Power status monitoring method and system for protocol signal processing module

Abstract

The present invention discloses a power status monitoring method and system for a protocol signal processing module. By realizing real-time analysis of power consumption fluctuations of the signal processing module, subtle power consumption changes are captured, and changes in power demand are accurately predicted, the power distribution efficiency of the power system is optimized and adjusted to adapt to various load scenarios. The system quickly responds to instantaneous fluctuations that occur during signal processing, improves the power management efficiency and system stability of the power system, and thus ensures accurate and efficient power distribution in adaptive control scenarios, avoiding signal processing interruptions or equipment overloads caused by sudden load changes.

Description

Technical Field

[0001] The present invention relates to the technical field of power management, and in particular to a method and system for monitoring the power status of a protocol signal processing module. Background Art

[0002] In modern communications and data processing, power management technology is crucial for ensuring the stable operation of signal processing modules. As protocol signal processing requirements become increasingly complex, the efficiency of power supply and distribution directly impacts system performance and reliability. Ensuring efficient use of power resources in dynamically changing operating environments has become a core issue in this field.

[0003] However, existing power management methods often lack adaptability when addressing the real-time demands of signal processing modules. Many solutions rely heavily on pre-set power supply modes, making it difficult to quickly respond to transient fluctuations in signal processing. This is especially true when dealing with diverse loads such as complex protocol signals, as they lack the ability to accurately detect subtle changes in power consumption. This limitation leads to uneven resource allocation and can even cause system instability.

[0004] Looking deeper, the core challenge facing this field lies in achieving real-time analysis and prediction of power consumption fluctuations in signal processing modules. Due to the dynamic nature of signal processing tasks, power consumption fluctuations are highly uncertain, making traditional static power supply strategies unsatisfactory. This uncertainty further complicates capturing subtle power consumption changes, as these fluctuations are often hidden within the overall load. Failure to accurately identify and quantify these fluctuations can lead to biased power distribution, impacting overall system performance.

[0005] Therefore, how to analyze the power consumption fluctuations of the signal processing module in real time, accurately predict changes in power demand, and capture tiny changes in power consumption with the help of optimization methods has become a key issue in improving power management efficiency and system stability. Summary of the Invention

[0006] The present invention provides a power status monitoring method and system for a protocol signal processing module, so as to realize real-time analysis of power consumption fluctuations of the signal processing module, capture subtle power consumption changes, accurately predict changes in power demand, optimize and adjust the power distribution efficiency of the power system to adapt to various load scenarios, quickly respond to instantaneous fluctuations that occur during signal processing, and improve the power management efficiency and system stability of the power system.

[0007] The present invention provides a power status monitoring method for a protocol signal processing module, which is applied to an electric power system and includes: collecting operating data of a signal processing module associated with the electric power system to obtain a power consumption fluctuation data set, wherein the power consumption fluctuation data set is used to extract small fluctuation characteristics corresponding to instantaneous fluctuations in power consumption; inputting the small fluctuation characteristics in the power consumption fluctuation data set into a dynamic power status prediction model of the electric power system to obtain a power status prediction result that is suitable for a variety of load scenarios, wherein the power status prediction result is used to schedule and adjust the power resources of the electric power system, and the dynamic power status prediction model is obtained by training based on a preset small fluctuation data set, and the small fluctuation characteristics are input into a dynamic power status prediction model of the electric power system to obtain a power status prediction result that is suitable for a variety of load scenarios, wherein the power status prediction result is used to schedule and adjust the power resources of the electric power system, and the dynamic power status prediction model is obtained by training based on a preset small fluctuation data set. The fluctuation data set is the historical data of the small fluctuation characteristics; the power supply status prediction result is input into the power system, the power supply allocation efficiency of the power system is optimized and adjusted, and the target power supply mode and target resource allocation strategy required by the power system under the current operating state are determined; based on the target resource allocation strategy, the power system is dynamically switched to adapt to the target power supply mode, and feedback data of the power system operation is obtained; based on the feedback data, the effect of resource allocation balance of the power system is evaluated, and the effect evaluation result of resource allocation balance is determined; based on the effect evaluation result, the monitoring mechanism with real-time analysis capability is updated to determine the final resource allocation strategy.

[0008] According to a power status monitoring method for a protocol signal processing module provided by the present invention, the method for determining the small fluctuation characteristics includes: based on the power consumption fluctuation data set, using a time series analysis algorithm to extract characteristics of dynamic load changes, and determining a load change analysis result of the load change, wherein the load change analysis result includes a periodic pattern, a random fluctuation pattern and a trend development pattern; based on the load change analysis result, determining a key time point corresponding to an instantaneous fluctuation response, wherein the key time point is determined by the maximum value of a weighted second-order derivative of the load change analysis result; based on the key time point, determining a data record of the instantaneous fluctuation response; based on the data record of the instantaneous fluctuation response, starting a refined analysis module for capturing small changes, and extracting small fluctuation characteristics from the data record.

[0009] According to a power status monitoring method for a protocol signal processing module provided by the present invention, the data record based on the instantaneous fluctuation response is started to analyze the refined changes and extract the small fluctuation characteristics from the data record, including: determining the characteristic data of the small fluctuations in the data record through the refined analysis module for capturing small changes; based on the characteristic data, using a statistical analysis method to calculate the distribution of small changes and determine the distribution law of small changes; inputting the distribution law into a preset fluctuation model to determine the degree of influence associated with the distribution law; and determining the small fluctuation characteristics based on the numerical range of the degree of influence; wherein the preset fluctuation model is trained based on historical data of the distribution law of small changes.

[0010] According to a power status monitoring method for a protocol signal processing module provided by the present invention, the method adopts a statistical analysis method based on characteristic data to calculate the distribution of small changes and determine the distribution law of small changes, including: based on the characteristic data, adopting a statistical analysis method to determine the change amount between any two adjacent time points in the characteristic data and the standard deviation of the change amount; based on the change amount and the standard deviation, calculating the probability distribution of small changes and determining the distribution law.

[0011] According to a power status monitoring method for a protocol signal processing module provided by the present invention, the method of determining the slight fluctuation characteristics based on the numerical range of the impact degree includes: based on the numerical range, using a support vector machine algorithm to classify the impact degree to obtain an impact level; if the impact level reaches a preset attention level, performing an in-depth analysis of the impact level to determine potential fluctuation trend characteristics, wherein the fluctuation trend characteristics are used to indicate the possible development direction of subsequent fluctuations; and determining the slight fluctuation characteristics based on the fluctuation trend characteristics.

[0012] According to a power status monitoring method for a protocol signal processing module provided by the present invention, the refined analysis module for capturing small changes determines characteristic data of small fluctuations in the data record, including: determining the fluctuation amplitude of the instantaneous fluctuation response from the data record using a preset fluctuation amplitude calculation algorithm; if the fluctuation amplitude exceeds a preset threshold range, extracting small fluctuation characteristics from the data record using the refined analysis module for capturing small changes.

[0013] According to a power status monitoring method for a protocol signal processing module provided by the present invention, based on the power consumption fluctuation data set, a time series analysis algorithm is used to extract features of dynamic load changes to determine the load change analysis results of the load changes, including: based on the power consumption fluctuation data set, a time series analysis algorithm is used to extract features of dynamic load changes to separate trend components, periodic components and residual components corresponding to the load change rules; based on the periodic components, the frequency characteristics of the periodic components are extracted, and the periodic components corresponding to the frequency characteristics that are higher than a preset threshold are classified as periodic patterns; based on the residual components, the statistical distribution of the residual components is determined, and the residual components corresponding to the statistical distribution that deviate from the normal distribution are marked as random fluctuation patterns; and the trend components are smoothed using a sliding window technology to obtain a trend development pattern of the load changes.

[0014] The present invention also provides a power status monitoring system for a protocol signal processing module, comprising: an acquisition module for acquiring operating data of a signal processing module associated with an electric power system to obtain a power consumption fluctuation data set, wherein the power consumption fluctuation data set is used to extract small fluctuation characteristics corresponding to instantaneous fluctuations in power consumption; a prediction module for inputting the small fluctuation characteristics in the power consumption fluctuation data set into a dynamic power status prediction model of the electric power system to obtain a power status prediction result that is adaptable to a variety of load scenarios, wherein the power status prediction result is used to schedule and adjust the electric power resources of the electric power system, and the dynamic power status prediction model is obtained by training based on a preset small fluctuation data set, and the small fluctuation data set is the small fluctuation characteristics. an adjustment module for inputting the power state prediction result into the power system, optimizing and adjusting the power distribution efficiency of the power system, and determining the target power supply mode and target resource allocation strategy required by the power system under the current operating state; a switching module for dynamically switching the target power supply mode to the power system based on the target resource allocation strategy, and obtaining feedback data of the power system operation; an evaluation module for evaluating the effect of resource allocation balance of the power system based on the feedback data, and determining an evaluation result of the effect of resource allocation balance; an updating module for updating the monitoring mechanism of real-time analysis capability based on the effect evaluation result, and determining a final resource allocation strategy.

[0015] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a power status monitoring method of the protocol signal processing module as described above is implemented.

[0016] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the power status monitoring method of the protocol signal processing module described in any one of the above is implemented.

[0017] The present invention provides a power status monitoring method and system for a protocol signal processing module, comprising: collecting operating data of a signal processing module associated with the power system to obtain a power consumption fluctuation data set, wherein the power consumption fluctuation data set is used to extract small fluctuation characteristics corresponding to instantaneous fluctuations in power consumption; inputting the small fluctuation characteristics in the power consumption fluctuation data set into a dynamic power status prediction model of the power system to obtain a power status prediction result that is adaptable to a variety of load scenarios, wherein the power status prediction result is used to schedule and adjust the power resources of the power system, and the dynamic power status prediction model is obtained by training based on a preset small fluctuation data set, and the small fluctuation data set is The present invention analyzes the power consumption fluctuations of the signal processing module in real time, captures subtle power consumption changes, accurately predicts changes in power demand, optimizes and adjusts the power distribution efficiency of the power system, adapts it to various load scenarios, quickly responds to instantaneous fluctuations that occur during signal processing, improves the power management efficiency and system stability of the power system, ensures that power distribution is accurate and efficient under adaptive control scenarios, and avoids signal processing interruptions or equipment overloads caused by sudden load changes. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is one of the flow charts of the power status monitoring method of the protocol signal processing module provided by an embodiment of the present invention;

[0019] Figure 2 This is a second flow chart of a method for monitoring the power status of a protocol signal processing module provided by an embodiment of the present invention;

[0020] Figure 3 This is a third flow chart of a method for monitoring the power status of a protocol signal processing module provided by an embodiment of the present invention;

[0021] Figure 4This is a fourth flow chart of a method for monitoring the power status of a protocol signal processing module provided by an embodiment of the present invention;

[0022] Figure 5 This is a fifth flow chart of a method for monitoring the power status of a protocol signal processing module provided by an embodiment of the present invention;

[0023] Figure 6 This is a sixth flow chart of a method for monitoring the power status of a protocol signal processing module provided by an embodiment of the present invention;

[0024] Figure 7 This is the seventh flow chart of the power status monitoring method of the protocol signal processing module provided by an embodiment of the present invention;

[0025] Figure 8 It is a structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0026] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0027] Reference Figure 1 , an embodiment of the present invention provides a power status monitoring method for a protocol signal processing module, comprising the following steps: Step 100, collecting the operating data of the signal processing module associated with the power system to obtain a power consumption fluctuation data set, wherein the power consumption fluctuation data set is used to extract the small fluctuation characteristics corresponding to the instantaneous fluctuation of power consumption; by continuously collecting the operating data of the signal processing module, current state information related to the power consumption fluctuation is obtained, and a fluctuation characteristic data set is determined. According to the collected fluctuation characteristic data set, the load changes in different time periods are segmented and processed to obtain a classified load change data group. The classified load change data group is screened using a preset threshold. If the load change in a certain time period exceeds the threshold range, it is marked as an abnormal fluctuation data segment, and the distribution of the abnormal fluctuation data segment is obtained. By performing a time series analysis on the distribution of the abnormal fluctuation data segment, the K-means clustering algorithm is used to divide the abnormal fluctuation data segment into patterns, and the power consumption fluctuation data sets under different modes are determined.

[0028] For example, by collecting operational data from the signal processing module, a high-precision sensor is first used to collect the module's voltage and current data once per second. For example, within a certain time period, a voltage of 3.3V and a current of 0.5A are recorded, and the instantaneous power consumption is calculated to be 1.65W. This data is then transmitted in real time to a cloud database via an embedded system, forming a power consumption history record with a timestamp. Simultaneously, based on the current state information, the system automatically extracts the ambient temperature (e.g., 25°C) and operating frequency (e.g., 100MHz) during the operation of the signal processing module and associates this with the power consumption data for subsequent analysis of the impact of the environment on power consumption. Next, the system uses a time-window-based classification algorithm to analyze load variations within different time periods, segmenting the data into hourly windows. For example, in the peak load data from 8:00 AM to 9:00 AM on weekdays, power consumption fluctuates between 1.5W and 2.0W, while power consumption during non-working hours from 10:00 PM to 11:00 PM remains stable at around 0.8W. Using a K-means clustering algorithm, these load variations are categorized as high, medium, and low load, and stored in separate data tables for easy query and analysis. During this process, the system also incorporates business-related data, such as the task type of the signal processing module (for example, power consumption for voice processing tasks fluctuates significantly, with an average of 1.8W). By cross-analyzing this with the load classification results, it was found that high-load periods are often associated with complex tasks, thereby optimizing the dataset structure. Finally, the power consumption fluctuation dataset obtained contains multi-dimensional information such as time, power consumption value, load category and task type. For example, a record shows that at 9:00, the power consumption was 1.9W, the load category was high load, and the task type was video processing. Through statistical analysis, the system found that the average power consumption under high load was 1.85W and the standard deviation was 0.15, indicating that the fluctuation was relatively concentrated, providing a reliable basis for the construction of the subsequent power consumption prediction model.

[0029] Step 200: Input the minor fluctuation characteristics from the power consumption fluctuation dataset into a dynamic power supply state prediction model of the power system to obtain a power supply state prediction result that is adaptable to multiple load scenarios. The power supply state prediction result is used to schedule and adjust power resources in the power system. The dynamic power supply state prediction model is trained based on a preset minor fluctuation dataset, which is historical data of the minor fluctuation characteristics. In this embodiment, the dynamic power supply state prediction model is trained based on the preset minor fluctuation dataset. The training method of the dynamic power supply state prediction model is obtained by the following steps: obtaining historical data corresponding to minor fluctuation characteristics as a preset minor fluctuation dataset; inputting the minor fluctuation dataset into a preset initial model based on the minor fluctuation characteristics; and training the initial model using a random forest algorithm to obtain an initial prediction model. The initial prediction model analyzes the load diversity characteristics and extracts multi-scenario load samples from the preset minor fluctuation dataset to obtain a diverse load feature set. Then, using the diverse load feature set, the model parameter configuration of the initial prediction model is optimized using a gradient boosting decision tree algorithm to determine the optimized initial prediction model. If the prediction accuracy of the optimized initial prediction model on the validation set of the small fluctuation data set is lower than a preset threshold, the learning rate and tree depth of the optimized initial prediction model are adjusted, and the model is retrained to finally obtain a dynamic power state prediction model.

[0030] Therefore, after obtaining the power consumption fluctuation data set, small fluctuation features are extracted from the power consumption fluctuation data set. Then, the small fluctuation features corresponding to the power consumption fluctuation data set are input into the dynamic power supply state prediction model to predict the power supply state of the power system for various load scenarios to which the small fluctuation features are applicable, and obtain the power supply state prediction results of various load scenarios output by the dynamic power supply state prediction model, wherein the power supply state prediction results are used to schedule and adjust the power resources of the power system.

[0031] For example, in analyzing load distribution data with minor changes, we first collect load data. Assume the dataset contains 1,000 load samples, covering low load (0-20% CPU utilization), medium load (20-60%), and high load (60-100%) scenarios. The data is stored as a time series with a sampling frequency of once per second. We classify the load data using the K-means clustering algorithm (K=3), setting the initial cluster center to a random seed, the number of iterations to 100, and the convergence threshold to 0.001. We obtain statistical characteristics for the three load distribution categories, such as means of 15%, 45%, and 80%, and standard deviations of 5%, 10%, and 8%, respectively. Based on this, we adjust the parameters of the prediction model, selecting a long short-term memory (LSTM) network as the prediction model, with settings for the number of hidden units to 128, a learning rate to 0.001, a batch size to 32, and a training epoch to 50. To address the challenge of workload diversity, data augmentation techniques were used to generate a dataset with 10% noise perturbation (approximately 100 new samples). Combined with transfer learning, the model used the weights of a pretrained model in a low-load scenario to initialize prediction errors in high-load scenarios. During training, mean squared error (MSE) was used as the loss function, the Adam optimizer was used, and the validation set accounted for 20%. The final model achieved an MSE of 0.015 on the test set, a 40% improvement over the baseline model (MSE = 0.025). To adapt to diverse workloads, an adaptive learning rate adjustment strategy was introduced. If the validation set loss did not decrease for five consecutive rounds, the learning rate was reduced by 0.5 times, with a minimum of 0.0001. Prediction results showed that the model achieved prediction accuracy of 92% in high-load scenarios, 95% in medium-load scenarios, and 90% in low-load scenarios. Comparative analysis showed that the model was most adaptable to medium-load scenarios due to their more concentrated data distribution. In business scenarios such as cloud computing resource allocation, the prediction results can be used to dynamically adjust virtual machine resources, reducing resource waste by 5% and improving system stability.

[0032] Step 300: Input the power state prediction results into the power system, optimize and adjust the power system's power allocation efficiency, and determine the target power supply mode and target resource allocation strategy required for the power system's current operating state. After obtaining the power state prediction results, perform a preliminary analysis of the power system's dynamic load changes and signal processing requirements to obtain a characteristic distribution of load changes. Based on the characteristic distribution of load changes, a matching analysis is performed using a pre-established load model to determine the target power supply mode currently adapted for the power system. If the currently adapted target power supply mode deviates from the historical mode in historical data, the resource allocation ratio is adjusted through calculation and analysis to obtain a preliminary resource allocation plan. Based on the preliminary resource allocation plan, combined with real-time data on signal processing requirements, a support vector machine algorithm is used to optimize and adjust the resource allocation to determine the feasibility of the plan. If the feasibility of the resource allocation plan is verified, a secondary calibration is performed based on the dynamic load change trend of the power system to obtain adjusted resource allocation parameters. Using the adjusted resource allocation parameters, a final power supply mode configuration is generated to determine the target resource allocation strategy for the power system.

[0033] For example, in the process of optimizing power distribution efficiency, dynamic load changes are first analyzed using a predictive model. Assuming that current system operating data indicates that the load will increase from 50% to 80% within the next 10 minutes, the prediction algorithm uses time series analysis, utilizing the past hour's load data (sampled once per minute, for a total of 60 data points) to predict future load trends using an ARIMA model. The calculated load growth rate is 3.5% per minute, with a prediction error of ±5%. Next, considering signal processing requirements, the power consumption requirements of the current signal processing module are analyzed. Assuming the signal processing task requires processing 1,000 data packets per second, with a single packet processing power consumption of 0.2 watts, the total power consumption is calculated to be 200 watts. The current power supply mode is a fixed output of 300 watts, resulting in an efficiency of only 66.7%. Then, based on the predicted results and power consumption requirements, the system calculates the current required power supply mode and uses the Dynamic Voltage and Frequency Scaling (DVFS) algorithm to adjust the supply voltage from 12V to 10V and the frequency from 1.2GHz to 1.0GHz, reducing power consumption to 220 watts and increasing efficiency to 90.9%. This ensures that signal processing latency does not exceed 2 milliseconds (the current latency is 1.8 milliseconds, meeting the requirement). Finally, the system determines the optimal resource allocation plan. Using a resource scheduling algorithm, the system allocates the excess 80 watts of power to the backup battery charging module, achieving 85% charging efficiency and an estimated storage capacity of 680 watt-hours within 10 minutes. The system also monitors load changes. If the load exceeds the predicted range by ±5%, a real-time adjustment mechanism is triggered, recalculating power supply parameters to ensure system stability.

[0034] Step 400: Based on the target resource allocation strategy, dynamically switch the power system to the target power supply mode and obtain feedback data on the power system's operation. Based on the target power supply mode, adjust the power supply mode of the signal processing module to the target power supply mode. While operating in the target power supply mode, control the power supply module to operate in accordance with the target resource allocation strategy. While operating in accordance with the target resource allocation strategy, obtain current operating status data as system operation feedback data. By comparing the operating status data before and after the adjustment, analyze the impact of the power supply mode switch on system operational stability and determine whether the switching solution meets stability requirements.

[0035] For example, in implementing adaptive dynamic switching of power supply modes to ensure stable system operation, the optimal resource allocation plan is first developed by real-time monitoring of system load data. Assuming a total system load of 1000 watts, divided into three sub-modules: 400 watts, 300 watts, and 300 watts, a load balancing algorithm is used to adjust the resource allocation ratios to 40%, 30%, and 30%. By calculating the deviation between each module's actual power consumption and the target power consumption (for example, the deviation for module one is 400-400 = 0 watts), the allocation error is ensured to be within ±5%. The system automatically updates the allocation ratio every 5 seconds based on load changes. Next, dynamic power supply mode switching is implemented, using a preset power threshold (e.g., 800 watts) as the switching condition. When the total load exceeds the threshold, the system automatically switches from energy-saving mode to high-performance mode. The switching process is completed through software-controlled power management units, with the switching time controlled within 100 milliseconds. Voltage fluctuation data before and after the switch (e.g., from 12.0V to 12.2V) is also recorded. Next, the system monitors power consumption fluctuations in real time, collecting data every second. For example, if a particular monitoring session reveals a power consumption fluctuation range of ±50 watts, Fourier transform analysis of the fluctuation frequency reveals a primary fluctuation period of 2 seconds. This determines whether this is an abnormal fluctuation (an alarm is triggered if the frequency exceeds a preset value of 0.8 Hz). Finally, system operation feedback data is obtained, using sensors to collect operating temperature (assuming 45 degrees Celsius) and current (assuming 10 amps). Combined with historical data analysis (for example, the average temperature over the past hour was 40 degrees Celsius), the system determines whether it is stable. If the temperature deviation exceeds 5 degrees Celsius, the cooling mechanism is triggered, automatically adjusting the fan speed to 3000 rpm to maintain system stability.

[0036] Step 500, based on the feedback data, evaluate the effect of resource allocation balance of the power system and determine the effect evaluation result of resource allocation balance; collect real-time feedback data through the system, adopt the time series analysis method to calculate the balance index of resource allocation, and obtain the effect evaluation result of distribution balance, which is used as the distribution balance index of power resources. Based on the analysis results, this embodiment adjusts the prediction model parameters to adapt to diversified load scenarios, optimizes power distribution efficiency, and can also realize dynamic switching of power supply modes, and evaluates the resource allocation balance effect through feedback data, and continuously improves the allocation strategy. Ultimately, a cyclically optimized power management configuration is formed to effectively respond to the continuous changes in signal processing requirements and improve system operation stability and energy utilization efficiency.

[0037] Step 600: Based on the effect evaluation result, update the monitoring mechanism of the real-time analysis capability and determine the final resource allocation strategy.

[0038] If the indicator values ​​in the effect evaluation results fall below a preset threshold, a deviation analysis algorithm is used to identify deviation patterns in resource allocation and determine the deviation distribution. Based on the deviation distribution, a support vector machine algorithm is used to classify the instantaneous fluctuation types and obtain a set of fluctuation characteristics. Using this set of fluctuation characteristics, the dynamic response model of the control parameters is adjusted, and the optimized parameter values ​​are calculated to generate the control parameters. The control parameters are then used to update the resource allocation strategy, calculate resource utilization and allocation balance indicators, and obtain an improved resource allocation strategy. Furthermore, the improved resource allocation strategy can be used to monitor system feedback data in real time and test the accuracy of the instantaneous fluctuation response to determine the effectiveness of the strategy optimization. If the strategy optimization effect does not reach the preset threshold, the dynamic adjustment mechanism is iteratively updated, the control parameters are recalculated, and the final resource allocation strategy is determined.

[0039] For example, when evaluating resource allocation balance, system operational feedback data can be analyzed. For example, consider collecting CPU usage data for a server cluster over the past 24 hours. Suppose the cluster has 10 servers, with an average CPU usage of 60% per server. However, two servers have a usage as high as 85%, while the other two have a usage of only 30%, with a standard deviation of 20.5. This indicates significant resource imbalance. Based on this information, the system automatically runs a balance assessment algorithm to calculate the variance of resource allocation and, based on a preset threshold (for example, a standard deviation less than 10), determines whether the current allocation is reasonable. If the analysis indicates that the variance exceeds the threshold, adjustments are required. To address potential deviations, the system automatically identifies transient fluctuations. For example, if a server's CPU usage spikes from 60% to 90% due to a traffic burst, the system uses a time series analysis algorithm to calculate the volatility (assuming the volatility threshold is 15%). If this exceeds the threshold, a response mechanism is triggered to adjust control parameters, such as increasing the resource scheduling priority from the default value of 0.5 to 0.8 to speed up resource allocation. The improved allocation strategy was validated through simulation tests. For example, after reallocating resources, the standard deviation of CPU utilization across 10 servers dropped to 8.2, keeping volatility below 10%. The system automatically records and compares the data before and after the adjustment, generating an optimization report showing an improvement of approximately 60% in balance. To ensure a rigorous logical chain, if resource allocation involves multiple business scenarios, business priority weights can be introduced. For example, if a business accounts for 40% of peak traffic, the system will automatically allocate an additional 20% of backup resources to it, ensuring a more accurate response to fluctuations. The overall strategy is continuously iterated and optimized, driven by data.

[0040] In another embodiment, according to the improved resource allocation strategy, a corresponding adjustment strategy template is obtained to determine the priority order of adaptation. According to the priority order, combined with the current power management state, a dynamic adjustment process is executed. If the adjusted state does not meet the preset standard, the analysis process is executed cyclically to obtain new adjustment parameters. According to the new adjustment parameters, the power management configuration scheme is updated, the configuration effect is verified by simulating the operating environment, and the final configuration data is determined. According to the final configuration data, the resource allocation strategy is continuously optimized, and the support vector machine model in machine learning is used to predict long-term demand changes to obtain prediction trend results. According to the prediction trend results, the sensitivity parameters of the monitoring mechanism are adjusted, and the above process is executed cyclically until the system runs stably, and the resource allocation strategy corresponding to the stable operation of the system is used as the final resource allocation strategy.

[0041] For example, the improved allocation strategy first uses a real-time data acquisition system to acquire the signal processing module's power consumption data and processing requirements. Assuming the current signal processing module consumes 50 watts per second and requires 1,000 packets per second, the built-in load analysis algorithm calculates the current power consumption to processing requirement ratio to be 0.05 watts per packet. Combined with historical data analysis, this ratio is found to be higher than the average of 0.04 watts per packet. The system automatically triggers the optimized allocation strategy, adjusting the power allocation priority to high-performance mode and increasing the supply voltage by 10% to 5.5 volts to improve processing efficiency. Subsequently, the monitoring mechanism with real-time analysis capabilities is updated, using embedded sensors to collect core module temperature and current data once per second. Assuming a temperature of 65 degrees Celsius and a current of 2.5 amps, the preset thermal management algorithm calculates a thermal load index of 0.8 (below the safety threshold of 1.0). The system determines that frequency reduction is not necessary and continues to maintain the current configuration. To address the constant fluctuations in signal processing demands, the system initiates a dynamic adjustment cycle, executing a demand forecasting model every five minutes. Based on the fluctuations in packet processing volume over the past hour (assuming a range of 800 to 1200 packets per second), a linear regression algorithm is used to predict that the processing demand for the next ten minutes will be 1100 packets per second. This prediction then adjusts the power management strategy, pre-allocating the power supply to 55 watts to accommodate potential peaks. When finalizing the power management configuration, the system integrates these analysis results and uses a weighted calculation (60% for performance and 40% for energy efficiency) to determine the optimal configuration, setting the supply voltage to 5.4 volts and the power cap to 58 watts. This adjustment is then logged in a database for future optimization.

[0042] The present invention provides a power status monitoring method and system for a protocol signal processing module, comprising: collecting operating data of a signal processing module associated with the power system to obtain a power consumption fluctuation data set, wherein the power consumption fluctuation data set is used to extract small fluctuation characteristics corresponding to instantaneous fluctuations in power consumption; inputting the small fluctuation characteristics in the power consumption fluctuation data set into a dynamic power status prediction model of the power system to obtain a power status prediction result that is adaptable to a variety of load scenarios, wherein the power status prediction result is used to schedule and adjust the power resources of the power system, and the dynamic power status prediction model is obtained by training based on a preset small fluctuation data set, and the small fluctuation data set is The present invention analyzes the power consumption fluctuations of the signal processing module in real time, captures subtle power consumption changes, accurately predicts changes in power demand, optimizes and adjusts the power distribution efficiency of the power system, adapts it to various load scenarios, quickly responds to instantaneous fluctuations that occur during signal processing, improves the power management efficiency and system stability of the power system, ensures that power distribution is accurate and efficient under adaptive control scenarios, and avoids signal processing interruptions or equipment overloads caused by sudden load changes.

[0043] In one embodiment, see Figure 2The method for determining the characteristics of minor fluctuations includes: Step 101: Based on the power consumption fluctuation dataset, a time series analysis algorithm is used to extract features of dynamic load changes, and a load change analysis result of the load change is determined, wherein the load change analysis result includes a periodic pattern, a random fluctuation pattern, and a trend development pattern; by continuously collecting load changes, a dynamic data stream is obtained from the system, and a time series analysis method is used to obtain a load change analysis result of the load change, wherein the load change analysis result is a change pattern of power consumption fluctuation, which includes a periodic pattern, a random fluctuation pattern, and a trend development pattern. For example, the power consumption fluctuation pattern represented by the load change analysis result can be an annual or monthly growth rate of a power indicator, or an increase specific to a certain day of the week, or a random fluctuation pattern. Subsequently, based on the load change analysis result of the load change, a real-time monitoring technology is used to dynamically track signal processing requirements and obtain data records of instantaneous fluctuations, wherein the data records represent data that fluctuates in power consumption fluctuation data and medium power consumption data within a certain time period.

[0044] For example, by analyzing load variation trends, a monitoring mechanism with real-time analysis capabilities is constructed. This begins with data collection. Assume that the system collects power consumption fluctuation data every second. Data points include CPU utilization (percentage) and memory usage (GB). For example, at a certain moment, the CPU utilization is 75.5% and the memory usage is 8.2GB. Using a sliding window algorithm with a window size of 60 seconds, the mean and standard deviation of the power consumption fluctuation data within the window are calculated. For example, the mean CPU utilization over the past 60 seconds is 70.2%, and the standard deviation is 3.5%. To capture transient fluctuations, an exponential moving average (EMA) algorithm is used with a smoothing factor α set to 0.3. The EMA value of the current load is calculated. For example, the current CPU utilization EMA is 72.8%. The CPU utilization, standard deviation, and EMA values ​​are the results of the load variation analysis and represent the changing trends of the signal processing module.

[0045] Step 102: Based on the load change analysis results, determine the key time point corresponding to the instantaneous fluctuation response, wherein the key time point is determined by the maximum value of the weighted second-order derivative of the load change analysis results; through continuous collection of load changes, obtain dynamic data streams from the system, and apply time series analysis methods to obtain trend characteristics of load changes. Based on the trend characteristics of load changes, use real-time monitoring technology to dynamically track signal processing requirements and determine the key time points of demand changes. Specifically, the key time point corresponding to the instantaneous fluctuation response is calculated by the maximum value of the weighted second-order derivative of the load change analysis results. The principle is to identify the key time points of load fluctuations by finding the maximum value of the weighted second-order derivative of the load change analysis results.

[0046] Step 103 determines the data records of the instantaneous fluctuation response based on the key time points. At the key time points of the demand change, the data records of the instantaneous fluctuations are obtained and, using a pre-established fluctuation amplitude calculation model, a determination is made as to whether the fluctuations exceed a preset threshold. If the data records of the instantaneous fluctuations exceed the preset threshold, data filtering techniques are used to remove noise from the abnormal fluctuations, resulting in processed fluctuation data as the data records of the instantaneous fluctuation response.

[0047] Step 104 : Based on the data record of the instantaneous fluctuation response, a detailed analysis module for capturing small changes is started to extract small fluctuation features from the data record.

[0048] After extracting the data records corresponding to the transient fluctuation response, the transient fluctuation information from the data records representing the transient fluctuation response is collected. Specifically, a pre-established threshold range can be used to compare the transient fluctuation information with the threshold range to determine whether the transient fluctuation information exceeds the preset threshold range, thereby obtaining a preliminary fluctuation determination result. If the fluctuation determination result indicates that the transient fluctuation exceeds the preset threshold range, the refined analysis module is activated. Through the refined analysis module, subtle changes are captured and processed, subtle fluctuation features are extracted from the data records of the transient fluctuation response, and specific characteristic information of the subtle changes is determined.

[0049] This embodiment accurately predicts changes in power demand by analyzing the power consumption fluctuations of the signal processing module in real time. Using optimization techniques, it precisely captures subtle power consumption changes and predicts power state results that adapt to various load scenarios. This allows for a rapid response to transient fluctuations that occur during signal processing, improving power management efficiency and system stability. By collecting a power consumption fluctuation dataset from operational data and extracting load variation analysis results using a time series analysis algorithm, the transient fluctuation response is extracted based on key time points in the load variation analysis results, generating data records corresponding to the transient fluctuation response. This improves the accuracy of the transient fluctuation response data extraction, thereby enhancing the accuracy of subsequent small fluctuation feature extraction.

[0050] In one embodiment, see Figure 3 The data record based on the instantaneous fluctuation response starts a refined analysis module for capturing small changes and extracts small fluctuation features from the data record, including:

[0051] Step 1041, determining characteristic data of small fluctuations in the data record through a refined analysis module for capturing small changes;

[0052] Step 1042: Based on the characteristic data, a statistical analysis method is used to calculate the distribution of small changes and determine the distribution pattern of small changes;

[0053] Step 1043: input the distribution law into a preset fluctuation model to determine the degree of influence associated with the distribution law;

[0054] Step 1044: determining the minor fluctuation characteristics based on the numerical range of the impact degree;

[0055] The preset fluctuation model is obtained by training based on historical data of the distribution law of small changes.

[0056] Based on the extracted data records of transient change responses, characteristic data of the small fluctuations contained therein is first determined. Based on the characteristic data of the small fluctuations, statistical analysis methods are used to calculate the distribution of the small fluctuations and distribution pattern data. The distribution pattern data is equivalent to the distribution pattern that characterizes the small fluctuations. The distribution pattern data is then used to determine whether the distribution conforms to a preset fluctuation model. If the distribution conforms to the preset fluctuation model, further data extraction is performed to obtain an impact parameter associated with the distribution. Based on the impact data associated with the distribution, a specific numerical range of the impact is determined. Based on the specific numerical range of the impact, a support vector machine algorithm is used to process the impact to determine the characteristics of the small fluctuations. The preset fluctuation model is a linear model. By inputting distribution parameters, distribution conditions, or distribution pattern parameters into the preset fluctuation model, the corresponding output is the impact associated with the distribution pattern. Specifically, based on the distribution of the small changes in the time series, the impact of each small fluctuation feature of the small change on the time series is predicted. The impact is the degree to which each small fluctuation feature of the small change is expressed in the time series, thereby accurately capturing the small fluctuation characteristics of small changes in the power consumption data. The preset volatility model is trained based on historical data of small change distribution patterns.

[0057] In this embodiment, the distribution of small changes is calculated based on the statistical analysis method, and the distribution law of small changes is determined, so that small fluctuation characteristics are extracted based on the distribution law of small changes. This method of extracting the distribution law of small changes and then extracting small fluctuation characteristics using the statistical analysis method can accurately capture subtle power consumption changes and improve the accuracy and screening efficiency of the screened small fluctuation characteristics.

[0058] In one embodiment, see Figure 4 The method of using statistical analysis methods based on characteristic data to calculate the distribution of small changes and determine the distribution law of small changes includes:

[0059] Step 10421: Based on the characteristic data, using a statistical analysis method, determine the amount of change between any two adjacent time points in the characteristic data and the standard deviation of the amount of change;

[0060] Step 10422: Calculate the probability distribution of small changes based on the change amount and the standard deviation, and determine the distribution law.

[0061] Based on the characteristic data of small fluctuations, a statistical analysis method is used to calculate the distribution of small changes. Specifically, based on the characteristic data of small fluctuations, a statistical analysis method is used to calculate the amount of change between any two adjacent time points in the characteristic data and the standard deviation of the amount of change. A preset threshold for small changes is obtained. Then, based on the above-mentioned amount of change and standard deviation, combined with the preset threshold for small changes, the probability distribution of small changes is calculated to determine the distribution law. This method determines the distribution law corresponding to the small fluctuation feature through the amount of change between adjacent time points in the small change characteristic data and the standard deviation of the amount of change, thereby accurately capturing subtle power consumption changes and improving the accuracy of small fluctuation feature recognition.

[0062] In this embodiment, the distribution law corresponding to the small fluctuation feature is determined by the change amount between adjacent time points in the small change feature data and the standard deviation of the change amount, thereby accurately capturing subtle power consumption changes and improving the accuracy of small fluctuation feature recognition.

[0063] In one embodiment, see Figure 5 , the determining of the small fluctuation characteristics based on the numerical range of the impact degree includes:

[0064] Step 10441: Based on the numerical range, a support vector machine algorithm is used to classify the impact degree to obtain an impact level.

[0065] Step 10442: If the impact level reaches a preset concern level, perform an in-depth analysis of the impact level to determine potential fluctuation trend characteristics, wherein the fluctuation trend characteristics are used to indicate the possible development direction of subsequent fluctuations;

[0066] Step 10443: Determine the minute fluctuation feature based on the fluctuation trend feature.

[0067] Based on the extracted fluctuation characteristics, statistical analysis methods are used to calculate the distribution of small changes, obtain distribution pattern data, and determine whether the distribution conforms to the preset fluctuation model. If the distribution conforms to the preset fluctuation model, further data extraction is performed to obtain the impact degree parameters related to the distribution and determine the specific numerical range of the impact degree. Based on the specific numerical range of the impact degree, the support vector machine algorithm is used to classify the impact degree, obtain the classified impact level, and determine whether the impact level reaches the preset level of concern. If the impact level reaches the preset level of concern, in-depth analysis of the classified impact level data is performed to obtain potential fluctuation trend characteristics and determine the possible development direction of subsequent fluctuations. Through continuous monitoring of fluctuation trend characteristics, time series analysis methods are used to obtain dynamic data of trend changes, determine whether the fluctuation trend is stable, and obtain the final fluctuation monitoring results.

[0068] For example, when the transient fluctuation response exceeds a preset threshold, the system automatically activates the refined analysis module to extract features and assess the impact of minor changes. For example, in a power system, assume the transient voltage fluctuation threshold is ±5V. The currently collected data indicates a voltage fluctuation of 6.2V, exceeding the threshold and triggering refined analysis. First, the system uses a fast Fourier transform (FFT) algorithm to perform frequency domain decomposition on the collected voltage signal (sampling frequency 10kHz, window length 1024 points). This extracts minor fluctuation components in the frequency range of 0.1Hz to 100Hz, and determines the spectral energy distribution. A significant peak is detected at 2Hz, accounting for 12.3% of the energy. Next, the module uses a wavelet transform (using the Daubechies-4 wavelet basis with 5 decomposition levels) to perform time-frequency analysis on the signal, identifying the time domain distribution of the minor fluctuations. The calculated amplitude of the fluctuations has a mean of 0.8V and a variance of 0.15V², indicating that the fluctuations are concentrated within a specific time period (e.g., 0.2 to 0.3 seconds). The system then uses the principal component analysis (PCA) algorithm to reduce the dimensionality of the extracted features, retaining 95% of the variance contribution. This generates a principal component eigenvector, which is used to quantify the impact of fluctuations on system stability. Combined with historical data (voltage records from the past hour), a Kalman filter is used to predict the fluctuation trend over the next 10 seconds, estimating the impact to be a 2.5% increase in system load. A distribution curve is generated (normal distribution, mean 0.8V, standard deviation 0.39V). Finally, the system stores the analysis results in a database and pushes them to the monitoring system via an API for subsequent business optimization, such as adjusting load distribution to mitigate the impact of fluctuations.

[0069] In this embodiment, by determining the impact level that has reached a certain level of preset attention level, an in-depth analysis of the impact level is performed to further explore potential fluctuation trend characteristics, thereby extracting small fluctuation characteristics and further improving the accuracy of small fluctuation feature recognition.

[0070] In one embodiment, see Figure 6 The refined analysis module for capturing small changes determines characteristic data of small fluctuations in the data record, including:

[0071] Step 10411, determining the fluctuation amplitude of the instantaneous fluctuation response by using a preset fluctuation amplitude calculation algorithm to record the data of the instantaneous fluctuation response;

[0072] Step 10412: If the fluctuation amplitude exceeds a preset threshold range, a fine-grained analysis module for capturing small changes is used to extract small fluctuation features from the data record.

[0073] At key points in demand fluctuations, data records of instantaneous fluctuations are obtained, and a pre-established fluctuation amplitude calculation algorithm is used to determine whether the fluctuations exceed the preset threshold range. If the data records of instantaneous fluctuations exceed the preset threshold range, data filtering technology is used to denoise the abnormal fluctuations to obtain processed fluctuation data. Based on the processed fluctuation data, a support vector machine algorithm is used to classify the fluctuation amplitudes and determine the risk level of the fluctuation amplitude. Based on the risk level of the fluctuation amplitude and combined with range control logic, if the risk level exceeds the preset standard, a real-time alarm mechanism is triggered and the specific time point of the alarm trigger is obtained. At the specific time point of the alarm trigger, data backtracking technology is used to conduct an in-depth analysis of the historical records of load changes to determine whether there are potential trend anomalies.

[0074] Furthermore, the fluctuation amplitude calculation algorithm includes an absolute fluctuation amplitude calculation method, a relative fluctuation amplitude method, a windowed fluctuation amplitude or a multi-scale fluctuation amplitude analysis method, etc. In one embodiment, the absolute fluctuation amplitude calculation method is explained as an implementation method. The principle of the absolute fluctuation amplitude calculation method is to calculate the fluctuation amplitude of the instantaneous fluctuation response by the absolute value of the difference between adjacent time points in the data record of the instantaneous fluctuation.

[0075] This embodiment establishes a real-time monitoring mechanism to capture instantaneous fluctuation responses, and activates a small change analysis module when a threshold is exceeded to extract small fluctuation features, thereby accurately capturing subtle power consumption changes.

[0076] In one embodiment, see Figure 7 , based on the power consumption fluctuation data set, using a time series analysis algorithm to extract features of dynamic load changes and determine a load change analysis result of the load change, including:

[0077] Step 1011 , based on the power consumption fluctuation data set, a time series analysis algorithm is used to extract features of dynamic load changes, and a trend component, a periodic component, and a residual component corresponding to the load change pattern are separated;

[0078] Step 1012: extracting frequency features of the periodic components based on the periodic components, and classifying the periodic components corresponding to the frequency features that are higher than a preset threshold as periodic patterns;

[0079] Step 1013: determining a statistical distribution of the residual components based on the residual components, and marking the residual components corresponding to the statistical distribution that deviate from the normal distribution as random fluctuation patterns;

[0080] Step 1014: Smoothing the trend component using a sliding window technique to obtain a load change trend development model;

[0081] The load change analysis result includes the periodic pattern, the random fluctuation pattern and the trend development pattern.

[0082] A power consumption fluctuation dataset is obtained and preprocessed to remove noise and outliers, resulting in a cleaned dataset. A time series decomposition method is used to separate the load variation analysis results from the cleaned dataset, including trend components, periodic components, and residual components. Based on these load variation analysis results, load variation patterns are identified, including periodic patterns, random fluctuation patterns, and trend development patterns. Based on the load variation patterns, the frequency and amplitude characteristics of the periodic components are extracted to determine the significance of the periodic patterns. If the stability of the frequency characteristics exceeds a preset threshold, the pattern is classified as periodic. The statistical distribution of the residual components is analyzed to identify the fluctuation characteristics of the random patterns. If the distribution of the residual components deviates from a normal distribution, the pattern is labeled as a random fluctuation pattern. A sliding window technique is used to smooth the trend components and obtain the main trend curve of the load variation. Based on the main trend curve, the upper and lower limits of the fluctuation range are calculated to determine the trend development pattern of the dynamic load. By integrating the periodic pattern, random fluctuation pattern, and trend development pattern, a comprehensive characteristic description of the load variation is constructed, resulting in a complete load variation analysis result.

[0083] For example, consider a power consumption fluctuation dataset containing 1,440 data points of server power consumption (in watts) recorded every minute over a 24-hour period, ranging from 100 to 500 watts. First, smooth the time series using a moving average algorithm with a 5-minute window size to obtain a smoothed series to reduce noise.

[0084] For example, within a 5-minute window, the power consumption values ​​were 200, 210, 205, 208, and 215, respectively, resulting in a smoothed value of 207.6 watts. Next, feature extraction was performed based on the smoothed sequence, and the time series was decomposed using the Fast Fourier Transform (FFT) algorithm to identify periodic patterns. FFT analysis revealed that the main frequency component corresponded to a 60-minute period, indicating significant hourly load fluctuations with an amplitude of approximately 50 watts, reflecting the periodic nature of server task scheduling. To distinguish between periodicity and randomness, an autoregressive integrated moving average (ARIMA) model (p=1, d=1, q=1) was used to fit the time series. Residual analysis revealed a standard deviation of 15 watts for the random fluctuations, indicating that the randomness was primarily caused by external interference or non-periodic tasks. The main trend was extracted using a Hodrick-Prescott filter, and the decomposition of the trend components revealed an overall slow upward trend in power consumption, from an initial 200 watts to 230 watts at the end of the 24-hour period. The fluctuation range, determined by the standard deviation and the period amplitude, was ±65 watts. During the analysis process, periodic fluctuations are related to server scheduled tasks, and random fluctuations may be related to user access peaks. The load prediction model can be further optimized by combining business scenarios.

[0085] The power status monitoring system of the protocol signal processing module provided by the present invention is described below. The power status monitoring system of the protocol signal processing module described below and the power status monitoring method of the protocol signal processing module described above can be referred to each other.

[0086] The present invention provides a power status monitoring system for a protocol signal processing module, comprising: an acquisition module for acquiring operating data of a signal processing module associated with a power system to obtain a power consumption fluctuation data set, wherein the power consumption fluctuation data set is used to extract minute fluctuation features corresponding to instantaneous fluctuations in power consumption;

[0087] a prediction module, configured to input the minor fluctuation characteristics in the power consumption fluctuation dataset into a dynamic power supply state prediction model of the power system to obtain a power supply state prediction result that is adaptable to a variety of load scenarios, wherein the power supply state prediction result is used to schedule and adjust power resources of the power system, and the dynamic power supply state prediction model is trained based on a preset minor fluctuation dataset, which is historical data of the minor fluctuation characteristics;

[0088] an adjustment module, configured to input the power state prediction result into the power system, optimize and adjust the power distribution efficiency of the power system, and determine the target power supply mode and target resource allocation strategy required by the power system under the current operating state;

[0089] a switching module, configured to implement dynamic switching of the target power supply mode for the power system based on the target resource allocation strategy, and obtain feedback data on the operation of the power system;

[0090] An evaluation module, configured to evaluate the effect of resource allocation balancing of the power system based on the feedback data, and determine an evaluation result of the effect of resource allocation balancing;

[0091] The updating module is used to update the monitoring mechanism of the real-time analysis capability based on the effect evaluation result and determine the final resource allocation strategy.

[0092] Figure 8 An example of a physical structure diagram of an electronic device is shown below. Figure 8 As shown, the electronic device may include: a processor 810, a communication interface 820, a memory 830, and a communication bus 840, wherein the processor 810, the communication interface 820, and the memory 830 communicate with each other via the communication bus 840. The processor 810 may call the logic instructions in the memory 830 to execute the power status monitoring method of the protocol signal processing module.

[0093] Furthermore, the logic instructions in the aforementioned memory 830 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product, stored in a storage medium, includes instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, mobile hard drives, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0094] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the power status monitoring method of the protocol signal processing module provided by the above methods.

[0095] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0096] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0097] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for monitoring the power status of a protocol signal processing module, characterized in that: The method is applied to an electric power system and includes: collecting operating data of a signal processing module associated with the electric power system to obtain a power consumption fluctuation data set, wherein the power consumption fluctuation data set is used to extract small fluctuation characteristics corresponding to instantaneous fluctuations in power consumption; inputting the small fluctuation characteristics into a dynamic power supply state prediction model of the electric power system to obtain a power supply state prediction result that is adaptable to multiple load scenarios, wherein the power supply state prediction result is used to schedule and adjust power resources of the electric power system, wherein the dynamic power supply state prediction model is trained based on a preset small fluctuation data set, and the small fluctuation data set is historical data of the small fluctuation characteristics; inputting the power supply state prediction result into the electric power system, optimizing and adjusting the power distribution efficiency of the electric power system, and determining a target power supply mode and a target resource allocation strategy required for the electric power system under a current operating state; based on the target resource allocation strategy, implementing dynamic switching of the target power supply mode for the electric power system and obtaining feedback data on the operation of the electric power system; based on the feedback data, evaluating the effect of resource allocation balance of the electric power system and determining an evaluation result of the resource allocation balance effect; and based on the evaluation result, updating a monitoring mechanism with real-time analysis capabilities and determining a final resource allocation strategy.

2. The power status monitoring method of the protocol signal processing module according to claim 1, characterized in that: The method for determining the characteristics of the small fluctuations includes: based on the power consumption fluctuation data set, using a time series analysis algorithm to extract characteristics of dynamic load changes, and determining a load change analysis result of the load change, wherein the load change analysis result includes a periodic pattern, a random fluctuation pattern, and a trend development pattern; based on the load change analysis result, determining a key time point corresponding to the instantaneous fluctuation response, wherein the key time point is determined by the maximum value of the weighted second-order derivative of the load change analysis result; based on the key time point, determining the data record of the instantaneous fluctuation response; based on the data record of the instantaneous fluctuation response, starting a refined analysis module for capturing small changes, and extracting small fluctuation characteristics from the data record.

3. The power status monitoring method of the protocol signal processing module according to claim 2, characterized in that: The data record based on the instantaneous fluctuation response starts a refined analysis module for capturing small changes, and extracts small fluctuation features from the data record, including: determining characteristic data of small fluctuations in the data record through the refined analysis module for capturing small changes; based on the characteristic data, using a statistical analysis method to calculate the distribution of small changes and determine the distribution law of small changes; inputting the distribution law into a preset fluctuation model to determine the degree of influence associated with the distribution law; and determining the small fluctuation features based on the numerical range of the influence degree; wherein the preset fluctuation model is trained based on historical data of the distribution law of small changes.

4. The power status monitoring method of the protocol signal processing module according to claim 3, characterized in that: The method of using a statistical analysis method based on the characteristic data to calculate the distribution of small changes and determine the distribution law of small changes includes: using a statistical analysis method based on the characteristic data to determine the change between any two adjacent time points in the characteristic data and the standard deviation of the change; and calculating the probability distribution of small changes based on the change and the standard deviation to determine the distribution law.

5. The power status monitoring method of the protocol signal processing module according to claim 3, characterized in that: The method of determining the small fluctuation characteristics based on the numerical range of the impact degree includes: based on the numerical range, using a support vector machine algorithm to classify the impact degree to obtain an impact level; if the impact level reaches a preset attention level, performing an in-depth analysis of the impact level to determine potential fluctuation trend characteristics, wherein the fluctuation trend characteristics are used to indicate the possible development direction of subsequent fluctuations; and determining the small fluctuation characteristics based on the fluctuation trend characteristics.

6. The power status monitoring method of the protocol signal processing module according to claim 3, characterized in that: The refined analysis module for capturing small changes determines characteristic data of small fluctuations in the data record, including: determining the fluctuation amplitude of the instantaneous fluctuation response data record through a preset fluctuation amplitude calculation algorithm; if the fluctuation amplitude exceeds a preset threshold range, extracting small fluctuation characteristics from the data record through the refined analysis module for capturing small changes.

7. The power status monitoring method of the protocol signal processing module according to claim 2, characterized in that: The method uses a time series analysis algorithm based on the power consumption fluctuation data set to extract features of dynamic load changes and determine the load change analysis results of the load changes, including: based on the power consumption fluctuation data set, using a time series analysis algorithm to extract features of dynamic load changes, and separating trend components, periodic components and residual components corresponding to the load change rules; based on the periodic components, extracting the frequency characteristics of the periodic components, and classifying the periodic components corresponding to the frequency characteristics that are higher than a preset threshold as periodic patterns; based on the residual components, determining the statistical distribution of the residual components, and marking the residual components corresponding to the statistical distribution that deviate from the normal distribution as random fluctuation patterns; and using a sliding window technology to smooth the trend components to obtain a trend development pattern of load changes.

8. A power status monitoring system for a protocol signal processing module, characterized in that: include: An acquisition module is used to collect the operating data of the signal processing module associated with the power system to obtain a power consumption fluctuation data set, wherein the power consumption fluctuation data set is used to extract the small fluctuation characteristics corresponding to the instantaneous fluctuation of power consumption; a prediction module is used to input the small fluctuation characteristics in the power consumption fluctuation data set into the dynamic power supply state prediction model of the power system to obtain a power supply state prediction result that is suitable for a variety of load scenarios, wherein the power supply state prediction result is used to schedule and adjust the power resources of the power system, and the dynamic power supply state prediction model is obtained by training based on a preset small fluctuation data set, and the small fluctuation data set is the historical data of the small fluctuation characteristics; an adjustment module is used to input the small fluctuation characteristics in the power consumption fluctuation data set into the dynamic power supply state prediction model of the power system to obtain a power supply state prediction result that is suitable for a variety of load scenarios, wherein the power supply state prediction result is used to schedule and adjust the power resources of the power system, and the dynamic power supply state prediction model is obtained by training based on a preset small fluctuation data set, and the small fluctuation data set is the historical data of the small fluctuation characteristics; The power status prediction result is input into the power system, the power distribution efficiency of the power system is optimized and adjusted, and the target power supply mode and target resource allocation strategy required by the power system under the current operating state are determined; a switching module is used to implement dynamic switching of the target power supply mode adaptation to the power system based on the target resource allocation strategy, and obtain feedback data of the power system operation; an evaluation module is used to evaluate the effect of resource allocation balance of the power system based on the feedback data, and determine the effect evaluation result of resource allocation balance; an update module is used to update the monitoring mechanism of real-time analysis capability based on the effect evaluation result, and determine the final resource allocation strategy.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the power status monitoring method of the protocol signal processing module according to any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the power status monitoring method of the protocol signal processing module according to any one of claims 1 to 7 is implemented.

Images