Time sequence data arrangement method, system and equipment of power equipment and medium

Abstract

The invention discloses a time sequence data arrangement method and system of power equipment, equipment and a medium. The method comprises the following steps: acquiring monitoring time sequence data of the power equipment; extracting statistical features and structural features of the power equipment from the monitoring time sequence data based on a sliding time window, calculating a time sequence feature difference degree between a current time window and a historical time window based on the statistical features and the structural features, and if the time sequence feature difference degree is greater than a self-adaptive threshold value, determining that the power equipment is in a self-adaptive state. If yes, a target arrangement strategy is obtained through matching from a preset arrangement strategy library according to the time sequence characteristic difference degree, the monitoring time sequence data is arranged based on the target arrangement strategy, and a time sequence data arrangement result of the power equipment is obtained, and the adaptive threshold is adaptively generated according to the statistical characteristics and the structural characteristics, so that the accuracy of rearrangement of the monitoring time sequence data of the power equipment can be improved by implementing the method and the device.

Description

Technical Field

[0001] This invention relates to the field of power systems, and more particularly to a method, system, device, and medium for arranging time-series data for power equipment. Background Technology

[0002] Time-series data from power equipment monitoring is the core basis for power grid operation status perception, fault early warning, and power analysis. However, in actual data acquisition, such data often suffers from problems such as timestamp errors, missing data points, and noise interference due to communication interference (e.g., abnormal SIM card signals), equipment transients, and heterogeneous multi-source systems. Unprocessed raw sequences will destroy the inherent temporal correlation and structural consistency of the data. If directly used for subsequent calculations or modeling, it will lead to serious consequences such as misjudgment of status, distortion of energy efficiency analysis, and delayed early warning.

[0003] Existing technologies typically employ orchestration methods based on fixed thresholds and static rules: that is, fluctuation thresholds (such as upper limits of variance) or simple rules (such as fixed time windows) are manually preset. When data exceeds the threshold, a pre-defined single processing strategy (such as linear interpolation or fixed-frequency resampling) is triggered. These methods are ill-suited to the complex and ever-changing operating conditions of power equipment, and cannot accurately identify abnormal patterns. Furthermore, static strategies cannot distinguish between noise and sudden changes in actual operating conditions, easily leading to mis-orchestration (such as over-smoothing of real fluctuations) or omissions in orchestration (such as failure to handle implicit timing inaccuracies). Summary of the Invention

[0004] This invention provides a method, system, device, and medium for arranging time-series data of power equipment, which can improve the accuracy of re-arranging time-series data of power equipment monitoring.

[0005] An embodiment of the present invention provides a method for time-series data arrangement of power equipment, comprising: Acquire monitoring time-series data of power equipment; Based on a sliding time window, statistical and structural features of power equipment are extracted from the monitoring time-series data. Based on the statistical and structural features, the time-series feature difference degree between the current time window and the historical time window is calculated. If the time-series feature difference degree is greater than an adaptive threshold, a target orchestration strategy is matched from a preset orchestration strategy library according to the time-series feature difference degree, and the monitoring time-series data is orchestrated based on the target orchestration strategy to obtain the time-series data orchestration result of the power equipment. The adaptive threshold is adaptively generated based on the statistical and structural features.

[0006] This invention, through simultaneous extraction of statistical and structural characteristics of data, constructs a more comprehensive and refined feature profile of monitoring data. Compared to methods using only a single type of feature, this dual-dimensional feature extraction can more accurately capture the inherent patterns and structural anomalies in data caused by signal interruptions, equipment operating condition switching, etc., providing a reliable perceptual basis for subsequent accurate decision-making. By quantifying the deviation between current data features and historical normal / stable state features, it replaces the subjective mode of relying on human experience to judge whether data is "abnormal" or "needs processing" in traditional methods, achieving precise quantitative identification of data deviation states. This provides an accurate and objective judgment basis for triggering re-arrangement, avoiding false triggers or missed triggers. By setting an adaptive threshold... The value can be dynamically adjusted based on the real-time statistical distribution characteristics of the data and the inherent operating standards of the power system, ensuring that the "threshold" for triggering re-arrangement matches the current actual operating scenario and data quality. By matching the most suitable re-arrangement strategy according to the specific difference pattern, it ensures that the arranged actions taken can most effectively correct the specific types of data defects currently identified, thereby directly improving the pertinence and accuracy of a single re-arrangement action. By calling the strategy that is precisely matched to the problem to process the data, it can most efficiently and directly repair the identified problems in the data, thereby outputting an optimized data sequence that is more accurate in terms of timing logic and numerical rationality, providing higher quality data input for downstream applications such as power analysis and status assessment.

[0007] Furthermore, the extraction of statistical and structural features of power equipment from the monitoring time-series data based on a sliding time window includes: The monitoring time series data is cleaned and preprocessed with unified timestamps to obtain the processing result; Based on a preset sliding time window, data segments within the current time window are sequentially extracted from the processing results, and the statistical characteristics of the data segments are calculated. The statistical characteristics include one or more combinations of the following: mean, variance, standard deviation, and rate of change. Calculate the structural features of the data segment.

[0008] By simultaneously extracting the statistical and structural characteristics of the data, a more comprehensive and detailed feature profile of the monitoring data is constructed. Compared with methods that use only a single type of feature, this dual-dimensional feature extraction can more accurately capture the inherent patterns and structural anomalies in the data caused by signal interruption, equipment operating condition switching, etc., providing a reliable perceptual basis for subsequent accurate decision-making.

[0009] Further, calculating the structural features of the data segment includes: The number of peaks in the data segment that exceed a preset amplitude threshold is counted, where the peaks are local extreme points; The data segment is divided into several equal-width intervals, and the distribution concentration of data points within each equal-width interval is calculated to determine the data density. Calculate the difference between adjacent data points in the data segment to determine continuity features based on the difference; The structural features of the power equipment are determined based on the number of peak points, the data density, and the continuity characteristics.

[0010] Further, the step of calculating the temporal feature difference degree between the current time window and the historical time window based on the statistical features and the structural features includes: The statistical features and structural features of the current time window are combined to form the current feature vector; Calculate the weighted feature difference between the current feature vector and the historical feature vector corresponding to the historical time window to obtain the time-series feature difference.

[0011] By quantifying the degree of deviation between the current data characteristics and the historical normal / stable state characteristics, this replaces the subjective mode of relying on human experience to judge whether the data is "abnormal" or "needs processing" in traditional methods. It achieves accurate quantitative identification of data deviation state, provides accurate and objective judgment basis for triggering re-arrangement, and avoids false triggering or missed triggering.

[0012] Further, the step of matching the target orchestration strategy from a preset orchestration strategy library based on the temporal feature difference includes: The orchestration strategy library is pre-built, and the orchestration strategy library includes at least four of the following: dynamic window resampling strategy, abnormal data smoothing rearrangement strategy, heterogeneous data temporal alignment strategy, and key node retention rearrangement strategy. Based on the structural features, feature change patterns are determined, and target orchestration strategies are selected or combined from the orchestration strategy library based on the feature change patterns and the temporal feature difference degree.

[0013] By matching the most suitable re-arrangement strategy based on the specific difference pattern, this ensures that the arranged actions can most effectively correct the specific types of data defects currently identified, thereby directly improving the targeting and accuracy of a single re-arrangement action. Further, the step of selecting or combining a target orchestration strategy from the orchestration strategy library based on the feature change pattern and the temporal feature difference includes: If the number of peak points exceeds the first set threshold, the abnormal data smoothing rearrangement strategy is matched first; if not applicable, the key node retention rearrangement strategy is matched. When the data density is lower than the second set threshold, the dynamic window resampling strategy is matched first; if it is not applicable, the abnormal data smoothing rearrangement strategy is matched. When the continuity feature is lower than the third set threshold, the heterogeneous data time-series alignment strategy is matched first; if it is not applicable, the dynamic window resampling strategy is matched. If the feature change pattern meets the preset conditions, then two or more orchestration strategies selected from the orchestration strategy library are combined to obtain the target orchestration strategy.

[0014] Furthermore, the step of orchestrating the monitoring time-series data based on the target orchestration strategy to obtain the time-series data orchestration result of the power equipment further includes: receiving the performance feedback index output after executing the time-series data orchestration result, and dynamically adjusting the window parameters of the sliding time window based on the performance feedback index.

[0015] Another embodiment of the present invention provides a timing data orchestration system for power equipment, comprising: The acquisition module is used to acquire monitoring time-series data of power equipment; The orchestration module is used to extract statistical and structural features of power equipment from the monitoring time-series data based on a sliding time window. Based on the statistical and structural features, it calculates the time-series feature difference degree between the current time window and the historical time window. If the time-series feature difference degree is greater than an adaptive threshold, it matches a target orchestration strategy from a preset orchestration strategy library according to the time-series feature difference degree, and orchestrates the monitoring time-series data based on the target orchestration strategy to obtain the time-series data orchestration result of the power equipment. The adaptive threshold is adaptively generated based on the statistical and structural features.

[0016] Another embodiment of the present invention provides a terminal device, including: a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the steps of the timing data arrangement method for power equipment of the present invention.

[0017] Another embodiment of the present invention provides a computer-readable storage medium item, including: a stored computer program, which, when the computer program is running, controls the device where the computer-readable storage medium is located to perform steps such as the timing data arrangement method of the power equipment of the present invention. Attached Figure Description

[0018] To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0019] Figure 1 This is a flowchart illustrating an embodiment of the timing data arrangement method for power equipment provided in this application; Figure 2 This is a flowchart illustrating one embodiment of steps S201 to S203 provided in this application; Figure 3 This is a flowchart illustrating one embodiment of steps S301 to S302 provided in this application; Figure 4 This is a schematic diagram of an embodiment of the timing data orchestration system for power equipment provided in this application. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.

[0022] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly defined.

[0023] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0024] In the description of the embodiments in this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.

[0025] In the description of the embodiments of this application, the term "multiple" refers to two or more (including two), similarly, "multiple sets" refers to two or more (including two sets), and "multiple pieces" refers to two or more (including two pieces).

[0026] In the description of the embodiments of this application, unless otherwise expressly specified and limited, technical terms such as "installation," "connection," "joining," and "fixing" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. For those skilled in the art, the specific meaning of the above terms in the embodiments of this application can be understood according to the specific circumstances.

[0027] Time-series data from power equipment monitoring is a core basis for power grid operation status perception, fault early warning, and power analysis. However, in actual data acquisition, this type of data often suffers from various problems due to interference, thus requiring time-series data arrangement and processing. Existing technologies typically employ arrangement methods based on fixed thresholds and static rules, which are difficult to adapt to the complex and ever-changing operating conditions of power equipment, resulting in low arrangement accuracy.

[0028] See Figure 1 To improve the accuracy of time-series data re-arrangement for power equipment monitoring, an embodiment of the present invention provides a method for time-series data arrangement for power equipment, including steps S101 to S102. Step S101: Obtain monitoring time series data of power equipment; In some embodiments, multi-source heterogeneous monitoring time-series data are received from the power equipment monitoring system. This data may specifically originate from various sensors installed on the power equipment (such as voltage, current, and temperature sensors), files reported by the SCADA (Supervisory Control and Data Acquisition) system, and L-files recording the equipment's operating status.

[0029] Step S102: Extract statistical and structural features of power equipment from the monitoring time series data based on a sliding time window. Calculate the time series feature difference degree between the current time window and historical time windows based on the statistical and structural features. If the time series feature difference degree is greater than an adaptive threshold, match a target orchestration strategy from a preset orchestration strategy library based on the time series feature difference degree, and orchestrate the monitoring time series data based on the target orchestration strategy to obtain the time series data orchestration result of the power equipment. The adaptive threshold is adaptively generated based on the statistical and structural features.

[0030] Please refer to Figure 2 In some embodiments, the step of extracting statistical and structural features of power equipment from the monitoring time-series data based on a sliding time window includes steps S201 to S203: Step S201: Perform data cleaning and timestamp unification preprocessing on the monitoring time series data to obtain the processing result; In some embodiments, to ensure data quality and consistency, after acquiring the monitoring time-series data, the data is first cleaned, including interpolating and filling missing values ​​and filtering out obvious noise to eliminate data errors introduced by abnormal acquisition or transmission interference, thus obtaining the cleaned results. Secondly, the timestamp format and precision of the cleaned results are standardized to ensure that all data are aligned under the same time base, providing standardized input data for subsequent feature extraction based on time windows.

[0031] Step S202: Based on a preset sliding time window, sequentially extract data segments within the current time window from the processing results, and calculate the statistical characteristics of the data segments, wherein the statistical characteristics include one or more combinations of the following: mean, variance, standard deviation, and rate of change; In some embodiments, after preprocessing is completed, the system sequentially extracts data segments within the current window through a preset sliding time window. By advancing segment by segment through the sliding window, the system can continuously acquire the statistical characteristics of time series data within a local time range, including mean, variance, standard deviation, and rate of change.

[0032] It should be noted that these statistical characteristics can reflect the central tendency, dispersion, and dynamic changes of the data within a window period. Among them, the mean represents the average level of the data, the variance and standard deviation reflect the magnitude of data fluctuations, and the rate of change reflects the degree of drastic change of the data over time.

[0033] Step S203: Calculate the structural features of the data segment.

[0034] In some embodiments, step S203 includes: counting the number of peaks in the data segment that exceed a preset amplitude threshold, wherein the peaks are local extrema; dividing the data segment into several equal-width intervals and calculating the distribution concentration of data points in each equal-width interval to determine the data density; calculating the difference between adjacent data points in the data segment to determine continuity features based on the difference; and determining the structural features of the power equipment based on the number of peaks, the data density, and the continuity features.

[0035] In some embodiments, firstly, the time series data points within the current sliding window are... (where i = 2, 3, ..., n−1) is traversed, excluding the first and last points to avoid boundary errors, and peak value determination is performed. When this point is a local maximum, At this time, minute fluctuations can be filtered out, and points that meet the above conditions are counted to obtain the number of effective peaks within the current window. The calculation formula is as follows: In the formula, This represents the number of peak points. As an adaptive threshold, it is typically set to 2-3 times the standard deviation of the data; This is an indicator function; its value is 1 if the condition is true, and 0 otherwise. Next, the data within the current window is divided into K equal-width intervals, and the number of data points in each interval is counted. And calculate the data density D, the calculation formula is: Then, for the data points within the window... (where i = 2, 3, ..., n−1), calculate the first-order difference sequence: And based on the difference sequence Calculate its standard deviation And based on the standard deviation, a continuity index is determined, calculated using the following formula: , among which, if A smaller value indicates that the data changes smoothly and has good continuity. A larger value for C indicates drastic data fluctuations and poor continuity; the closer C is to 1, the better the continuity. Finally, the calculated number of peak points... The structural characteristics of the power equipment can be determined by summing the data density D and the continuity feature C.

[0036] It should be noted that the number of local peaks characterizes the degree of fluctuation in time-series data within a sliding window, reflecting potential sudden states or abnormal events during power equipment operation. Data density assesses the degree of concentration of time-series data within a numerical range, reflecting whether the data exhibits clustering or dispersion trends, and is suitable for identifying data distribution drift. Temporal continuity measures the degree of smoothness of data changes over time, reflecting the stability of the power equipment's operating state.

[0037] By simultaneously extracting the statistical and structural characteristics of the data, a more comprehensive and detailed feature profile of the monitoring data is constructed. Compared with methods that use only a single type of feature, this dual-dimensional feature extraction can more accurately capture the inherent patterns and structural anomalies in the data caused by signal interruption, equipment operating condition switching, etc., providing a reliable perceptual basis for subsequent accurate decision-making.

[0038] In some embodiments, calculating the temporal feature difference degree between the current time window and the historical time window based on the statistical features and the structural features includes: combining the statistical features and the structural features of the current time window to form a current feature vector; and calculating the weighted feature difference degree between the current feature vector and the historical feature vector corresponding to the historical time window to obtain the temporal feature difference degree. Specifically, firstly, the statistical features and structural features extracted within the current sliding time window are combined in a predetermined order to form a multidimensional current feature vector. Historical feature vectors It is constructed in a similar manner. Then, weighted Euclidean distance or weighted cosine similarity is used to calculate the weighted feature difference between the current feature vector and the historical feature vectors. The calculation formula is as follows: ,in, These are the weighting coefficients for each feature.

[0039] It should be noted that this invention also determines whether data distribution drift or scene switching has occurred. If potential changes in data characteristics are detected, such as changes in the operating status of power equipment causing data characteristics to change over time, an adaptive re-orchestration strategy adjustment is triggered. This avoids performing unnecessary complex processing on stable data and ensures that the re-orchestration strategy matches the current data characteristics. Specifically, this can be achieved by calculating the feature change rate. The calculation formula is: ,in, The weighted feature difference of the current window. The weighted feature difference of the historical window, when the feature difference rate changes across multiple consecutive windows. If the threshold T1 is exceeded, a distribution shift is determined to have occurred. If the rate of change of feature difference across multiple consecutive windows is... If the threshold T2 is exceeded, a scene switch is determined to have occurred. The thresholds T1 and T2 can be set as needed, and this application does not impose any restrictions.

[0040] By quantifying the degree of deviation between the current data characteristics and the historical normal / stable state characteristics, this replaces the subjective mode of relying on human experience to judge whether the data is "abnormal" or "needs processing" in traditional methods. It achieves accurate quantitative identification of data deviation state, provides accurate and objective judgment basis for triggering re-arrangement, and avoids false triggering or missed triggering.

[0041] Please refer to Figure 3 In some embodiments, the step of matching the target orchestration strategy from a preset orchestration strategy library based on the temporal feature difference includes steps S301 to S302: Step S301: The orchestration strategy library is pre-built, and the orchestration strategy library includes at least four of the following: dynamic window resampling strategy, abnormal data smoothing rearrangement strategy, heterogeneous data temporal alignment strategy, and key node retention rearrangement strategy. In some embodiments, firstly, based on the feature extraction results (i.e., statistical features and structural features), a real-time threshold is generated using an adaptive algorithm (such as a dynamic thresholding method based on statistical distribution) to replace the manually preset threshold. If the temporal feature difference is greater than this adaptive threshold, a matching operation is triggered.

[0042] In some embodiments, to achieve adaptive re-orchestration of time series data, an orchestration strategy library containing various basic processing logics needs to be pre-built. The orchestration strategy library is a predefined, callable set of strategies designed to cover common quality problems and orchestration requirements of time series data. Specifically, the library includes at least the following four core strategies: 1) Dynamic window resampling strategy, whose core function is to dynamically adjust the sampling frequency or aggregation time window based on the data change rate, suitable for scenarios with uneven data density or where data granularity needs to be changed; 2) Outlier smoothing and rearrangement strategy, which uses filtering algorithms (such as moving average, median filtering) or statistical outlier correction methods to identify and process noise and outliers in the data to improve the smoothness of the sequence; 3) Heterogeneous data time-series alignment strategy, which is used to process data from different sources with different collection times or timestamp precisions, aligning multi-source data on the time axis through timestamp interpolation, normalization, or event-based synchronization algorithms; 4) Key node retention and rearrangement strategy, which aims to identify and retain important data points in the sequence (such as local extrema, inflection points that meet specific business rules), avoiding the loss of key information during noise reduction or compression. When building this strategy library, each strategy is encapsulated as an independent module with clear inputs, outputs, and adjustable parameters, providing a foundation for subsequent intelligent selection and invocation. Step S302: Determine the feature change pattern based on the structural features, and select or combine the target orchestration strategy from the orchestration strategy library based on the feature change pattern and the temporal feature difference degree.

[0043] In some embodiments, when structural features are calculated, it is necessary to determine the feature change pattern. For example, when the number of local peaks increases significantly, i.e., exceeds a first threshold dynamically calculated based on historical data, it is determined to be a peak surge pattern; when the data density index shows that the data distribution becomes dispersed, i.e., the concentration is lower than a dynamically generated second threshold, it is determined to be a density dispersion pattern; when the time series continuity index drops sharply, i.e., the variance of the difference sequence exceeds a third threshold, it is determined to be a continuity disruption pattern. The system has a built-in preset pattern-policy mapping rule table, which defines the preferred and alternative policies corresponding to each feature change pattern. The pattern-policy mapping rule table is shown in the table below: By matching the most suitable re-arrangement strategy based on the specific difference pattern, this ensures that the arranged actions can most effectively correct the specific types of data defects currently identified, thereby directly improving the targeting and accuracy of a single re-arrangement action. In some embodiments, selecting or combining orchestration strategies from the orchestration strategy library based on the feature change pattern and the temporal feature difference degree to form a target orchestration strategy includes: if the number of peak points exceeds a first preset threshold, preferentially matching the abnormal data smoothing rearrangement strategy; if not applicable, matching the key node retention rearrangement strategy; when the data density is lower than a second preset threshold, preferentially matching the dynamic window resampling strategy; if not applicable, matching the abnormal data smoothing rearrangement strategy; when the continuity feature is lower than a third preset threshold, preferentially matching the heterogeneous data temporal alignment strategy; if not applicable, matching the dynamic window resampling strategy; if the feature change pattern meets preset conditions, combining two or more orchestration strategies selected from the orchestration strategy library to obtain the target orchestration strategy. Specifically, for peak surge patterns, the system preferentially matches the abnormal data smoothing rearrangement strategy to suppress abnormal fluctuations; if the smoothing effect is not ideal or business requirements require retaining peaks, the key node retention rearrangement strategy is enabled as an alternative. Secondly, the temporal feature difference degree, as a scalar of overall change magnitude, is used for decision triggering and strategy combination determination. The strategy adjustment is triggered only when the feature difference exceeds the adaptively generated global threshold. If the difference is extremely high or multiple structural features are detected to change significantly at the same time (i.e., the preset multi-feature composite change conditions are met), the system will select the preferred strategy based on the main change pattern and may combine it with other related strategies (e.g., perform temporal alignment and smooth rearrangement simultaneously).

[0044] It should be noted that feature variation needs to be calculated during matching. And based on the feature change, a strategy is used to select scores. ,in, To select weights, choose the strategy or strategy combination with the highest score.

[0045] In some embodiments, the monitoring time series data is arranged based on the target arrangement strategy to obtain the time series data arrangement result of the power equipment. Specifically, after the target arrangement strategy is selected, its internal parameters (such as the window size of resampling and the intensity coefficient of the smoothing algorithm) will be adaptively fine-tuned according to the real-time statistical characteristics of the current window (such as the rate of change and the noise level) to ensure the accurate matching between the arrangement action and the data characteristics.

[0046] In some embodiments, the process of orchestrating the monitoring time-series data based on the target orchestration strategy to obtain the time-series data orchestration result of the power equipment further includes: receiving performance feedback indicators output after executing the time-series data orchestration result, and dynamically adjusting the window parameters of the sliding time window based on the performance feedback indicators. Specifically, firstly, performance feedback indicators output after the execution of downstream tasks are received, such as the smoothness of the time-series data curve (which can be measured by the angle change between adjacent data points), the accuracy of the downstream modeling task, processing latency, etc. These indicators are used to evaluate the actual effect of the current re-orchestration strategy and parameters. The performance feedback indicators are analyzed. If the feedback indicators do not meet the preset target (such as insufficient smoothness or decreased task accuracy), a parameter adjustment mechanism is triggered to dynamically optimize the window parameters of the sliding time window, including but not limited to window size and sliding step size. At the same time, the weight allocation in feature extraction and the threshold setting in the strategy selection logic can also be iteratively adjusted according to the feedback results.

[0047] This invention, through simultaneous extraction of statistical and structural characteristics of data, constructs a more comprehensive and refined feature profile of monitoring data. Compared to methods using only a single type of feature, this dual-dimensional feature extraction can more accurately capture the inherent patterns and structural anomalies in data caused by signal interruptions, equipment operating condition switching, etc., providing a reliable perceptual basis for subsequent accurate decision-making. By quantifying the deviation between current data features and historical normal / stable state features, it replaces the subjective mode of relying on human experience to judge whether data is "abnormal" or "needs processing" in traditional methods, achieving precise quantitative identification of data deviation states. This provides an accurate and objective judgment basis for triggering re-arrangement, avoiding false triggers or missed triggers. By setting an adaptive threshold... The value can be dynamically adjusted based on the real-time statistical distribution characteristics of the data and the inherent operating standards of the power system, ensuring that the "threshold" for triggering re-arrangement matches the current actual operating scenario and data quality. By matching the most suitable re-arrangement strategy according to the specific difference pattern, it ensures that the arranged actions taken can most effectively correct the specific types of data defects currently identified, thereby directly improving the pertinence and accuracy of a single re-arrangement action. By calling the strategy that is precisely matched to the problem to process the data, it can most efficiently and directly repair the identified problems in the data, thereby outputting an optimized data sequence that is more accurate in terms of timing logic and numerical rationality, providing higher quality data input for downstream applications such as power analysis and status assessment.

[0048] like Figure 4 As shown, based on the above method embodiments, corresponding apparatus embodiments are provided; An embodiment of the present invention provides a timing data orchestration system for power equipment, comprising: The acquisition module 100 is used to acquire monitoring time-series data of power equipment; The orchestration module 200 is used to extract statistical and structural features of power equipment from the monitoring time-series data based on a sliding time window. Based on the statistical and structural features, it calculates the time-series feature difference degree between the current time window and the historical time window. If the time-series feature difference degree is greater than an adaptive threshold, it matches a target orchestration strategy from a preset orchestration strategy library according to the time-series feature difference degree, and orchestrates the monitoring time-series data based on the target orchestration strategy to obtain the time-series data orchestration result of the power equipment. The adaptive threshold is adaptively generated based on the statistical and structural features.

[0049] It is understood that the above-described device embodiments correspond to the method embodiments of the present invention, and can implement the timing data arrangement method for power equipment provided by any of the above-described method embodiments of the present invention.

[0050] It should be noted that the device embodiments described above are merely illustrative, and some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can specifically be implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0051] Based on the above-described embodiments of the timing data orchestration method for power equipment, another embodiment of the present invention provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the timing data orchestration method for power equipment according to any embodiment of the present invention.

[0052] For example, in this embodiment, the computer program can be divided into one or more modules, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the terminal device.

[0053] The terminal device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0054] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the terminal device, connecting all parts of the terminal device via various interfaces and lines.

[0055] Based on the above-described method embodiments, another embodiment of the present invention provides a computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to execute the timing data arrangement method for power equipment described in any of the above-described method embodiments of the present invention.

[0056] The modules / units integrated in the device / terminal equipment, if implemented as software functional units and sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0057] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for arranging time-series data of power equipment, characterized in that, include: Acquire monitoring time-series data of power equipment; Based on a sliding time window, statistical and structural features of power equipment are extracted from the monitoring time-series data. Based on the statistical and structural features, the time-series feature difference degree between the current time window and the historical time window is calculated. If the time-series feature difference degree is greater than an adaptive threshold, a target orchestration strategy is matched from a preset orchestration strategy library according to the time-series feature difference degree, and the monitoring time-series data is orchestrated based on the target orchestration strategy to obtain the time-series data orchestration result of the power equipment. The adaptive threshold is adaptively generated based on the statistical and structural features.

2. The method for arranging time-series data of power equipment according to claim 1, characterized in that, The extraction of statistical and structural features of power equipment from the monitoring time-series data based on a sliding time window includes: The monitoring time series data is cleaned and preprocessed with unified timestamps to obtain the processing result; Based on a preset sliding time window, data segments within the current time window are sequentially extracted from the processing results, and the statistical characteristics of the data segments are calculated. The statistical characteristics include one or more combinations of the following: mean, variance, standard deviation, and rate of change. Calculate the structural features of the data segment.

3. The method for arranging time-series data of power equipment according to claim 2, characterized in that, The calculation of the structural features of the data segment includes: The number of peaks in the data segment that exceed a preset amplitude threshold is counted, where the peaks are local extreme points; The data segment is divided into several equal-width intervals, and the distribution concentration of data points within each equal-width interval is calculated to determine the data density. Calculate the difference between adjacent data points in the data segment to determine continuity features based on the difference; The structural features of the power equipment are determined based on the number of peak points, the data density, and the continuity characteristics.

4. The method for arranging time-series data of power equipment according to claim 1, characterized in that, The step of calculating the temporal feature difference degree between the current time window and the historical time window based on the statistical features and the structural features includes: The statistical features and structural features of the current time window are combined to form the current feature vector; Calculate the weighted feature difference between the current feature vector and the historical feature vector corresponding to the historical time window to obtain the time-series feature difference.

5. The method for arranging time-series data of power equipment according to claim 3, characterized in that, The step of matching the target orchestration strategy from a preset orchestration strategy library based on the temporal feature difference includes: The orchestration strategy library is pre-built, and the orchestration strategy library includes at least four of the following: dynamic window resampling strategy, abnormal data smoothing rearrangement strategy, heterogeneous data temporal alignment strategy, and key node retention rearrangement strategy. Based on the structural features, feature change patterns are determined, and target orchestration strategies are selected or combined from the orchestration strategy library based on the feature change patterns and the temporal feature difference degree.

6. The method for arranging time-series data of power equipment according to claim 5, characterized in that, The step of selecting or combining a target orchestration strategy from the orchestration strategy library based on the feature change pattern and the temporal feature difference includes: If the number of peak points exceeds the first set threshold, the abnormal data smoothing rearrangement strategy is matched first; if not applicable, the key node retention rearrangement strategy is matched. When the data density is lower than the second set threshold, the dynamic window resampling strategy is matched first; if it is not applicable, the abnormal data smoothing rearrangement strategy is matched. When the continuity feature is lower than the third set threshold, the heterogeneous data time-series alignment strategy is matched first; if it is not applicable, the dynamic window resampling strategy is matched. If the feature change pattern meets the preset conditions, then two or more orchestration strategies selected from the orchestration strategy library are combined to obtain the target orchestration strategy.

7. The method for arranging time-series data of power equipment according to any one of claims 1-6, characterized in that, The process of orchestrating the monitoring time-series data based on the target orchestration strategy to obtain the time-series data orchestration result of the power equipment further includes: receiving the performance feedback index output after executing the time-series data orchestration result, and dynamically adjusting the window parameters of the sliding time window based on the performance feedback index.

8. A timing data arrangement system for power equipment, characterized in that, include: The acquisition module is used to acquire monitoring time-series data of power equipment; The orchestration module is used to extract statistical and structural features of power equipment from the monitoring time-series data based on a sliding time window. Based on the statistical and structural features, it calculates the time-series feature difference degree between the current time window and the historical time window. If the time-series feature difference degree is greater than an adaptive threshold, it matches a target orchestration strategy from a preset orchestration strategy library according to the time-series feature difference degree, and orchestrates the monitoring time-series data based on the target orchestration strategy to obtain the time-series data orchestration result of the power equipment. The adaptive threshold is adaptively generated based on the statistical and structural features.

9. A terminal device, characterized in that, include: One or more processors; A memory, coupled to the processor, for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the steps of the timing data orchestration method for power equipment as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, include: A stored computer program, wherein, when the computer program is executed, it controls the device containing the computer-readable storage medium to perform the steps of the timing data orchestration method for power equipment as described in any one of claims 1-7.

Images