A data acquisition method and system based on frequency adaptive regulation

Abstract

The present application discloses a data acquisition method and system based on frequency adaptive adjustment, which relates to the field of data acquisition. The method includes: obtaining a monitoring indicator acquisition value set; extracting local time series data of the monitoring indicator acquisition value set under various sliding windows, and calculating multi-scale difference entropy values; extracting a real-time multi-dimensional monitoring indicator matrix corresponding to the acquisition moment closest to the current time from the monitoring indicator acquisition value set, and calculating the Mahalanobis distance between the real-time multi-dimensional monitoring indicator matrix and the historical multi-dimensional monitoring indicator mean matrix; constructing the input state of a reinforcement learning model based on a first acquisition frequency, Mahalanobis distance, and multi-scale difference entropy values ​​to determine a target action through reinforcement learning; determining a second acquisition frequency based on the acquisition frequency adjustment information indicated by the first acquisition frequency and the target action. Thus, the multi-scale time series features are integrated with the real-time deviation calculation, and the reinforcement learning model is used to realize intelligent adaptive adjustment of the sampling frequency.

Description

Technical Field

[0001] The present application relates to the field of data acquisition technology, and in particular to a data acquisition method and system based on frequency adaptive regulation. Background Art

[0002] With the rapid development of data centers and cloud platforms, real-time collection and monitoring of key indicators in system operation (such as CPU, memory, disk I / O, network traffic, etc.) has become increasingly important.

[0003] Currently, most data collection methods use a fixed sampling frequency, which makes it difficult to adjust sampling accuracy based on real-time data fluctuations. This results in the loss of information about abnormal events in highly dynamic scenarios, while often leading to resource waste during stable periods. Specifically, for certain monitoring indicators in a stable state, the computational, storage, and transmission pressures associated with frequent data collection far exceed actual requirements, adding unnecessary burdens. In some cases, the data values ​​at adjacent moments differ minimally, resulting in a large amount of redundancy between the data and a waste of storage space.

[0004] There are also some related adaptive data collection mechanisms that rely on preset monitoring thresholds or anomaly triggers. Once a monitored indicator exceeds or falls below a set threshold, the system will actively switch to high-frequency sampling mode to collect more detailed data on the anomaly. However, these methods are highly dependent on trigger conditions and often can only capture one-time or relatively sudden anomalies, making it difficult to respond promptly to hidden problems that accumulate over time. Summary of the Invention

[0005] The present application provides a data acquisition method, system, storage medium, computer program product and electronic device based on frequency adaptive regulation, which are used to at least solve the hidden problems of the current related technologies that rely on preset thresholds or single abnormal trigger mechanisms, such as the difficulty in balancing one-time bursts and gradual accumulation, the accumulation of redundant data and the waste of system resources.

[0006] In a first aspect, an embodiment of the present application provides a data acquisition method based on frequency adaptive regulation, comprising: obtaining a monitoring indicator acquisition value set, wherein the monitoring indicator acquisition value set includes multiple acquisition moments and corresponding multi-dimensional monitoring indicator acquisition values ​​obtained based on a first acquisition frequency; extracting local time series data of the monitoring indicator acquisition value set under various types of sliding windows, and calculating the scale difference entropy values ​​corresponding to each of the local time series data, and generating multi-scale difference entropy values ​​by fusion; each type of sliding window has a corresponding time scale; extracting the real-time multi-dimensional monitoring indicator acquisition value corresponding to the acquisition moment closest to the current time from the monitoring indicator acquisition value set; dimensional monitoring indicator matrix, and calculate the Mahalanobis distance between the real-time multidimensional monitoring indicator matrix and the historical multidimensional monitoring indicator mean matrix; the historical multidimensional monitoring indicator mean matrix represents the mean of the multidimensional monitoring indicator matrix at each acquisition moment in a preset historical time period; the input state of the reinforcement learning model is constructed according to the first acquisition frequency, the Mahalanobis distance and the multi-scale difference entropy value, so as to determine the target action through reinforcement learning; the action space of the reinforcement learning model is defined according to the sampling frequency adjustment strategy; the second acquisition frequency is determined according to the first acquisition frequency and the acquisition frequency adjustment information indicated by the target action.

[0007] In a second aspect, an embodiment of the present application provides a data acquisition system based on frequency adaptive regulation, comprising: an original indicator acquisition unit, for acquiring a monitoring indicator acquisition value set, wherein the monitoring indicator acquisition value set includes multiple acquisition moments and corresponding multi-dimensional monitoring indicator acquisition values ​​obtained based on a first acquisition frequency; a multi-scale difference entropy extraction unit, for extracting local time series data of the monitoring indicator acquisition value set under various types of sliding windows, and calculating the scale difference entropy values ​​corresponding to each of the local time series data, and generating multi-scale difference entropy values ​​by fusion; each type of sliding window has a corresponding time scale; a Mahalanobis distance calculation unit, for extracting the acquisition moment closest to the current time from the monitoring indicator acquisition value set. A corresponding real-time multidimensional monitoring indicator matrix is ​​obtained, and the Mahalanobis distance between the real-time multidimensional monitoring indicator matrix and the historical multidimensional monitoring indicator mean matrix is ​​calculated; the historical multidimensional monitoring indicator mean matrix represents the mean of the multidimensional monitoring indicator matrix at each acquisition moment within a preset historical time period; a reinforcement learning optimization unit is used to construct the input state of the reinforcement learning model according to the first acquisition frequency, the Mahalanobis distance and the multi-scale difference entropy value, so as to determine the target action by reinforcement learning; the action space of the reinforcement learning model is defined according to the sampling frequency adjustment strategy; an acquisition frequency adjustment unit is used to determine the second acquisition frequency according to the first acquisition frequency and the acquisition frequency adjustment information indicated by the target action.

[0008] In a third aspect, an electronic device is provided, comprising: 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 perform the steps of the data acquisition method based on frequency adaptive adjustment of any embodiment of the present application.

[0009] In a fourth aspect, an embodiment of the present application provides a storage medium on which a computer program is stored, characterized in that when the program is executed by a processor, the steps of the data acquisition method based on frequency adaptive adjustment of any embodiment of the present application are implemented.

[0010] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the data acquisition method based on frequency adaptive adjustment of any embodiment of the present application.

[0011] The data acquisition method and system based on frequency adaptive regulation provided by this application can produce at least the following technical effects:

[0012] (1) The multi-scale difference entropy and Mahalanobis distance are used to capture the multi-scale temporal complexity and real-time change trend of the system's current monitoring indicators, dynamically construct the system state, and learn the reward relationship between historical states and behaviors through the reinforcement learning model. The optimal sampling frequency adjustment strategy can be adaptively output, so that the system can improve the sampling accuracy when the data fluctuates violently and reduce the sampling frequency when the system is stable, thereby effectively avoiding the coexistence of abnormal information loss and resource waste.

[0013] (2) Compared with the traditional method that relies on a single anomaly threshold trigger mechanism, by introducing the Mahalanobis distance, a statistical distance between multi-dimensional indicators, to judge the degree of deviation between real-time data and historical patterns, the system has a stronger overall perception ability and can discover potential anomalies with nonlinear correlations in the data. In addition, the introduction of multi-scale differential entropy enables the system to evaluate the complexity and volatility of data at multiple time granularities, and has the ability to detect anomaly features that evolve slowly and accumulate gradually, thereby significantly improving the system's recognition effect on non-sudden anomalies.

[0014] (3) Through the continuous updating of the relationship between system state, action and reward through the reinforcement learning model, it has the ability of online learning and adaptive adjustment, and can continuously optimize the indicator collection strategy under different business load modes, operating environments and abnormal behavior modes, forming a closed-loop optimized data collection mechanism, which is significantly better than the traditional method that relies on manually set strategies or static rules.

[0015] This technical solution integrates multi-scale time series features with real-time deviation calculation, and uses a reinforcement learning model to achieve intelligent adaptive adjustment of the sampling frequency. This optimizes resource utilization, improves the sensitivity of anomaly detection, and enhances the system's self-learning capabilities, providing an efficient and accurate intelligent solution for the monitoring systems of data centers and cloud platforms. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0017] Figure 1 A flow chart showing an example of a data acquisition method based on frequency adaptive adjustment according to an embodiment of the present application is shown;

[0018] Figure 2 A schematic diagram showing an example of a state transition action in a reinforcement learning model;

[0019] Figure 3 An operational flow chart of an example of calculating multi-scale difference entropy values ​​according to an embodiment of the present application is shown;

[0020] Figure 4 An operational flow chart illustrating an example of calculating Mahalanobis distance according to an embodiment of the present application is shown;

[0021] Figure 5 A structural block diagram of an example of a data acquisition system based on frequency adaptive adjustment according to an embodiment of the present application is shown;

[0022] Figure 6 This is a schematic structural diagram of an embodiment of an electronic device of the present application. DETAILED DESCRIPTION

[0023] To make the purpose, technical solutions, and advantages of the embodiments of this application more clear, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0024] It's important to note that in network telemetry for data centers and cloud platforms, the sampling frequency directly impacts both measurement cost and accuracy. If the sampling frequency is too low, critical information (such as microbursts) may be missed when metrics undergo significant changes, resulting in inaccurate measurements. However, if the sampling frequency is too high, measurement overhead increases. Therefore, developing adaptive frequency-varying measurement algorithms that adapt to dynamic network changes is crucial for reducing measurement costs.

[0025] Among current technologies, Moneo is an intuitive adaptive frequency control method that adjusts the measurement frequency based on changes in traffic peaks to reduce measurement overhead. However, it cannot accurately detect sudden performance spikes and dips. Furthermore, it lacks a theoretical basis for frequency conversion, and there's no way to guarantee the accuracy of inferred fine-grained data after coarse-grained data is collected using frequency conversion. Furthermore, some experts and scholars have proposed an adaptive sampling method using artificial neural networks to reduce traffic measurement overhead, but this method is only applicable to smooth traffic. Cuckoo sampling is an adaptive sampling method based on Cuckoo hashing that adjusts the sampling rate based on hash table memory usage. Event-driven adaptive sampling methods typically require the monitoring system to proactively schedule sensor nodes to increase the sampling frequency only when specific abnormal events are triggered. To improve event monitoring efficiency and resource utilization, e-Sampling adjusts the sampling rate based on the frequency characteristics of the signal. Adaptive sampling methods based on compressed sensing estimate the minimum required sampling rate within each sampling window, given a given perceptual quality, and adjust the sensor sampling frequency accordingly. However, perceptual quality estimation incurs high computational costs. On the other hand, the two-stage pattern sampling method collects basic information about historical data in the first stage and uses this information to determine the sampling frequency in the second stage. Although these methods have explored variable-frequency measurement, existing research lacks theoretical basis, making it difficult to achieve a balance between measurement accuracy and cost, and unable to capture microbursts.

[0026] It should be understood that the purpose of the above description of the current related art is only to facilitate the public to better understand the inventive spirit and motivation of this application, and is not to be construed as limiting this application. In addition, the technical solutions described in the above-mentioned current related art are not prior art and may also be undisclosed technical solutions, such as solutions under research or in the laboratory stage.

[0027] In the technical solutions of this application, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved shall comply with the provisions of relevant laws and regulations and shall not violate public order and good morals.

[0028] Figure 1A flow chart of an example of a data acquisition method based on frequency adaptive adjustment according to an embodiment of the present application is shown.

[0029] The execution entity of the method of the embodiment of the present application can be any controller or processor with computing or processing capabilities. Specifically, it can be implemented by a data acquisition management platform or data acquisition management system. By constructing input features based on multidimensional statistical difference measurement and temporal complexity analysis, and combining reinforcement learning to achieve intelligent adjustment of sampling frequency, it effectively improves the robustness, agility, and resource efficiency of the data acquisition system in complex operating environments.

[0030] In some examples, it can be integrated into an electronic device or terminal through software, hardware, or a combination of software and hardware, and the type of terminal or electronic device can be diverse, such as a mobile phone, tablet computer, or desktop computer, etc.

[0031] like Figure 1 As shown, in step S110, a monitoring indicator collection value set is obtained, and the monitoring indicator collection value set includes multiple collection moments obtained based on the first collection frequency and corresponding multi-dimensional monitoring indicator collection values.

[0032] Specifically, the system collects multi-dimensional system monitoring metrics in real time from multiple running entities (such as physical servers, virtual machines, containers, database instances, etc.) distributed within a data center or cloud platform at a preset first collection frequency (e.g., 10 seconds or 30 seconds). Each sampling point (or sampling moment) generates a multi-dimensional vector. It should be understood that the metric dimensions indicated by the multi-dimensional vector at each sampling point can be diverse, such as CPU frequency or application performance metrics, and this is not a limitation.

[0033] In some examples of the embodiments of the present application, the monitoring indicator dimension indicated by the multi-dimensional monitoring indicator collection value includes at least one of the following: computing resource indicator, memory resource indicator, storage resource indicator or network resource indicator.

[0034] More specifically, computing resource indicators include CPU utilization, computing task processing power, thread occupancy, etc., which reflect the usage of computing resources and can help determine whether the system has a processing capacity bottleneck. Memory resource indicators include memory usage, memory allocation and recovery, swap space usage, etc., which help understand system performance and avoid memory leaks. Storage resource indicators include disk read and write speeds, disk space usage, I / O wait time, etc., which can help identify disk I / O bottlenecks and insufficient storage problems. Network resource indicators include network bandwidth, latency, packet loss rate, number of connections, etc., which can reflect delays or bandwidth bottlenecks in the data transmission process. As a result, the collected data covers different dimensions of computing, memory, storage and network resources, ensuring comprehensive monitoring of system status.

[0035] During the data collection phase, it is necessary to ensure that the system can efficiently acquire the aforementioned multi-dimensional monitoring indicators and store the collected data with timestamps to accurately record time series characteristics. Furthermore, during each data collection phase, the collected data is verified. For example, self-checks are performed by comparing the timestamps of two consecutive data collections and fluctuations in the collected values ​​to ensure that no data is lost during the data collection process.

[0036] In step S120, the local time series data of the monitoring indicator collection value set under each type of sliding window are extracted, and the scale difference entropy value corresponding to each local time series data is calculated, and the multi-scale difference entropy value is generated by fusion, and each type of sliding window has a corresponding time scale.

[0037] In some embodiments, for the acquired monitoring indicator collection value set, the system constructs sliding windows of multiple time scales, each type of sliding window corresponding to a different time span (e.g., short-term window, medium-term window, and long-term window), such as 1 minute, 5 minutes, and 10 minutes. Then, within each type of sliding window, the corresponding local time series data sequence is extracted, and then a specific entropy algorithm (e.g., sample entropy, permutation entropy, fuzzy entropy algorithm, etc.) is used to calculate the scale difference entropy value of the time series data. This entropy value reflects the complexity and uncertainty of data fluctuations at the corresponding time scale. Furthermore, the difference entropy values ​​calculated at different time scales are fused to obtain a multi-scale difference entropy value, so that the multi-scale entropy value can effectively capture the subtle changes and mutations of the data within different time ranges, and can effectively produce the comprehensive characteristics of the system's short-term fluctuations and long-term trends. At the same time, it also avoids the risk of "overfitting" a single window size to a specific behavior pattern, thereby improving the generalization ability of the system.

[0038] In step S130 , the real-time multidimensional monitoring indicator matrix corresponding to the collection time closest to the current time is extracted from the monitoring indicator collection value set, and the Mahalanobis distance between the real-time multidimensional monitoring indicator matrix and the historical multidimensional monitoring indicator mean matrix is ​​calculated.

[0039] Here, the historical multidimensional monitoring indicator mean matrix represents the mean of the multidimensional monitoring indicator matrix at each sampling moment within a preset historical time period. Specifically, the sampling moment closest to the current time is selected from the monitoring indicator collection value set to construct a real-time multidimensional monitoring indicator matrix, with each column in the matrix corresponding to a monitoring indicator. Simultaneously, the mean of the multidimensional indicators at each sampling moment is calculated from the data collected within the preset historical time period to form a historical multidimensional monitoring indicator mean matrix, which is used to represent the normal operating status of the system. Furthermore, the real-time matrix is ​​compared with the historical mean matrix using the Mahalanobis distance calculation method to quantify the degree of deviation between the current state and the historical norm.

[0040] It should be noted that the Mahalanobis distance is a distance metric used to measure the similarity between multidimensional data points. Unlike the Euclidean distance, the Mahalanobis distance considers the covariance structure of the data, making it better able to handle correlations and scale differences in high-dimensional data. Furthermore, because the Mahalanobis distance considers the covariance between various indicators, it allows the system to comprehensively consider the correlations between various indicators when dealing with multidimensional data while quickly detecting anomalies. Its results are more suitable for reflecting the actual distances between multidimensional data.

[0041] In step S140, an input state of a reinforcement learning model is constructed according to the first acquisition frequency, the Mahalanobis distance, and the multi-scale difference entropy value to determine the target action through reinforcement learning. The action space of the reinforcement learning model is defined according to the sampling frequency adjustment strategy.

[0042] In some embodiments, the first acquisition frequency, Mahalanobis distance, and multi-scale difference entropy value currently used for sampling of the system are fused to form a multi-dimensional state vector, which fully reflects the real-time changes and historical deviations of the current monitoring data. The Mahalanobis distance is used to reflect the difference between real-time multi-dimensional monitoring data and historical data. If the Mahalanobis distance is large, it means that the current data deviates greatly from the historical state, which may indicate that an abnormality or load change has occurred in the system, and the reinforcement learning model may need to consider increasing the sampling frequency. The volatility and complexity of the system data are measured by the multi-scale difference entropy value. A higher difference entropy value indicates that the data changes greatly, and more frequent sampling may be required to capture the changes; a lower difference entropy value indicates that the system is stable and the sampling frequency can be appropriately reduced.

[0043] Based on this state vector, a decision-making model is then established using reinforcement learning algorithms (such as deep Q-networks and policy gradients). During training, the model continuously optimizes its strategy based on the feedback from the actions in the environment, ensuring that it can output the optimal sampling frequency adjustment action in different environments. During the inference phase, the target action output from the optimized action space is used to determine the corresponding sampling frequency adjustment information, such as increasing the sampling frequency by a certain amount, decreasing the sampling frequency by a certain amount, or maintaining the sampling frequency unchanged.

[0044] Figure 2 A schematic diagram showing an example of state transition actions in a reinforcement learning model.

[0045] like Figure 2 As shown, it involves multiple states S The action corresponding to the state transition in the state space of 1~Sn, for example a 1 means from S 1 to S 2. State transition action, a 2 means from S 2 to S 1 action, a 3 means from S 1 to S 3 state transition, and so on. Here, the corresponding state migration can occur based on the strategy, and each state transition strategy can be used to generate different transitions. S 1 transfer strategy, action can occur a 2 or a 3.

[0046] It should be noted that the range of states that a state can transfer to (also called transferable states) may be limited or conditional, for example S 1~ S None of the 3 S 4~ S n occurs between states, and for states S The state that 1 can transfer to is S 2 and S 3, etc.

[0047] In some embodiments, each action has a corresponding action reward, and each action reward can be determined based on a preset reward function. Generally, if the transfer reward is larger, it can be considered that the transfer action is more valuable, and the system will give priority to executing this action. For example, if the action a The reward corresponding to 1 is greater than a The reward corresponding to 3 indicates the transfer action a 1 is more valuable.

[0048] In an embodiment of the present application, the action of the reinforcement learning model is used to express the adjustment strategy for the acquisition frequency, which includes increasing the sampling frequency by a certain amount, reducing the sampling frequency by a certain amount, or maintaining the sampling frequency unchanged. In this way, the sampling frequency adjustment decision process is regarded as an "action selection" problem, and the reinforcement learning model is used to learn the optimal sampling frequency adjustment strategy through continuous trial and error. The input state of the reinforcement learning model consists of the first acquisition frequency currently used by the system and the real-time calculated Mahalanobis distance and multi-scale difference entropy value, which together affect the sampling frequency adjustment decision.

[0049] In step S150 , a second acquisition frequency is determined according to the first acquisition frequency and acquisition frequency adjustment information indicated by the target action.

[0050] More specifically, based on the target action output by the reinforcement learning model, the system interprets the corresponding sampling frequency adjustment information and dynamically transforms the first acquisition frequency to obtain a new second acquisition frequency. For example, the new acquisition frequency can be significantly increased to capture data details of sudden anomalies, or it can be appropriately decreased to reduce invalid data collection when the system is stable, or it can remain unchanged to maintain the current monitoring indicator collection status. Thus, efficient resource management is achieved through adaptive adjustment of the sampling frequency, forming a good data acquisition feedback loop that can ensure that critical data is not lost in a highly dynamic environment while reducing the computing, storage, and transmission burdens during stable phases.

[0051] Through the embodiments of the present application, real-time multi-scale data feature extraction is integrated with reinforcement learning decision-making. By combining multi-scale difference entropy and Mahalanobis distance to comprehensively reflect the system status and potential anomalies, the sampling frequency can be quickly increased when an abnormality occurs to ensure the capture of key data. The reinforcement learning model is called to continuously correct the strategy through actual sampling feedback, realizing a complete closed loop from data acquisition, feature extraction to decision feedback, and realizing intelligent adaptive adjustment of data acquisition frequency.

[0052] It should be noted that, in the embodiments of the present application, the action space of the reinforcement learning model is relatively small, involving only three types of actions (i.e., increasing, decreasing, or maintaining the acquisition frequency), and the Deep Q-Network (DQN) can effectively deal with the problem of discrete action space. Therefore, in some examples of the embodiments of the present application, the reinforcement learning model adopts the Deep Q-Network. In addition, the Deep Q-Network can also use the Experience Replay technology, using the experience replay buffer to store samples of historical state-action-reward-next state, and randomly sample training to break the sample correlation and thereby reduce the correlation between samples, thereby accelerating training convergence. Thus, by using the architecture of the Deep Q-Network, the system can efficiently learn strategies with the help of a large number of training samples, avoid the limitations brought about by manually setting the sampling frequency, and improve the efficiency of online decision-making.

[0053] As described above, the policy network of the reinforcement learning model is trained through training samples generated by the interaction between the agent and the environment, and its optimization direction is determined by the goal defined by the reward function.

[0054] In some examples of the embodiments of the present application, a reward function is used to reflect the reward that can be obtained by the "state transition" caused by taking an action.

[0055] Specifically, the reward function is expressed as follows:

[0056] , formula (1)

[0057] Where, Is the reward function, indicating that in the current state The reward generated by the state transition action; represents the Mahalanobis distance term, represents the multi-scale difference entropy term, Represents the sampling frequency deviation term.

[0058] , Formula (2)

[0059] Where, Indicates the current status The corresponding Mahalanobis distance is Indicates the next state The corresponding Mahalanobis distance is Represents the Mahalanobis distance reward correction coefficient.

[0060] It should be noted that in formula (2), the Mahalanobis distance term mainly measures the distance between the state before and after the state transition (i.e., the state arrive ) monitors the difference between real-time monitoring data and historical data benchmarks. When the gap between the current state data and the historical mean decreases after a certain action, it indicates that the action helped restore the system to normal; conversely, it indicates that the current action did not achieve the expected effect. This effectively captures the degree of deviation between the system's abnormal and normal states, allowing the positive feedback generated by the action to guide the agent to minimize the gap between the state and the historical normal state during state transition, thereby improving the response speed of anomaly detection and system stability restoration.

[0061] , Formula (3)

[0062] Where, Indicates the current status The corresponding multi-scale difference entropy value, Indicates the next state The corresponding multi-scale difference entropy value, Represents the multi-scale difference entropy reward correction coefficient.

[0063] It should be noted that the multi-scale differential entropy term reflects the volatility and uncertainty of data at different time scales, and thus reflects the complexity of the system state. Large system fluctuations often indicate potential anomalies. By using the difference in differential entropy before and after state transition, it can be assessed whether the data volatility is improved after taking a certain action. Specifically, during the training process, the model compares the state and The entropy value captures improvements or deteriorations in data stability. This allows for a more comprehensive capture of system fluctuations in the short and long term, enabling monitoring strategies to better adapt to different dynamic situations, further reducing the risk of false positives caused by data fluctuations, and more accurately guiding the agent to adjust its sampling frequency.

[0064] , Formula (4)

[0065] Where, represents the ideal sampling frequency indicated by the label of the training sample, Indicates the current status The corresponding acquisition frequency, Indicates the next state The corresponding acquisition frequency, Indicates the sampling frequency deviation correction factor.

[0066] In formula (4), This measure measures the deviation between the current system sampling frequency and the preset target sampling frequency. Its core objective is to quantify the impact of actions on the sampling frequency adjustment. By comparing the sampling frequency deviation before and after state transitions and using the square of the normalized difference to penalize excessive deviations, the agent can mitigate overly frequent or insufficient frequency adjustments, keeping the sampling frequency as close to the ideal target as possible. This ensures that the system tends toward the target frequency when adjusting the sampling frequency, optimizes the balance of data collection, and helps reduce the number of frequency adjustments, thereby reducing resource waste.

[0067] Regarding the explanation of the various correction coefficients involved in the above reward function, and Mainly controls the impact of Mahalanobis distance and multi-scale difference entropy on rewards. Generally, data with larger fluctuations require a higher sampling frequency, so these two coefficients are larger. Penalties for adjusting sampling frequency deviations ensure that the system avoids unnecessary frequency adjustments. More specifically, these correction coefficients can be optimized through the sample training process, allowing the model to adjust the sampling frequency based on the actual system conditions.

[0068] By using the reward function provided in the embodiment of the present application, the state is calculated arrive The changes in various indicators are not just focused on the state evaluation at a single moment, but truly capture the transition effect of the system state after the action is executed. This allows the agent to learn to take the optimal action in different situations through training, which can not only improve abnormal conditions but also ensure a balance between sampling frequency and resource consumption. In addition, the three key indicators of Mahalanobis distance, multi-scale difference entropy, and sampling frequency deviation system load are simultaneously introduced into the reward function, forming a joint feedback mechanism for anomaly detection, data fluctuation balance, and resource management. The agent can automatically increase the sampling frequency when data fluctuations are large and automatically reduce the sampling frequency when the load is high, realizing system indicator monitoring based on adaptive frequency adjustment.

[0069] Figure 3 The following is a flowchart showing an example of calculating the multi-scale difference entropy value according to an embodiment of the present application. In the embodiment of the present application, a non-parametric entropy calculation method is adopted.

[0070] like Figure 3 As shown, in step S310, each local time series data is processed based on the kernel density estimation method to respectively determine the scale probability density function at the corresponding time scale.

[0071] It should be noted that Kernel Density Estimation (KDE) is a nonparametric probability density estimation method used to estimate continuous probability density functions from discrete data. In the context of multidimensional surveillance data, KDE can help estimate the distribution of local time series data. Furthermore, KDE is particularly well-suited for processing surveillance data with high levels of noise, effectively removing meaningless fluctuations and extracting the true patterns of data change.

[0072] Specifically, by smoothing the data in each time scale window, the probability density function of the data in the window can be obtained using the kernel density estimation method: , which reveals the volatility and randomness of the data.

[0073] , Formula (5)

[0074] Where, For the The probability density function of the data in the time scale window is the probability density function of the local time series data at the point The probability density at ; represents the bandwidth parameter of the kernel density estimate, which determines the degree of data smoothing; Indicates the The total number of sampling points within the class time scale window, is the Gaussian kernel function, Indicates the corresponding The first The value of the sampling point.

[0075] Therefore, by applying KDE, the calculation of differential entropy does not rely on traditional discretization or binning methods, avoids the inaccuracy caused by data cutting, and can handle smoother continuous data distribution.

[0076] In step S320 , based on the scale probability density function of each time scale, the sliding window difference entropy of the corresponding time scale is calculated respectively.

[0077] Here, differential entropy is used to measure the "complexity" or "randomness" of a dataset; higher differential entropy indicates greater data variability. In the calculation of multiscale differential entropy, the data distribution within each timescale window is evaluated. Calculating the differential entropy of the data within that window effectively quantifies volatility at different timescales and reveals the stability of the system at each timescale.

[0078] Specifically, for the data of each time scale window, the difference entropy is calculated according to the probability density obtained by KDE, and then the distribution of each window is processed using the formula of difference entropy to ensure that the volatility and change pattern of the data can be reflected.

[0079] , Formula (6)

[0080] Where, Indicates the Sliding window difference entropy corresponding to the class time scale window.

[0081] Therefore, by calculating differential entropy, the complexity of local time series data can be captured. Especially when faced with complex multidimensional data, differential entropy provides a method for quantifying and comparing volatility and uncertainty, helping the system detect changes at different levels, such as short-term fluctuations, medium-term trends, and long-term stability. In addition, in the embodiments of the present application, differential entropy calculation does not rely solely on data from a single dimension, but instead combines data from multiple monitoring indicator dimensions (such as computing resources, memory resources, storage resources, and network resources), enabling the system to comprehensively evaluate the interactions of multidimensional monitoring data.

[0082] In step S330 , weighted summation is performed on the difference entropies of the sliding windows to obtain corresponding multi-scale difference entropy values.

[0083] Here, weighted summation combines the differential entropies of different time scales according to their importance, yielding a comprehensive multi-scale differential entropy value. By assigning different weights to each time scale, the contribution of each scale to the final result can be more flexibly adjusted in different dynamic environments.

[0084] Specifically, the timescale weights are calculated using the inverse ratio of the window size or an exponential decay method, ensuring that longer timescale windows have a greater impact on the final differential entropy value. By taking the weighted sum of the differential entropies across all timescales, the resulting multi-scale differential entropy value fully reflects the dynamic fluctuations of the system state.

[0085] , formula (7)

[0086] , formula (8)

[0087] Where, Indicates the The weight of the class time scale window indicates the contribution of this time scale to the final multi-scale difference entropy value; Represents the attenuation factor, which is used to control the influence of time scale on weight; represents the multi-scale difference entropy value, represents the total number of classes at the time scale, Indicates the The time scale size of the class time scale window; represents the category index of the time scale window, is the normalization factor for all time scale weights, which is used to ensure that the sum of the weights of time scale windows under all categories is 1.

[0088] This weighted summation approach enables the fusion of multi-scale information, enabling the system to more comprehensively perceive data volatility across different timeframes. Longer timescale windows can capture long-term trends, while shorter timescale windows can accurately capture instantaneous fluctuations, eliminating the shortcomings of a single timescale and improving the system's monitoring and anomaly detection capabilities.

[0089] Compared to the traditional fixed sampling frequency and simple threshold judgment method, the multi-scale differential entropy calculation used in the embodiment of the present application can comprehensively analyze the dynamic changes of the system. By using the kernel density estimation method to process the data distribution, complex data patterns can be more accurately reflected, reducing the errors caused by data segmentation and discretization in traditional methods. In addition, the advantage of differential entropy in abnormal data processing enables the system to maintain high accuracy and robustness when facing complex and noisy environments.

[0090] Figure 4 An operational flowchart of an example of calculating Mahalanobis distance according to an embodiment of the present application is shown.

[0091] It should be noted that in traditional Mahalanobis distance calculations, the distance is calculated based on data differences, without considering the correlation between the data. However, in actual monitoring systems, various resources (such as CPU, memory, storage, network, etc.) may be highly correlated. Therefore, in the embodiments of the present application, the covariance matrix is ​​incorporated into the calculation of the Mahalanobis distance. The covariance matrix reflects the joint variation pattern between various monitoring indicators. Therefore, the calculated Mahalanobis distance not only reflects the difference between the current data and the historical data mean, but also takes into account the coordinated changes between different monitoring indicators.

[0092] Specifically, first select the monitoring data of the nearest moment. Assuming that the system sampling frequency is (for example, 1Hz), then in the system time stream, the monitoring data closest to the current moment can be selected by the timestamp. The nearest moment refers to the time from the current timestamp. The data point at the shortest sampling moment.

[0093] Formula (8)

[0094] Where, Indicates the timestamp of the current moment. The timestamp representing each data point at the moment of collection.

[0095] In this way, the system can ensure that each time an anomaly detection is performed, the monitoring data closest to the current moment is used, ensuring real-time and accuracy.

[0096] like Figure 4 As shown, in step S410, the mean of each monitoring indicator within a preset historical time period is calculated to obtain a historical multi-dimensional monitoring indicator mean matrix.

[0097] Here, by calculating the mean of historical data, we obtain a baseline of normal operation for comparison with real-time data. The historical mean represents the normal operation of the system over the past period of time and forms the basis for current data analysis.

[0098] Specifically, the historical time period can be calculated The mean of all sampling points in the , so that the mean matrix It contains the average value of each monitoring indicator within the historical time period. The calculation can flexibly adjust the time period length (for example, the past 5 minutes, the past 24 hours, etc.) to meet different analysis needs.

[0099] , formula (9)

[0100] Where, Represents the mean matrix of historical multi-dimensional monitoring indicators within the historical time period, Indicates the historical period The multi-dimensional monitoring indicator matrix at each collection moment, Represents the total number of data points in the historical time period.

[0101] Therefore, by constructing a historical mean matrix, a benchmark for the normal state of the system is provided.

[0102] In step S420 , the covariance matrix is ​​calculated.

[0103] Here, the covariance matrix reflects the correlation between various monitoring indicators. For multidimensional monitoring data, if there are correlations between different resource indicators, directly calculating their Euclidean distance will not accurately capture the overall changes in the system. However, the Mahalanobis distance, by introducing the covariance matrix, can consider the relationships between different dimensions.

[0104] Specifically, the covariance matrix of historical data is calculated, which measures the joint changes between various monitoring indicators. For example, CPU and memory usage may show a certain positive correlation, and the covariance matrix can accurately capture this relationship.

[0105] , formula (10)

[0106] Where, is the covariance matrix, which represents the covariance between various monitoring indicators; Represents a vector transpose operation.

[0107] Therefore, the covariance matrix reveals the mutual influence between different monitoring dimensions, and the Mahalanobis distance can eliminate the deviations between different monitoring indicators due to different units and scales, ensuring the accuracy of the calculation.

[0108] In step S430 , the Mahalanobis distance between the real-time multi-dimensional monitoring indicator matrix and the historical multi-dimensional monitoring indicator mean matrix is ​​calculated.

[0109] The Mahalanobis distance is used to measure the difference between current data and historical data. The Mahalanobis distance takes into account the covariance matrix of the data, so it can comprehensively evaluate the changes in multidimensional data and take into account the correlation between different dimensions.

[0110] Specifically, the Mahalanobis distance is calculated using the difference between real-time and historical data. The inverse of the covariance matrix can be incorporated into the calculation to eliminate the effects of scale inconsistencies between dimensions. A larger Mahalanobis distance indicates that the current data deviates further from historical data, potentially indicating anomalies.

[0111] , formula (11)

[0112] Where, represents the real-time multi-dimensional monitoring indicator matrix, express and The Mahalanobis distance between Represents the covariance matrix The inverse matrix of .

[0113] Therefore, by calculating the Mahalanobis distance, the system can accurately assess the difference between the current state and the normal historical state, thereby improving the accuracy of anomaly detection. In particular, the Mahalanobis distance calculated in conjunction with the covariance matrix can comprehensively consider changes in multi-dimensional monitoring indicators and provide more accurate monitoring feedback.

[0114] Through the embodiments of the present application, the Mahalanobis distance and covariance matrix are introduced. By comparing real-time data with historical means and combining the covariance matrix, the global state of the system can be comprehensively evaluated, which can better reflect the overall health of the system, rather than just the fluctuations in a single dimension. Based on the covariance matrix, the system can capture the correlation between the various monitoring indicators and take it into account in the Mahalanobis distance calculation, achieving a comprehensive consideration of the coordinated changes of multi-dimensional resources, especially in the case of highly correlated monitoring data (such as CPU and memory, disk I / O and network bandwidth, etc.), so that the system can more accurately identify anomalies and avoid the synergistic effects that may be ignored by single-dimensional monitoring methods.

[0115] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity of description, they are all expressed as a series of combined actions, but those skilled in the art should be aware that this application is not limited by the order of the actions described, because according to this application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by this application. In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0116] Figure 5 A structural block diagram of an example of a data acquisition system based on frequency adaptive adjustment according to an embodiment of the present application is shown.

[0117] like Figure 5 As shown, the data acquisition system 500 based on frequency adaptive adjustment includes an original indicator acquisition unit 510, a multi-scale difference entropy extraction unit 520, a Mahalanobis distance calculation unit 530, a reinforcement learning optimization unit 540 and an acquisition frequency adjustment unit 550.

[0118] The original indicator acquisition unit 510 is used to acquire a monitoring indicator collection value set, where the monitoring indicator collection value set includes multiple collection moments obtained based on a first collection frequency and corresponding multi-dimensional monitoring indicator collection values.

[0119] The multi-scale difference entropy extraction unit 520 is used to extract the local time series data of the monitoring indicator collection value set under each type of sliding window, and calculate the scale difference entropy value corresponding to each of the local time series data, and generate a multi-scale difference entropy value by fusion; each type of sliding window has a corresponding time scale.

[0120] The Mahalanobis distance calculation unit 530 is used to extract the real-time multidimensional monitoring indicator matrix corresponding to the collection moment closest to the current time from the monitoring indicator collection value set, and calculate the Mahalanobis distance between the real-time multidimensional monitoring indicator matrix and the historical multidimensional monitoring indicator mean matrix; the historical multidimensional monitoring indicator mean matrix represents the mean of the multidimensional monitoring indicator matrix at each collection moment within a preset historical time period.

[0121] The reinforcement learning optimization unit 540 is used to construct the input state of the reinforcement learning model based on the first acquisition frequency, the Mahalanobis distance and the multi-scale difference entropy value, so as to determine the target action through reinforcement learning; the action space of the reinforcement learning model is defined according to the sampling frequency adjustment strategy.

[0122] The acquisition frequency adjustment unit 550 is configured to determine a second acquisition frequency according to the first acquisition frequency and acquisition frequency adjustment information indicated by the target action.

[0123] In some embodiments, an embodiment of the present application provides a non-volatile computer-readable storage medium, in which one or more programs including execution instructions are stored. The execution instructions can be read and executed by an electronic device (including but not limited to a computer, a server, or a network device, etc.) to execute the steps of any of the above-mentioned data acquisition methods based on frequency adaptive adjustment in the present application.

[0124] In some embodiments, the embodiments of the present application also provide a computer program product, which includes a computer program stored on a non-volatile computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer is caused to perform any step of the above-mentioned data acquisition method based on frequency adaptive adjustment.

[0125] In some embodiments, an embodiment of the present application also provides an electronic device, comprising: 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 perform the steps of the data acquisition method based on frequency adaptive adjustment.

[0126] Figure 6 FIG. 1 is a schematic diagram of the hardware structure of an electronic device for executing a data acquisition method based on frequency adaptive adjustment provided by another embodiment of the present application. Figure 6 As shown, the device includes:

[0127] One or more processors 610 and memory 620, Figure 6 A processor 610 is taken as an example.

[0128] The device for executing the data acquisition method based on frequency adaptive regulation may further include: an input device 630 and an output device 640 .

[0129] The processor 610, the memory 620, the input device 630 and the output device 640 may be connected via a bus or other means. Figure 6 The bus connection is taken as an example.

[0130] Memory 620, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer executable programs, and modules, such as the program instructions / modules corresponding to the frequency adaptive adjustment-based data acquisition method in the embodiments of the present application. Processor 610 executes the non-volatile software programs, instructions, and modules stored in memory 620 to execute various server functional applications and data processing, thereby implementing the frequency adaptive adjustment-based data acquisition method in the above-mentioned method embodiment.

[0131] The memory 620 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and applications required for at least one function; the data storage area may store data created based on the use of the electronic device, etc. In addition, the memory 620 may include a high-speed random access memory and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory 620 may optionally include a memory remotely located relative to the processor 610, and these remote memories may be connected to the electronic device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0132] The input device 630 may receive input digital or character information and generate signals related to user settings and function control of the electronic device. The output device 640 may include a display device such as a display screen.

[0133] The one or more modules are stored in the memory 620 , and when executed by the one or more processors 610 , perform the data acquisition method based on frequency adaptive adjustment in any of the above method embodiments.

[0134] The above-mentioned product can execute the method provided in the embodiment of this application, and has the functional modules and beneficial effects corresponding to the execution method. For technical details not fully described in this embodiment, please refer to the method provided in the embodiment of this application.

[0135] The electronic devices of the embodiments of the present application exist in various forms, including but not limited to:

[0136] (1) Mobile communication devices: These devices are characterized by their mobile communication capabilities and their primary purpose is to provide voice and data communications. These terminals include smartphones, multimedia phones, feature phones, and low-end phones.

[0137] (2) Ultra-mobile personal computer devices: These devices fall under the category of personal computers and have computing and processing capabilities, and generally also have mobile Internet access. These terminals include PDAs, MIDs, and UMPCs.

[0138] (3) Portable entertainment devices: These devices can display and play multimedia content. They include audio and video players, handheld game consoles, e-books, smart toys, and portable car navigation devices.

[0139] (4) Other onboard electronic devices with data interaction functions, such as onboard computer devices installed in vehicles.

[0140] 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, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of this embodiment.

[0141] Through the description of the above embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a general hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the relevant technology, 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.

[0142] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application 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 embodiments of the present application.

Claims

1. A data acquisition method based on frequency adaptive regulation, comprising: Acquire a monitoring indicator collection value set, wherein the monitoring indicator collection value set includes multiple collection moments obtained based on the first collection frequency and corresponding multi-dimensional monitoring indicator collection values; Extracting the local time series data of the monitoring indicator collection value set under each type of sliding window, and calculating the scale difference entropy value corresponding to each of the local time series data, and generating a multi-scale difference entropy value by fusion; each type of sliding window has a corresponding time scale; Extracting a real-time multidimensional monitoring indicator matrix corresponding to the collection moment closest to the current time from the monitoring indicator collection value set, and calculating the Mahalanobis distance between the real-time multidimensional monitoring indicator matrix and the historical multidimensional monitoring indicator mean matrix; the historical multidimensional monitoring indicator mean matrix represents the mean value of the multidimensional monitoring indicator matrix at each collection moment within a preset historical time period; The first acquisition frequency, the Mahalanobis distance, and the multi-scale difference entropy value are combined into a state vector, which is input into a reinforcement learning model; the action space of the reinforcement learning model includes three types of discrete actions: an increase sampling frequency action for indicating the frequency increase amplitude, a decrease sampling frequency action for indicating the frequency decrease amplitude, and a maintain sampling frequency action for indicating that the acquisition frequency remains unchanged; and the reinforcement learning model selects an optimal action as a target action based on the state vector; A second acquisition frequency is determined according to the first acquisition frequency and acquisition frequency adjustment information indicated by the target action.

2. The method according to claim 1, wherein The monitoring indicator dimension indicated by the multi-dimensional monitoring indicator collection value includes at least one of the following: a computing resource indicator, a memory resource indicator, a storage resource indicator or a network resource indicator.

3. The method according to claim 1, wherein The calculating of the scale difference entropy values ​​corresponding to the respective local time series data and generating a multi-scale difference entropy value by fusing the scale difference entropy values ​​includes: The local time series data are processed based on the kernel density estimation method to respectively determine the scale probability density function at the corresponding time scale: ， Where, For the The probability density function of the data in the time scale window is the probability density function of the local time series data at the point The probability density at ; represents the bandwidth parameter of the kernel density estimate, which determines the degree of data smoothing; Indicates the The total number of sampling points within the class time scale window, is the Gaussian kernel function, Indicates the corresponding The first The value of each sampling point; Based on the scale probability density function of each time scale, the sliding window difference entropy of the corresponding time scale is calculated respectively: ， Where, Indicates the Sliding window difference entropy corresponding to the class time scale window; Perform weighted summation of the difference entropy of each sliding window to obtain the corresponding multi-scale difference entropy value: ， ， Where, Indicates the The weight of the class time scale window indicates the contribution of this time scale to the final multi-scale difference entropy value; Represents the attenuation factor, which is used to control the influence of time scale on weight; represents the multi-scale difference entropy value, represents the total number of classes at the time scale, Indicates the The time scale size of the class time scale window; represents the category index of the time scale window, is the normalization factor for all time scale weights, which is used to ensure that the sum of the weights of time scale windows under all categories is 1.

4. The method according to claim 3, wherein: The calculating of the Mahalanobis distance between the real-time multi-dimensional monitoring indicator matrix and the historical multi-dimensional monitoring indicator mean matrix includes: Calculate the mean of each monitoring indicator within the preset historical time period to obtain the historical multi-dimensional monitoring indicator mean matrix: ， Where, Represents the mean matrix of historical multi-dimensional monitoring indicators within the historical time period, Indicates the number of historical periods The multi-dimensional monitoring indicator matrix at each collection moment, Indicates the total number of data points in the historical time period; Compute the covariance matrix: ， Where, is the covariance matrix, which represents the covariance between various monitoring indicators; Represents a vector transpose operation; Calculate the Mahalanobis distance between the real-time multidimensional monitoring indicator matrix and the historical multidimensional monitoring indicator mean matrix: ， Where, represents the real-time multi-dimensional monitoring indicator matrix, express and The Mahalanobis distance between Represents the covariance matrix The inverse matrix of .

5. The method according to claim 4, wherein The policy network of the reinforcement learning model is trained through training samples generated by the interaction between the agent and the environment. Its optimization direction is determined by the goal defined by the reward function, where the reward function is expressed as follows: ， Where, Is the reward function, indicating that in the current state The reward generated by the state transition action; represents the Mahalanobis distance term, represents the multi-scale difference entropy term, represents the sampling frequency deviation term; ， Where, Indicates the current status The corresponding Mahalanobis distance is Indicates the next state The corresponding Mahalanobis distance is represents the Mahalanobis distance reward correction coefficient; ， Where, Indicates the current status The corresponding multi-scale difference entropy value, Indicates the next state The corresponding multi-scale difference entropy value, Represents the multi-scale difference entropy reward correction coefficient; ， Where, represents the ideal sampling frequency indicated by the label of the training sample, Indicates the current status The corresponding acquisition frequency, Indicates the next state The corresponding acquisition frequency, Indicates the sampling frequency deviation correction factor.

6. The method according to claim 1 or 5, wherein: The reinforcement learning model adopts a deep Q network.

7. A data acquisition system based on frequency adaptive regulation, comprising: an original indicator acquisition unit, configured to acquire a monitoring indicator collection value set, wherein the monitoring indicator collection value set includes a plurality of collection moments obtained based on a first collection frequency and corresponding multi-dimensional monitoring indicator collection values; A multi-scale difference entropy extraction unit is used to extract the local time series data of the monitoring indicator collection value set under each type of sliding window, and calculate the scale difference entropy value corresponding to each of the local time series data, and generate a multi-scale difference entropy value by fusion; each type of sliding window has a corresponding time scale; a Mahalanobis distance calculation unit for extracting a real-time multidimensional monitoring indicator matrix corresponding to a collection moment closest to the current time from the monitoring indicator collection value set, and calculating a Mahalanobis distance between the real-time multidimensional monitoring indicator matrix and a historical multidimensional monitoring indicator mean matrix; the historical multidimensional monitoring indicator mean matrix represents the mean value of the multidimensional monitoring indicator matrix at each collection moment within a preset historical time period; a reinforcement learning optimization unit, configured to combine the first acquisition frequency, the Mahalanobis distance, and the multi-scale difference entropy value into a state vector, and input the state vector into a reinforcement learning model; The action space of the reinforcement learning model includes three types of discrete actions: an increase sampling frequency action for indicating the frequency increase amplitude, a decrease sampling frequency action for indicating the frequency decrease amplitude, and a maintain sampling frequency action for indicating that the acquisition frequency remains unchanged; the optimal action is selected as the target action based on the state vector by the reinforcement learning model; An acquisition frequency adjustment unit is configured to determine a second acquisition frequency according to the first acquisition frequency and acquisition frequency adjustment information indicated by the target action.

Images