform systematic preprocessing to improve data quality and model performance. For example, data cleaning is performed on the collected load data to remove outliers and invalid data, ensuring the accuracy and consistency of the data. The system employs interpolation or forward imputation methods to fill in transient missing values, thereby ensuring the accuracy and temporal consistency of the input data. For different time-series load parameters, such as processor utilization as a percentage, sensor connection count as an integer, and data acquisition frequency as a Hertz, they can be mapped to a unified numerical range or converted to values ​​conforming to a standard normal distribution, such as [0,1], by using maximum or minimum normalization to eliminate the interference of dimensional differences on model training and improve the convergence speed and generalization ability of the time-series prediction model. Statistical analysis is performed on the cleaned and normalized load time-series data to extract key features that can characterize the dynamic changes in system load. For example, the mean, variance, and first difference can be calculated using a sliding window to depict short-term load fluctuation trends. Fourier transform, autocorrelation analysis, and other methods can be used to identify the periodic patterns of load changes. The extracted trend features and periodic features can together constitute the input feature vector of the load prediction model.

[0036] Based on the historical load information corresponding to the load change trend characteristics and the periodic characteristics, a predefined load prediction model is trained to determine the target load prediction model. The model parameters in the target load prediction model are adjusted based on the real-time load information corresponding to the load change trend characteristics and the periodic characteristics to obtain a time-series prediction model. This model is then used to predict the load change trend within a predefined time period, yielding a load change trend prediction result. In other words, preprocessed historical load data is used as the training set to supervisedly train the selected load prediction model. For example, the training set is indexed by equally spaced time points, where each sample contains processor occupancy information, sensor connection information, data acquisition frequency information, communication status information, and the amount of data to be processed collected at the same time. Using an LSTM (Long Short-Term Memory) model or a GRU (Gated Recurrent Unit) model can effectively capture long-term dependencies and complex nonlinear patterns in the data. Pro... The PHET model excels at predicting time series with clear trends and seasonality. During training, historical load information is input into the model, enabling it to learn the mapping relationship between historical multi-dimensional load states and future load states. The trained model is the target load prediction model. After training, the model's internal parameters can be continuously adjusted using algorithms such as backpropagation, combined with real-time load information, to obtain the time series prediction model. This model has the ability to predict the overall load change trend over a future period based on the real-time input multi-dimensional load sequence. To ensure the accuracy and reliability of the prediction, the time series prediction model needs continuous optimization and evaluation. Model performance is optimized through methods such as cross-validation and hyperparameter tuning, and the prediction results are quantitatively evaluated using metrics such as mean absolute error (MAE) and root mean square error (RMSE). Furthermore, the time series prediction model supports an online learning mechanism, enabling incremental updates based on the latest system operating data, thus adapting to the long-term evolution of load patterns and ensuring the effectiveness of the prediction model in practical applications.

[0037] In a specific embodiment, as another optional implementation method, the predefined time can be dynamically configured for different application scenarios or time periods. For example, in a base station scenario where the day and night load patterns differ significantly, a shorter prediction window can be set during the day to respond quickly to changes, such as 5 minutes, while a longer window can be set at night to achieve deeper and smoother energy saving, such as 30 minutes. This can be set through the configuration interface or policy file.

[0038] In a specific embodiment, if abnormal input data or low confidence of the time series prediction model is detected during the prediction process, the system can adopt a conservative strategy, such as maintaining the current power consumption state or making short-term predictions based on a simpler moving average algorithm, while issuing a log alarm to prompt that the data source needs to be checked or the time series prediction model needs to be retrained.

[0039] In one feasible implementation, step S20 may include steps A11 to A14: Step A11, data cleaning is performed based on the processor occupancy information, the sensor connection information, the data acquisition frequency information, the communication status information, and the amount of data to be processed, to determine the cleaning operation load parameters; it should be noted that the cleaning operation load parameters are the operation load parameters after data cleaning. For example, the processor occupancy information, sensor connection information, data acquisition frequency information, communication status information, and the amount of data to be processed can be filtered and corrected by means of range verification, mutation detection, and continuity analysis, to remove outliers that exceed physical limits, to correct invalid data caused by transmission errors, and to fill in missing values ​​caused by sampling loss using forward padding or linear interpolation.

[0040] Step A12: Normalize the cleaning operation load parameters to determine the normalized operation load parameters. It should be noted that the normalized operation load parameters are the result of mapping the cleaning operation load parameters of different physical units and time series to a unified numerical scale through mathematical transformation. For example, minimum or maximum normalization can be used to transform the cleaning operation load parameters to the [0, 1] interval, or to transform them into a distribution with a mean of 0 and a standard deviation of 1. Normalization eliminates the differences in the numerical range between parameters of different dimensions, so that the time series prediction model can learn the influence of each feature equally and effectively, and avoid certain large numerical features dominating model training.

[0041] Step A13: Determine the corresponding load change trend characteristic information and periodic characteristic information based on the normalized operating load parameters. It should be noted that the load change trend characteristic information is a quantitative indicator extracted from the normalized parameter sequence to characterize the direction and intensity of short-term load fluctuations. For example, the mean, variance, and first-order difference mean of the sequence are calculated through a sliding window. The periodic characteristic information is a repetitive pattern identified from historical load data through Fast Fourier Transform (FFT) or autocorrelation analysis. For example, the periodic fluctuation component caused by daily peaks or hourly regular data collection tasks.

[0042] Step A14: Based on the load change trend characteristic information and the periodic characteristic information, input the predefined load prediction model to predict the load change trend within a predefined time period, and obtain the load change trend prediction result.

[0043] It is understood that the load change trend prediction result can be a set of prediction data, including the predicted load level at each future time point, the predicted value sequence of key load indicators, the trend direction of load change, and the possible occurrence time of load peak or trough. The predicted load level at each future time point can be a high, medium, or low predicted load level for the next 5 minutes, 15 minutes, or 30 minutes. The predicted value sequence of key load indicators can be a comprehensive load index, and the trend direction of load change can be rising, falling, or stable.

[0044] In another feasible implementation, step S20 may include steps B11 to B13: Step B11, training a predefined load prediction model based on the load change trend feature information and the historical load information corresponding to the periodic feature information to determine the target load prediction model; it should be noted that the target load prediction model is a model with the ability to learn the basic load change pattern obtained by training the selected load prediction model architecture with historical load information of a longer period, such as historical load information of the past month.

[0045] It is understood that the target load prediction model has optimized its internal parameters to a relatively stable state by learning the correlation patterns, long-term trends and periodicity between load and various parameters in historical sequences, and is able to perform preliminary trend inference and state prediction on newly input load characteristics.

[0046] Step B12: Adjust the model parameters in the target load prediction model according to the real-time load information corresponding to the load change trend characteristic information and the periodic characteristic information to obtain the time series prediction model. It should be noted that the time series prediction model is obtained by continuously receiving the latest real-time load data stream from the system after the target load prediction model is put into online operation, and by slightly updating its model parameters based on incremental data through online learning or fine-tuning technology.

[0047] Step B13: Based on the load change trend characteristic information and the periodic characteristic information, input the time series prediction model to predict the load change trend within a predefined time period, and obtain the load change trend prediction result.

[0048] It is understandable that the time-series prediction model not only inherits the general patterns learned from historical data, but also enhances its adaptability to short-term load fluctuations and emerging patterns in the current operating environment through real-time adjustments, thereby making it practically used to make high-precision predictions of load change trends within a predefined time period in the future.

[0049] Step S30: Generate corresponding hibernation control strategy information and power consumption adjustment strategy information based on the load change trend prediction results. It should be noted that the hibernation control strategy information is a set of control instructions for the operating status of each functional module inside the FSU in order to coordinate the system power consumption and real-time performance requirements. For example, full-speed operation instruction, shallow hibernation instruction, deep hibernation instruction and complete shutdown instruction. The power consumption adjustment strategy information is a specific configuration scheme for dynamically adjusting the working voltage and operating frequency of the system main control chip to match the current and predicted load levels.

[0050] In a specific embodiment, as an optional implementation, multiple predefined functional modules are subjected to hibernation based on the load change trend prediction results to determine hibernation control strategy information; the operating frequency and power supply status of the main control chip are adjusted based on the load change trend prediction results and the hibernation control strategy information to determine power consumption adjustment strategy information. That is, the load level of the FSU in the future can be determined based on the load change trend prediction results. For example, the load change trend prediction results output a comprehensive load index prediction sequence for a predefined time step in the future. This sequence is mapped to the corresponding load level sequence according to a preset threshold range, and the change trend is determined by combining the slope analysis of the sequence. For example, if the load will steadily increase from medium to high within the next 10 minutes, the load level within the predefined time is obtained. The FSU is then adjusted according to the load level. The internal functional modules implement hierarchical hibernation control, adopting different operating and hibernation strategies for core modules, critical modules, and non-critical modules. At the same time, power consumption adjustment strategy information generates corresponding instructions to dynamically adjust the operating frequency and power supply status of the main control chip, so that the system power consumption matches the load level. For example, the operating frequency of the main control chip CPU is gradually reduced from the full speed of 1GHz to the energy-saving 400MHz, and its core voltage is reduced accordingly.

[0051] In a specific embodiment, as another optional implementation, if the load change trend prediction result indicates that the load will rise rapidly, the sleep control strategy information will execute a pre-wake-up strategy. Before the load reaches the threshold, several data processing modules in deep sleep will be woken up to standby state in advance to avoid response delays during performance surges. Correspondingly, the power consumption adjustment strategy information will instruct the power management unit to increase the voltage tolerance of the power supply rail in advance and instruct the main control chip CPU to lock the frequency at a higher performance level to prepare for the upcoming high load task. Both the sleep control strategy information and the power consumption adjustment strategy information can be encapsulated in JSON format, including fields such as target module, target status, adjustment parameters, and effective time window, and then directly sent to each execution unit. If the load change trend prediction result has low confidence or is abnormal, the system can downgrade to a conservative strategy based on the current real-time load, such as only performing a small frequency adjustment and maintaining the full operation of the core modules, while issuing alarm information that requires detailed verification.

[0052] Step S40: Control the system to operate normally based on the sleep control strategy information and the power consumption adjustment strategy information.

[0053] Understandably, hibernation control strategy information and power consumption regulation strategy information can be translated into specific control actions for FSU hardware and software modules to achieve actual switching of system power consumption and performance status, ensuring stable and efficient operation of the system under new load expectations.

[0054] In a specific embodiment, as an optional implementation method, the normal operation of the system based on the hibernation control strategy information and the power consumption adjustment strategy information is achieved through an integrated control strategy executor. The executor parses and schedules the received strategy information. According to the strategy content, the executor accurately issues a sequence of operation instructions, including module state switching, CPU frequency / voltage adjustment, etc., through the system bus, power management interface, clock control unit, and dedicated control registers of each functional module, and monitors the execution results to complete the closed-loop control from decision-making to implementation.

[0055] In a specific embodiment, as another optional implementation, when the system is running during a period of low load and is predicted to remain stable, according to the generated sleep control strategy, the control strategy executor will send a command to the designated non-critical sensor interface chip via the I2C / SPI bus to enter the low-power mode. At the same time, according to the power adjustment strategy, the main core frequency will be set to the lowest level through the CPU frequency adjustment driver of the operating system. After all instructions are executed, the system enters a low-power steady-state operating mode that matches the current and predicted load. If the strategy requires pre-wake-up or rapid state switching, the control strategy executor will use an asynchronous, phased delivery method. For example, it will first power on the module to be woken up by controlling the power switch, and after its power supply stabilizes, it will send an initialization command through the software interface to make it enter the standby state, and trigger it to start working at a preset precise time.

[0056] In a specific embodiment, the actual load changes are monitored based on the sleep control strategy information and the power consumption adjustment strategy information to determine the actual load change monitoring information. When the actual load change monitoring information exceeds a preset threshold or a predefined external wake-up command is received, multiple predefined functional modules are woken up from the sleep state in stages according to a predefined control strategy to determine the wake-up control strategy information. Based on the wake-up control strategy information and the power consumption adjustment strategy information, the operating frequency and power supply status of the main control chip are adjusted to determine the target power consumption adjustment strategy information. Based on the wake-up control strategy information and the target power consumption adjustment strategy information, the system is controlled to operate normally. That is, the system continuously monitors the actual load changes during operation. When a load change or external triggering condition is detected, the system wakes up the corresponding functional modules according to a preset strategy and restores them to a suitable operating state.

[0057] In one feasible implementation, step S40 may include steps C11 to C14: Step C11, based on the sleep control strategy information and the power consumption adjustment strategy information, the corresponding actual load change is monitored to determine the actual load change monitoring information; it should be noted that the actual load change monitoring information is a data set that continuously tracks and quantifies the actual operating load fluctuation after the system enters a specific power consumption state according to a predetermined strategy.

[0058] It is understood that the actual load change monitoring information is obtained in real time through the load data acquisition module, which acquires indicators such as the actual processor utilization, sensor data throughput, and communication load of the system after adjustment.

[0059] Step C12: When the actual load change monitoring information exceeds a preset threshold or a predefined external wake-up command is received, multiple predefined functional modules are woken up from the sleep state in stages according to the predefined control strategy, and wake-up control strategy information is determined. It should be noted that the wake-up control strategy information is a step-by-step wake-up command plan for each functional module generated when the system needs to exit the low-power state due to a surge in actual load or in response to external events.

[0060] It is understood that the wake-up control strategy information clarifies the wake-up priority, wake-up sequence, initial working state and dependencies of different modules, in order to avoid simultaneous power-up or excessive instantaneous current while quickly restoring system performance.

[0061] Step C13: Based on the wake-up control strategy information and the power consumption adjustment strategy information, adjust the operating frequency and power supply status of the main control chip to determine the target power consumption adjustment strategy information; it should be noted that the target power consumption adjustment strategy information is a power consumption configuration scheme that recalculates and sets the operating point of the main control chip and related power supply circuits after the graded wake-up process is started in order to match the performance requirements of all or part of the functional modules that are about to be restored.

[0062] It is understandable that the target power consumption adjustment strategy information, based on the original power consumption adjustment strategy and combined with the new load expectation introduced by the wake-up strategy, calculates a new and higher CPU operating frequency and voltage value through dynamic voltage and frequency adjustment. At the same time, it is necessary to coordinate the power consumption status of peripheral units such as bus clock and memory controller to ensure that the system has sufficient computing power and energy efficiency to handle the expected load after wake-up.

[0063] Step C14: Control the system to operate normally based on the wake-up control strategy information and the target power consumption adjustment strategy information.

[0064] Understandably, based on the wake-up control strategy information and the target power consumption adjustment strategy information, the module wake-up operation and chip power consumption parameter reconfiguration can be executed sequentially and orderly through hardware interfaces such as the system control bus, power management unit, and clock controller. For example, wake-up signals are sent sequentially to each module in the sleep state and wait for them to be ready. The phase-locked loop of the main control chip and the register settings of the power management IC are adjusted according to the new target frequency and voltage value, so that the entire system can smoothly and quickly transition from a low-power state to a normal operating state that can meet the current actual load requirements.

[0065] This embodiment proposes a dynamic power consumption adjustment method based on load prediction and hierarchical sleep mode. The method acquires processor occupancy information, sensor connection information, data acquisition frequency information, communication status information, and the amount of data to be processed. Based on these information, a predefined load prediction model is input to predict the load change trend within a predefined time period, and the predicted load change trend is determined. Based on the predicted load change trend, corresponding sleep control strategy information and power consumption adjustment strategy information are generated. The system operates normally based on these two strategies. This application addresses the technical challenge of accurately and effectively balancing system load for dynamic power consumption regulation. Compared to existing technologies, it collects multi-dimensional load information, including processor usage, sensor connectivity, data acquisition frequency, communication status, and the amount of data to be processed. It then uses a time-series prediction model to predict future load trends and dynamically generates corresponding hierarchical sleep control and power consumption regulation strategies based on the prediction results. This allows for the regulation of system operation status, ensuring the continuous operation of core functions while enabling refined management of each module. This effectively reduces energy waste under light loads and avoids insufficient response under heavy loads, thereby improving overall system stability and energy efficiency and extending the device's runtime in energy storage power supply scenarios.

[0066] Based on the first embodiment of this application, in the second embodiment of this application, the same or similar content as the first embodiment can be referred to the above description, and will not be repeated hereafter.

[0067] In this embodiment, referring to Figure 2, which is a flowchart of the dynamic power consumption adjustment method based on load prediction and hierarchical hibernation provided in Embodiment 2 of this application, step S30 specifically includes steps S31 to S32: Step S31, hierarchical hibernation is performed on multiple predefined functional modules based on the load change trend prediction results to determine hibernation control strategy information; it can be understood that hierarchical hibernation predefines the internal functional modules of the FSU as core modules (such as main control communication link, emergency alarm processing unit), key modules (such as main environmental sensor data acquisition interface) and non-key modules (such as auxiliary log recording unit, backup sensor interface). For different predicted load levels, corresponding hibernation depth and strategy are preset for each type of module. By applying strict hibernation control to non-core modules, deep energy saving is achieved, while ensuring that core and key modules maintain necessary operation or are in a shallow hibernation state that can be quickly woken up, thereby achieving the best balance between reducing the overall power consumption of the system and ensuring the availability of key functions.

[0068] In a specific embodiment, as an optional implementation, the load level within a predefined time period is determined based on the load change trend prediction result; and multiple predefined functional modules are hibernated in a hibernation manner based on the load level to obtain hibernation control strategy information.

[0069] In a specific embodiment, as another optional implementation, when the load change trend prediction result indicates that the load level will be low in the future, the core communication module is set to remain active or standby, the critical data acquisition module is set to shallow sleep, and the non-critical module is set to deep sleep or completely shut down. When the load change trend prediction result indicates a rapid drop from high to low, the hierarchical sleep can be designed to be executed in stages with delays. For example, after the load is confirmed to be decreasing, the non-critical module is put into deep sleep. After a period of time is confirmed that the load is stable at a low level, the acquisition frequency of the critical module is reduced. The core module maintains basic connectivity throughout the process to prevent short-term load fluctuations from causing frequent switching of module states and improve system stability. The hierarchical classification of modules and corresponding strategies can be defined and adjusted through configuration files, so that the solution can be flexibly adapted to different FSU models or different application scenarios. For example, the definition of module importance may be different for communication base stations and unattended equipment rooms.

[0070] In a specific embodiment, if the system detects a significant deviation between the actual load and the prediction during the hierarchical hibernation process, for example, if the actual load is still higher than expected after a non-critical module goes into hibernation, the current hibernation strategy can be partially revoked or adjusted according to a preset feedback mechanism, such as waking up a critical module in advance to share the processing pressure.

[0071] In one feasible implementation, step S31 may include steps D11~D12: Step D11, determining the load level within a predefined time period based on the load change trend prediction result; it should be noted that the load level is the continuous or quantified load trend prediction value for a future period of time mapped to the corresponding state level according to a preset threshold range. The load level can be divided into three levels: high, medium and low, or further refined into multiple levels such as extremely high, high, medium, low, and idle. Each level will correspond to a clear set of functional module operation and power consumption benchmark strategies.

[0072] Step D12: Based on the load level, perform hibernation on multiple predefined functional modules to obtain hibernation control strategy information.

[0073] It should be noted that, assuming the system predefines three functional modules, such as the core communication module (module A), the key data acquisition module (module B), and the log recording module (module C), when the determined load level is low, module A maintains low-frequency standby (maintains heartbeat, reduces polling frequency), module B enters deep sleep (wakes up to sample once every 10 minutes), and module C enters a shut-down state. At this time, the obtained sleep control strategy information is: {Module A: State = Low-frequency standby, Parameter = Polling interval 5 seconds}; {Module B: State = Deep sleep, Parameter = Wake-up cycle 600 seconds}; {Module C: State = Shut-down}.

[0074] Step S32: Based on the load change trend prediction results and the sleep control strategy information, adjust the power consumption of the main control chip's operating frequency and power supply status to determine the power consumption adjustment strategy information.

[0075] Understandably, power consumption adjustment dynamically changes the operating parameters of the main control chip, including its operating frequency and power supply voltage / power state, based on the predicted future load level and the planned sleep state of functional modules. This allows the chip's computing power output and energy consumption level to be precisely matched with the actual and expected processing needs, thereby optimizing system energy efficiency.

[0076] In a specific embodiment, as an optional implementation, the operating frequency and power supply status of the main control chip are adjusted based on the load change trend prediction results, the sleep control strategy information, and the predefined load change trend standard results to determine the operating frequency information and power supply status information; power consumption adjustment strategy information is obtained based on the operating frequency information and the power supply status information.

[0077] In a specific embodiment, as another optional implementation, power consumption adjustment needs to be considered in conjunction with tiered hibernation. For example, the hibernation strategy decides to put the high-speed data sampling module into hibernation, resulting in a reduction in CPU interrupt processing load. When calculating the target frequency, the power consumption adjustment module will choose a frequency value lower than simply looking at the load level, achieving more extreme energy saving. At the same time, the power management unit puts the idle CPU cores into a deeper hibernation state. Therefore, the power consumption adjustment strategy can be forward-looking. If the load prediction result shows that the load will rise from low to medium in the next 2 minutes, the adjustment strategy will not immediately increase the frequency to the level corresponding to the medium load. Instead, it will plan a ramp-up curve or increase the frequency a short time window before the critical point to smooth power consumption fluctuations and ensure performance continuity.

[0078] In a specific embodiment, if a hardware anomaly such as excessively high chip temperature or unstable power supply is detected during the power consumption adjustment process, the adjustment strategy will have a safe rollback mechanism. For example, it may interrupt further frequency and voltage reduction operations, or even briefly increase the frequency to ensure the reliable completion of the computing task, while reporting the abnormal event to ensure the safety of system functions.

[0079] In one feasible implementation, step S32 may include steps E11 to E13: Step E11, based on the load change trend prediction result, the sleep control strategy information, and the predefined load change trend standard result, the operating frequency and power supply status of the main control chip are adjusted to determine the operating frequency information and power supply status information; it should be noted that, based on the load change trend prediction result, the sleep control strategy information, and the predefined load change trend standard result, the operating point of the main control chip that best matches the expected future load and the current module status can be calculated, i.e., the specific operating frequency information and power supply status information.

[0080] Step E12: Obtain power consumption adjustment strategy information based on the operating frequency information and the power supply status information.

[0081] It should be noted that the operating frequency information and power supply status information can be encapsulated into a set of control commands that can be directly recognized and executed by the underlying hardware driver or power management unit, i.e., power consumption regulation strategy information, which may include target values, execution timing, sequence, and necessary safety verification instructions.

[0082] This embodiment proposes a dynamic power consumption adjustment method based on load prediction and hierarchical hibernation. Based on the load change trend prediction results, multiple predefined functional modules are hierarchically hibernated to determine hibernation control strategy information. Based on the load change trend prediction results and the hibernation control strategy information, the operating frequency and power supply status of the main control chip are adjusted to determine power consumption adjustment strategy information. This solves the technical problem of how to accurately and effectively balance system load for dynamic power consumption adjustment. Compared with existing technologies, this application implements differentiated hibernation through hierarchical hibernation control. While ensuring the continuous availability of core functions, non-critical modules are placed in a deep low-power state, thereby effectively reducing system static power consumption and avoiding functional interruptions or data loss. Simultaneously, by dynamically adjusting the main control chip frequency and power supply status, the chip's computing power is precisely matched with the dynamic load, eliminating excessive computing power under low load and performance bottlenecks under high load, significantly improving energy utilization efficiency. Therefore, while ensuring system real-time performance and reliability, it significantly extends the device's runtime in energy storage scenarios and reduces overall operation and maintenance costs.

[0083] For example, to help understand the implementation flow of the dynamic power consumption adjustment method based on load prediction and hierarchical sleep obtained by combining this embodiment with the above embodiment one, please refer to Figure 3. Figure 3 provides a simplified flowchart of the dynamic power consumption adjustment method based on load prediction and hierarchical sleep. Specifically: Referring to embodiment one, processor occupancy information, sensor connection information, data acquisition frequency information, communication status information, and data volume information to be processed are obtained; based on the processor occupancy information, sensor connection information, data acquisition frequency information, communication status information, and data volume information to be processed, a predefined load prediction model is input to predict the load change trend within a predefined time period, and the load change trend prediction result is determined; based on the load change trend prediction result, corresponding sleep control strategy information and power consumption adjustment strategy information are generated; based on the sleep control strategy information and the power consumption adjustment strategy information, the system operates normally. Referring to embodiment two, based on the load change trend prediction result, multiple predefined functional modules are hierarchically put into sleep mode to determine the sleep control strategy information; based on the load change trend prediction result and the sleep control strategy information, the operating frequency and power supply status of the main control chip are adjusted to determine the power consumption adjustment strategy information. The system collects multi-dimensional load parameters of the FSU in real time. After preprocessing such as cleaning, normalization, and feature extraction, the collected data is used to train a predefined load prediction model to obtain a time-series prediction model. The time-series prediction model is then used for analysis to predict the load change trend and determine the load level in the future. Differentiated dynamic control is executed according to the predicted load level. Under low load, hierarchical sleep control is initiated, causing non-core modules to enter deep sleep and the system as a whole enters a low-power mode. Under medium load, a strategy of running some modules is adopted to optimize energy efficiency while ensuring performance, and the system is in a power balance mode. Under high load, all critical modules are ensured to run, and the system switches to normal operation mode to ensure processing capacity. Throughout the entire operation, the system continuously performs real-time operation monitoring and wake-up judgment. Once the actual load change exceeds the expectation or meets the wake-up condition, a state switch is immediately triggered. At the same time, the system uses the monitoring results as load change feedback for strategy correction and iterative optimization of the prediction model, thereby achieving closed-loop, adaptive, and dynamic power consumption fine management.

[0084] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the dynamic power consumption adjustment method based on load prediction and hierarchical sleep in this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0085] This application also provides a dynamic power consumption adjustment device based on load prediction and hierarchical sleep mode. Referring to Figure 4, the dynamic power consumption adjustment device based on load prediction and hierarchical sleep mode includes: an acquisition module 10, used to acquire processor occupancy information, sensor connection information, data acquisition frequency information, communication status information, and data volume information to be processed; a processing module 20, used to input a predefined load prediction model based on the processor occupancy information, sensor connection information, data acquisition frequency information, communication status information, and data volume information to be processed to predict the load change trend within a predefined time period, and determine the load change trend prediction result; the processing module 20 is also used to generate corresponding sleep control strategy information and power consumption adjustment strategy information based on the load change trend prediction result; and an execution module 30, used to control the system to operate normally based on the sleep control strategy information and the power consumption adjustment strategy information.

[0086] The processing module 20 is further configured to perform data cleaning based on the processor occupancy information, the sensor connection information, the data acquisition frequency information, the communication status information, and the amount of data to be processed, and determine the cleaning operation load parameters; perform normalization processing based on the cleaning operation load parameters to determine the normalized operation load parameters; determine the corresponding load change trend characteristic information and periodic characteristic information based on the normalized operation load parameters; and input the load change trend characteristic information and the periodic characteristic information into a predefined load prediction model to predict the load change trend within a predefined time period, thereby obtaining the load change trend prediction result.

[0087] The processing module 20 is further configured to train a predefined load prediction model based on the historical load information corresponding to the load change trend feature information and the periodic feature information to determine a target load prediction model; adjust the model parameters in the target load prediction model based on the real-time load information corresponding to the load change trend feature information and the periodic feature information to obtain a time-series prediction model; and predict the load change trend within a predefined time period based on the load change trend feature information and the periodic feature information input into the time-series prediction model to obtain a load change trend prediction result.

[0088] The processing module 20 is further configured to perform hierarchical hibernation on multiple predefined functional modules based on the load change trend prediction results, and determine hibernation control strategy information; and to adjust the power consumption of the main control chip's operating frequency and power supply status based on the load change trend prediction results and the hibernation control strategy information, and determine power consumption adjustment strategy information.

[0089] The processing module 20 is further configured to determine the load level within a predefined time period based on the load change trend prediction result; and to perform hibernation on multiple predefined functional modules based on the load level to obtain hibernation control strategy information.

[0090] The processing module 20 is further configured to adjust the operating frequency and power supply status of the main control chip based on the load change trend prediction result, the sleep control strategy information and the predefined load change trend standard result, and determine the operating frequency information and power supply status information; and obtain power consumption adjustment strategy information based on the operating frequency information and the power supply status information.

[0091] The execution module 30 is further configured to monitor the corresponding actual load changes based on the sleep control strategy information and the power consumption adjustment strategy information, and determine the actual load change monitoring information; when the actual load change monitoring information exceeds a preset threshold or a predefined external wake-up command is received, to wake up multiple predefined functional modules from the sleep state in stages according to a predefined control strategy, and determine the wake-up control strategy information; to adjust the power consumption of the main control chip's operating frequency and power supply status based on the wake-up control strategy information and the power consumption adjustment strategy information, and determine the target power consumption adjustment strategy information; and to control the system to operate normally based on the wake-up control strategy information and the target power consumption adjustment strategy information.

[0092] The dynamic power consumption regulation device based on load prediction and hierarchical sleep mode provided in this application, employing the dynamic power consumption regulation method based on load prediction and hierarchical sleep mode in the above embodiments, can solve the technical problem of how to accurately and effectively balance system load for dynamic power consumption regulation. Compared with the prior art, the beneficial effects of the dynamic power consumption regulation device based on load prediction and hierarchical sleep mode provided in this application are the same as those of the dynamic power consumption regulation method based on load prediction and hierarchical sleep mode provided in the above embodiments, and other technical features in the dynamic power consumption regulation device based on load prediction and hierarchical sleep mode are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0093] This application provides a dynamic power consumption regulation device based on load prediction and hierarchical sleep mode. The dynamic power consumption regulation device based on load prediction and hierarchical sleep mode includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the dynamic power consumption regulation method based on load prediction and hierarchical sleep mode in the above embodiment 1.

[0094] Referring to Figure 5 below, a schematic diagram of a dynamic power consumption regulation device based on load prediction and hierarchical sleep mode, suitable for implementing embodiments of this application, is shown. The dynamic power consumption regulation device based on load prediction and hierarchical sleep mode in embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. The dynamic power consumption regulation device based on load prediction and hierarchical sleep mode shown in Figure 5 is merely an example and should not impose any limitations on the functionality and scope of use of embodiments of this application.

[0095] As shown in Figure 5, the dynamic power consumption regulation device based on load prediction and hierarchical sleep may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in ROM (Read Only Memory) 1002 or a program loaded from storage device 1003 into RAM (Random Access Memory) 1004. RAM 1004 also stores various programs and data required for the operation of the dynamic power consumption regulation device based on load prediction and hierarchical sleep. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via bus 1005. Input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows the load-prediction and hierarchical sleep-based dynamic power regulation device to communicate wirelessly or wiredly with other devices to exchange data. Although the figure shows a load-prediction and hierarchical sleep-based dynamic power regulation device with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented alternatively.

[0096] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0097] The dynamic power consumption regulation device based on load prediction and hierarchical sleep mode provided in this application, employing the dynamic power consumption regulation method based on load prediction and hierarchical sleep mode in the above embodiments, can solve the technical problem of how to accurately and effectively balance system load for dynamic power consumption regulation. Compared with the prior art, the beneficial effects of the dynamic power consumption regulation device based on load prediction and hierarchical sleep mode provided in this application are the same as the beneficial effects of the dynamic power consumption regulation method based on load prediction and hierarchical sleep mode provided in the above embodiments, and other technical features in this dynamic power consumption regulation device based on load prediction and hierarchical sleep mode are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0098] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0099] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0100] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the dynamic power consumption regulation method based on load prediction and hierarchical sleep in the above embodiments.

[0101] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0102] The aforementioned computer-readable storage medium may be included in a dynamic power management device based on load prediction and hierarchical sleep; or it may exist independently and not assembled into a dynamic power management device based on load prediction and hierarchical sleep.

[0103] The aforementioned computer-readable storage medium carries one or more programs. When these programs are executed by a dynamic power consumption regulation device based on load prediction and hierarchical sleep, the device performs the following actions: acquires processor occupancy information, sensor connection information, data acquisition frequency information, communication status information, and data volume information to be processed; based on the processor occupancy information, sensor connection information, data acquisition frequency information, communication status information, and data volume information to be processed, it inputs a predefined load prediction model to predict the load change trend within a predefined time period and determines the load change trend prediction result; based on the load change trend prediction result, it generates corresponding sleep control strategy information and power consumption regulation strategy information; and based on the sleep control strategy information and power consumption regulation strategy information, it controls the system to operate normally.

[0104] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a Local Area Network (LAN) or a Wide Area Network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0105] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0106] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0107] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described dynamic power consumption regulation method based on load prediction and graded sleep, thereby solving the technical problem of how to accurately and effectively balance system load for dynamic power consumption regulation. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the dynamic power consumption regulation method based on load prediction and graded sleep provided in the above embodiments, and will not be repeated here.

[0108] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A dynamic power consumption regulation method based on load prediction and hierarchical sleep mode, characterized in that, The method includes: acquiring processor occupancy information, sensor connection information, data acquisition frequency information, communication status information, and unprocessed data volume information; inputting a predefined load prediction model based on the processor occupancy information, sensor connection information, data acquisition frequency information, communication status information, and unprocessed data volume information to predict the load change trend within a predefined time period, and determining the load change trend prediction result; generating corresponding sleep control strategy information and power consumption adjustment strategy information based on the load change trend prediction result; and controlling the system to operate normally based on the sleep control strategy information and the power consumption adjustment strategy information.

2. The method as described in claim 1, characterized in that, The step of predicting the load change trend within a predefined time period based on the processor occupancy information, sensor connection information, data acquisition frequency information, communication status information, and the amount of data to be processed by inputting a predefined load prediction model, and determining the load change trend prediction result includes: performing data cleaning based on the processor occupancy information, sensor connection information, data acquisition frequency information, communication status information, and the amount of data to be processed to determine the cleaning operation load parameters; performing normalization processing based on the cleaning operation load parameters to determine the normalized operation load parameters; determining the corresponding load change trend feature information and periodic feature information based on the normalized operation load parameters; and inputting the load change trend feature information and the periodic feature information into the predefined load prediction model to predict the load change trend within a predefined time period to obtain the load change trend prediction result.

3. The method as described in claim 2, characterized in that, The step of predicting the load change trend within a predefined time period based on the load change trend feature information and the periodic feature information to obtain the load change trend prediction result includes: training the predefined load prediction model according to the historical load information corresponding to the load change trend feature information and the periodic feature information to determine the target load prediction model; adjusting the model parameters in the target load prediction model according to the real-time load information corresponding to the load change trend feature information and the periodic feature information to obtain the time series prediction model; and predicting the load change trend within a predefined time period based on the load change trend feature information and the periodic feature information to obtain the load change trend prediction result.

4. The method as described in claim 1, characterized in that, The steps of generating corresponding sleep control strategy information and power consumption adjustment strategy information based on the load change trend prediction results include: performing hierarchical sleep on multiple predefined functional modules based on the load change trend prediction results to determine sleep control strategy information; and adjusting the power consumption of the main control chip's operating frequency and power supply status based on the load change trend prediction results and the sleep control strategy information to determine power consumption adjustment strategy information.

5. The method as described in claim 4, characterized in that, The step of performing hibernation in a tiered manner on multiple predefined functional modules based on the load change trend prediction results and determining hibernation control strategy information includes: determining the load level within a predefined time period based on the load change trend prediction results; and performing hibernation in a tiered manner on multiple predefined functional modules based on the load level to obtain hibernation control strategy information.

6. The method as described in claim 4, characterized in that, The step of adjusting the power consumption of the main control chip based on the load change trend prediction result and the sleep control strategy information, and determining the power consumption adjustment strategy information, includes: adjusting the operating frequency and power supply status of the main control chip based on the load change trend prediction result, the sleep control strategy information, and the predefined load change trend standard result, and determining the operating frequency information and power supply status information; and obtaining the power consumption adjustment strategy information based on the operating frequency information and the power supply status information.

7. The method as described in claim 1, characterized in that, After the step of controlling the system to operate normally based on the sleep control strategy information and the power consumption adjustment strategy information, the method further includes: monitoring the corresponding actual load changes based on the sleep control strategy information and the power consumption adjustment strategy information to determine the actual load change monitoring information; when the actual load change monitoring information exceeds a preset threshold or a predefined external wake-up command is received, waking up multiple predefined functional modules from the sleep state in stages according to a predefined control strategy to determine the wake-up control strategy information; adjusting the operating frequency and power supply status of the main control chip based on the wake-up control strategy information and the power consumption adjustment strategy information to determine the target power consumption adjustment strategy information; and controlling the system to operate normally based on the wake-up control strategy information and the target power consumption adjustment strategy information.

8. A dynamic power consumption regulation device based on load prediction and hierarchical sleep mode, characterized in that, The device includes: an acquisition module for acquiring processor occupancy information, sensor connection information, data acquisition frequency information, communication status information, and data volume information to be processed; a processing module for inputting a predefined load prediction model based on the processor occupancy information, sensor connection information, data acquisition frequency information, communication status information, and data volume information to be processed to predict the load change trend within a predefined time period, and determining the load change trend prediction result; the processing module is also used to generate corresponding sleep control strategy information and power consumption adjustment strategy information based on the load change trend prediction result; and an execution module for controlling the system to operate normally based on the sleep control strategy information and the power consumption adjustment strategy information.

9. A dynamic power consumption regulation device based on load prediction and hierarchical sleep mode, characterized in that, The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the dynamic power consumption regulation method based on load prediction and hierarchical sleep as described in any one of claims 1 to 7.

10. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the dynamic power consumption adjustment method based on load prediction and hierarchical sleep as described in any one of claims 1 to 7.

Images