CN113536042B - Time series abnormity detection method, device and equipment - Google Patents
Time series abnormity detection method, device and equipment Download PDFInfo
- Publication number
- CN113536042B CN113536042B CN202110853832.8A CN202110853832A CN113536042B CN 113536042 B CN113536042 B CN 113536042B CN 202110853832 A CN202110853832 A CN 202110853832A CN 113536042 B CN113536042 B CN 113536042B
- Authority
- CN
- China
- Prior art keywords
- data
- time
- sequence
- energy
- time sequence
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Testing And Monitoring For Control Systems (AREA)
- Debugging And Monitoring (AREA)
Abstract
The invention discloses a method, a device and equipment for detecting time sequence abnormity, wherein the method comprises the following steps: acquiring a time sequence and an energy sequence of the time sequence; carrying out stabilization operation processing on the energy sequence to obtain a processing result; and determining whether the time sequence is abnormal or not according to the statistical values of the data points to be detected and the target time sequence data in the processing result. Through the mode, the generality and the calculation performance of the correlation algorithm in the time series anomaly detection field are improved, a single data point is not abnormal, and a section of data mode is combined to be used for anomaly detection, so that the method is obviously improved.
Description
Technical Field
The present invention relates to the field of information processing technologies, and in particular, to a method, an apparatus, and a device for detecting time series anomalies.
Background
The time sequence data is the most common observation index in the field of operation and maintenance, and the abnormal detection refers to the judgment of abnormal points in data at equal intervals by using an algorithm, so that the foundation is laid for subsequent fault location and other processing.
The traditional time series abnormity detection method can be divided into three categories: a statistical class method, a machine learning class method, and a time series decomposition class method. Statistical methods are of limited applicability to data types. The parameter adjusting difficulty of the machine learning algorithm is high. The time series decomposition algorithm usually faces a mutation point problem and a periodic prediction problem in the decomposition process, and a trend fitting error and a periodic fitting error are easily attributed to noise, so that the algorithm result is influenced. Time series decomposition type algorithms typically decompose data into different orthogonal components, which are modeled separately for anomaly detection. Generally, the time series decomposition idea decomposes data into three parts of trend, periodicity and noise, and trains on the basis of data with a certain length so as to predict data in a future point time, and then compares the predicted value with an observed value so as to judge whether an abnormal point exists or not. The anomaly detection algorithm has strong interpretability, the applicable data types are more than those of statistical methods, but the calculation performance is poor, and the large-scale production environment is difficult to meet; such methods typically have more algorithm parameters, creating challenges for the user. Based on the characteristics of the methods, the Tima algorithm has the characteristics of multiple applicable data types, high operation speed, good algorithm effect and calculation resource saving.
However, in the practical process, the algorithm has a very good effect on single-point abnormality in time series abnormality detection, but the effect is often poor when a single data point is not abnormal and a section of data mode is combined to be abnormal.
Disclosure of Invention
The invention provides a method, a device and equipment for detecting time series abnormity. The method solves the problems of poor universality and poor calculation performance of a correlation algorithm based on the time series anomaly detection field in the prior art.
In order to solve the technical problems, the technical scheme of the invention is as follows:
a time series abnormality detection method includes:
acquiring a time sequence and an energy sequence of the time sequence;
carrying out stabilization operation processing on the energy sequence to obtain a processing result;
and determining whether the time sequence is abnormal or not according to the statistical values of the data points to be detected and the target time sequence data in the processing result.
Optionally, obtaining the energy sequence of the time sequence includes:
and performing square operation on each point in the time sequence to obtain an energy sequence of the time sequence.
Optionally, the smoothing operation processing is performed on the energy sequence to obtain a processing result, and the processing result includes:
and removing the trend data and/or the periodic data from the energy sequence to obtain a processing result.
Optionally, performing a trend data removing operation on the energy sequence, including:
performing trend data removal operation processing on the energy sequence through a formula G (t) = diff (g (t));
wherein, g (t) is data after removing trend data, diff represents difference operation, and g (t) is energy sequence.
Optionally, the removing operation processing of the periodic data on the energy sequence includes:
and taking the first preset time length as the periodic time window length, taking the average value of the data in one time window length of the energy sequence as the periodic numerical value of the next point outside the time window to remove, and taking the data in one period as a sliding window by using the data of the second preset time length until the periodic data of all the points are removed, wherein the second preset time length is less than the first preset time length.
Optionally, in the process of performing the periodic data removal operation on the energy sequence, the method further includes:
and for target data with the variance outside N times of standard deviation, calculating the mean value in the time window by replacing the target data with the mean value of all data in the time window of the target data, wherein N is a positive integer.
Optionally, the target time-series data includes: the data point to be detected in the current time period and the data point in the same time period as the current time period in at least one preset time period;
determining whether the time sequence is abnormal according to the statistical value of the data point to be detected and the target time sequence data in the processing result, wherein the determining step comprises the following steps:
obtaining a statistic of the target time series data;
and if the data point to be detected falls outside the upper and lower boundary ranges of the statistical result, determining that the time sequence is abnormal.
An embodiment of the present invention further provides a time series abnormality detection apparatus, including:
the acquisition module is used for acquiring a time sequence and an energy sequence of the time sequence;
the processing module is used for carrying out stabilization operation processing on the energy sequence to obtain a processing result; and determining whether the time sequence is abnormal or not according to the statistical values of the data points to be detected and the target time sequence data in the processing result.
An embodiment of the present invention further provides an electronic device, including: a processor, a memory storing a computer program which, when executed by the processor, performs the time series anomaly detection method as described above.
Embodiments of the present invention also provide a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the time-series anomaly detection method as described above.
The scheme of the invention at least comprises the following beneficial effects:
obtaining a time sequence and an energy sequence of the time sequence; carrying out stabilization operation processing on the energy sequence to obtain a processing result; and determining whether the time sequence is abnormal or not according to the statistical values of the data points to be detected and the target time sequence data in the processing result. The universality and the calculation performance of a correlation algorithm in the time series anomaly detection field are improved, a single data point is not abnormal, and a section of data mode is combined to be used for anomaly detection.
The foregoing description is only an overview of the technical solutions of the embodiments of the present invention, and the embodiments of the present invention can be implemented according to the content of the description in order to make the technical means of the embodiments of the present invention more clearly understood, and the detailed description of the embodiments of the present invention is provided below in order to make the foregoing and other objects, features, and advantages of the embodiments of the present invention more clearly understandable.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the embodiments of the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
FIG. 1 is a flow chart of a time series anomaly detection method provided by an embodiment of the invention;
fig. 2 is a diagram illustrating an anomaly detection result of the time-series anomaly detection method according to the embodiment of the present invention on a real data;
fig. 3 is a schematic structural diagram of a time-series abnormality detection apparatus according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a computing device provided in an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
As shown in fig. 1, an embodiment of the present invention provides a time series anomaly detection method, including:
and step 13, determining whether the time sequence is abnormal or not according to the statistical value of the data point to be detected and the target time sequence data in the processing result.
In the time series abnormality detection method according to this embodiment, a time series and an energy series of the time series are obtained; carrying out stabilization operation processing on the energy sequence to obtain a processing result; and determining whether the time sequence is abnormal or not according to the statistical values of the data points to be detected and the target time sequence data in the processing result. The universality and the calculation performance of the correlation algorithm in the time series anomaly detection field are improved, and the detection of the anomaly is obviously promoted when a single data point is not abnormal and a section of data mode is combined.
In an alternative embodiment of the present invention, step 11 comprises: and performing square operation on each point in the time sequence to obtain an energy sequence of the time sequence.
In this embodiment, it is assumed that given a time series f (t), it has a noise of ε (t) and follows a normal distribution. Then, a square operation is performed on each point in the time series f (t), and the transformed time series g (t) is regarded as the energy series of the original time series, i.e., g (t) = f (t). Time series exceptions are generally described as three classes, single point exceptions, contextual exceptions, and continuity exceptions. Considering that data point anomalies in a real-world signal are usually single point anomalies, the signal-to-noise ratio can be improved by using the energy sequence g (t).
In an alternative embodiment of the present invention, step 12 comprises: and removing the trend data and/or the periodic data from the energy sequence to obtain a processing result.
In this embodiment, after the energy data g (t) is obtained, the data is smoothed. The removal of trend data and periodic data is mainly performed. After the removal, the smoothing operation of the energy sequence is completed, and the smoothed data is denoted as s (t).
In an optional embodiment of the present invention, in step 12, performing a trend data removing operation on the energy sequence includes: performing trend data removal operation processing on the energy sequence through a formula G (t) = diff (g (t)); wherein, g (t) is data after removing trend data, diff represents difference operation, and g (t) is energy sequence.
In this embodiment, the trend data is removed by using mathematical difference operation, the difference operation can effectively record the information of the mutation point, express the mutation point as a point with a larger difference value, prepare data for the development of subsequent anomaly detection, and record g (t) as the data after trend removal, there are: g (t) = diff (g (t)), where diff denotes a differential operation.
In another optional embodiment of the present invention, in step 12, the removing operation of the periodic data on the energy sequence includes:
and taking a first preset time as a periodic time window length, taking the average value of data in one time window length of the energy sequence as a periodic numerical value of the next point outside the time window for removal, and applying data with a second preset time to the data in one period to make a sliding window until the periodicity of all the points is removed, wherein the second preset time is shorter than the first preset time.
In this embodiment, in the periodic component of the data, in consideration of the data characteristics of a general service, the first preset duration is the periodic time window length, the average value in one time window length is removed as the periodic numerical value of the next point outside the window, and for the data in one period, the data with the second preset duration is used as a sliding window until the periodicity of all the points is removed. The first preset time is usually 7 days, the second preset time is usually 1 day, the first preset time and the second preset time can be changed according to actual conditions, but the second preset time is always smaller than the first preset time.
In another optional embodiment of the present invention, in step 12, during the process of removing the periodic data from the energy sequence, the method may further include:
and for target data with the variance outside N times of standard deviation, calculating the mean value in the time window by replacing the target data with the mean value of all data in the time window of the target data, wherein N is a positive integer.
In this embodiment, during the processing of removing the periodic data by the energy sequence, the existence of the outlier may affect the effect of the removal period, so for the data whose variance is outside N times of the standard deviation, the mean value in the window is calculated by replacing the mean value in the window with the value, and thus the influence of the outlier is eliminated. N here may preferably be 3; the moving average also plays a role in noise reduction while removing periodic portions. This completes the smoothing operation of the energy sequence.
The smoothed data is denoted as s (t), and there are: s (t) = g (t) = ma (g (t)), g (t) = diff (g (t)), where ma denotes a moving average operation, and diff denotes a difference operation.
In yet another alternative embodiment of the present invention, the target time-series data includes: a data point to be detected in a current time period and a data point in at least one preset time period in the same time period as the current time period; step 13 may include:
obtaining a statistical value of the target time series data;
specifically, assuming that data has a daily cycle and a weekly cycle, a data point to be detected in the processing result, a data point of the past same time period and a data point of the corresponding time period of the last week are put together to form a group of new time series data, the time series data is a target time series, K-sigma method statistics is carried out on the time series data on the basis, the data point of the target time series needs to obey normal distribution, and then a mean value and a variance are calculated to obtain a statistical value of the target time series data.
And if the data point to be detected falls outside the upper and lower boundary ranges of the statistical result, determining that the time sequence is abnormal.
In this embodiment, after the statistical value of the target time series data is obtained, abnormality detection is performed on the data point to be detected in the processing result. If the data point to be detected falls outside the upper and lower boundary ranges of the K-sigma statistical result, the data point to be detected is regarded as abnormal; and if the data point to be detected falls within the upper and lower boundary ranges of the K-sigma statistical result, the data point is normal.
As shown in fig. 2, an abnormality detection result diagram of the time series abnormality detection method provided by the embodiment of the present invention on some real data shows that the time series abnormality detection method provided by the embodiment of the present invention is very effective through the determination result.
In the method, the mathematical ideas of data stabilization and statistics are fully combined, the data are not required to be decomposed into different orthogonal parts, and the data are modeled respectively to carry out anomaly detection. The embodiment of the invention firstly starts from the idea of time series decomposition, and performs the smoothing operation on the data by combining the difference and the moving average from the energy point of view; secondly, carrying out K-sigma method statistics on the stabilized data and the target time sequence data, determining whether the time sequence is abnormal, and finally, regarding the abnormal point of the stabilized data as the abnormal corresponding point in the original data. Because the energy-based differential smoothing technology is applied, the situation that the trend fitting error and the period fitting error are imposed on noise is avoided, and the problem caused by the abrupt change point is avoided to a great extent by the differential. Compared with other algorithms, the method has the advantages that the calculation time is greatly reduced, and the requirements on resources such as internal memory are low. The time series abnormality detection method has considerable universality and calculation performance. From the perspective of being applicable to data types, regarding single-point anomaly, the method disclosed by the invention is not only applicable to a stable time sequence, but also applicable to data with trend and periodicity, and the method disclosed by the invention is also characterized in that the method can adapt to mutation points. Compared with the Tima algorithm, the method can effectively detect the condition that a single data point in the time sequence is not abnormal and a section of data mode is combined to be abnormal, thereby being an effective supplement of the Tima algorithm. Meanwhile, the method is suitable for more data types and has the characteristic that other mainstream methods do not have, so that the method omits the processes of classifying time sequences and matching different algorithms and different types of data. The method has better operability in the actual landing process.
Fig. 3 is a schematic diagram illustrating a time-series abnormality detection apparatus according to an embodiment of the present invention. As shown in fig. 3, the apparatus includes:
an obtaining module 31, configured to obtain a time series and an energy series of the time series;
the processing module 32 is configured to perform a stabilizing operation on the energy sequence to obtain a processing result; and determining whether the time sequence is abnormal or not according to the statistical values of the data points to be detected and the target time sequence data in the processing result.
Optionally, the obtaining module 31 is configured to perform a square operation on each point in the time series to obtain an energy series of the time series.
Optionally, the processing module 32 is specifically configured to perform removing operation processing on the trend data and/or the periodic data on the energy sequence to obtain a processing result.
Optionally, the processing module 32 is further configured to perform trend data removing operation processing on the energy sequence through a formula g (t) = diff (g (t));
wherein, g (t) is data after removing trend data, diff represents difference operation, and g (t) is energy sequence.
Optionally, the processing module 32 is further configured to use a first preset time duration as a periodic time window duration, use an average value of data in one time window duration of the energy sequence as a periodic numerical value of a next point outside the time window to remove, and use data of a second preset time duration as a sliding window for the data in one period until the periodic data of all the points are removed, where the second preset time duration is smaller than the first preset time duration.
Optionally, in the process of performing the operation of removing the periodic data on the energy sequence, the method further includes:
and for target data with the variance outside N times of standard deviation, calculating the mean value in the time window by replacing the target data with the mean value of all data in the time window of the target data, wherein N is a positive integer.
Optionally, the target time-series data includes: the data point to be detected in the current time period and the data point in the same time period as the current time period in at least one preset time period;
the processing module 32 is further configured to obtain a statistic of the target time-series data; and if the data point to be detected falls outside the upper and lower boundary ranges of the statistical result, determining that the time sequence is abnormal.
It should be noted that this embodiment is an apparatus embodiment corresponding to the above method embodiment, and all the implementations in the above method embodiment are applicable to this apparatus embodiment, and the same technical effects can be achieved.
The embodiment of the invention provides a nonvolatile computer storage medium, wherein at least one executable instruction is stored in the computer storage medium, and the computer executable instruction can execute the time sequence abnormity detection method in any method embodiment.
Fig. 4 is a schematic structural diagram of a computing device according to an embodiment of the present invention, and the specific embodiment of the present invention does not limit the specific implementation of the computing device.
As shown in fig. 4, the computing device may include: a processor (processor), a Communications Interface (Communications Interface), a memory (memory), and a Communications bus.
Wherein: the processor, the communication interface, and the memory communicate with each other via a communication bus. A communication interface for communicating with network elements of other devices, such as clients or other servers. The processor is used for executing a program, and specifically can execute relevant steps in the time series abnormality detection method embodiment for the computing device.
In particular, the program may include program code comprising computer operating instructions.
The processor may be a central processing unit CPU or an application Specific Integrated circuit asic or one or more Integrated circuits configured to implement embodiments of the present invention. The computing device includes one or more processors, which may be the same type of processor, such as one or more CPUs; or may be different types of processors such as one or more CPUs and one or more ASICs.
And the memory is used for storing programs. The memory may comprise high-speed RAM memory, and may also include non-volatile memory (non-volatile memory), such as at least one disk memory.
The program may specifically be configured to cause the processor to execute the time-series abnormality detection method in any of the above-described method embodiments. For specific implementation of each step in the program, reference may be made to corresponding steps and corresponding descriptions in units in the foregoing time series abnormality detection method embodiments, which are not described herein again. It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described devices and modules may refer to the corresponding process descriptions in the foregoing method embodiments, and are not described herein again.
The algorithms or displays presented herein are not inherently related to any particular computer, virtual system, or other apparatus. Various general purpose systems may also be used with the teachings herein. The required structure for constructing such a system will be apparent from the description above. In addition, embodiments of the present invention are not directed to any particular programming language. It is appreciated that a variety of programming languages may be used to implement the teachings of embodiments of the present invention as described herein, and any descriptions of specific languages are provided above to disclose the best modes of embodiments of the invention.
In the description provided herein, numerous specific details are set forth. It is understood, however, that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.
Those skilled in the art will appreciate that the modules in the devices in an embodiment may be adaptively changed and arranged in one or more devices different from the embodiment. The modules or units or components of the embodiments may be combined into one module or unit or component, and furthermore they may be divided into a plurality of sub-modules or sub-units or sub-components. All of the features disclosed in this specification (including any accompanying abstract and drawings, etc.), and all of the processes or elements of any method or apparatus so disclosed, may be combined in any combination, except combinations where at least some of such features and/or processes or elements are mutually exclusive. Each feature disclosed in this specification (including the accompanying abstract and drawings), may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.
Furthermore, those skilled in the art will appreciate that while some embodiments herein include some features included in other embodiments, rather than other features, combinations of features of different embodiments are meant to be within the scope of the invention and form different embodiments.
The various component embodiments of the invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art will appreciate that a microprocessor or Digital Signal Processor (DSP) may be used in practice to implement some or all of the functionality of some or all of the components according to embodiments of the present invention. Embodiments of the present invention may also be embodied as device or system programs (e.g., computer programs and computer program products) for performing a portion or all of the methods described herein. Such programs implementing embodiments of the present invention may be stored on computer-readable media or may be in the form of one or more signals. Such a signal may be downloaded from an internet website or provided on a carrier signal or in any other form.
Claims (10)
1. A time series abnormality detection method is characterized by comprising:
acquiring a time sequence and an energy sequence of the time sequence, wherein each point in the time sequence is subjected to square operation to obtain the energy sequence of the time sequence;
carrying out stabilizing operation processing on the energy sequence to obtain a processing result, wherein the processing result comprises the following steps: removing operation processing of trend data and periodic data is carried out on the energy sequence to obtain a processing result, and removing operation processing of the periodic data is carried out on the energy sequence, and the removing operation processing comprises the following steps: taking a first preset time length as a periodic time window length, taking a data average value in one time window length of the energy sequence as a periodic numerical value of a next point outside the time window to remove, and using data with a second preset time length as a sliding window for the data in one period until the periodic data of all the points are removed, wherein the second preset time length is less than the first preset time length;
and determining whether the time sequence is abnormal according to the statistical values of the data points to be detected and the target time sequence data in the processing result, wherein the K-sigma method statistics is carried out on the target time sequence data, and then the mean value and the variance are calculated to obtain the statistical value of the target time sequence data.
2. The time-series abnormality detection method according to claim 1, characterized in that:
a time sequence f (t) with noise epsilon (t) and obeying normal distribution, then taking a square operation on each point in the time sequence f (t), and regarding the transformed time sequence g (t) as an energy sequence of the original time sequence, namely g (t)(t)＝f2(t)。
3. The time-series abnormality detection method according to claim 1, characterized in that:
the target time-series data follows a normal distribution.
4. The method according to claim 1, wherein performing a trend data removal operation on the energy series includes:
performing trend data removal operation processing on the energy sequence through a formula G (t) ═ diff (g (t));
wherein, g (t) is data after removing trend data, diff represents difference operation, and g (t) is energy sequence.
5. The time-series abnormality detection method according to claim 4, characterized in that:
the smoothed data is denoted as s (t), and there are: s (t) ═ g (t) — ma (g (t)), and g (t) ═ diff (g (t)), where ma denotes a moving average operation and diff denotes a difference operation.
6. The method according to claim 5, wherein the step of performing periodic data removal operation on the energy sequence further comprises:
and for target data with the variance outside N times of standard deviation, calculating the mean value in the time window by replacing the target data with the mean value of all data in the time window of the target data, wherein N is a positive integer.
7. The time-series abnormality detection method according to claim 1, characterized in that the target time-series data includes: the data point to be detected in the current time period and the data point in the same time period as the current time period in at least one preset time period;
determining whether the time sequence is abnormal according to the statistical value of the data point to be detected and the target time sequence data in the processing result, wherein the determining step comprises the following steps:
obtaining a statistical value of the target time series data;
and if the data point to be detected falls outside the upper and lower boundary ranges of the statistical value, determining that the time sequence is abnormal.
8. A time-series abnormality detection device characterized by comprising:
the acquisition module is used for acquiring a time sequence and an energy sequence of the time sequence, wherein each point in the time sequence is subjected to square operation to obtain the energy sequence of the time sequence;
the processing module is used for carrying out stabilization operation processing on the energy sequence to obtain a processing result, and comprises: removing operation processing of trend data and periodic data is carried out on the energy sequence to obtain a processing result, and removing operation processing of the periodic data is carried out on the energy sequence, and the removing operation processing comprises the following steps: taking a first preset time length as a periodic time window length, taking a data average value in one time window length of the energy sequence as a periodic numerical value of a next point outside the time window to remove, and using data with a second preset time length as a sliding window for the data in one period until the periodic data of all the points are removed, wherein the second preset time length is less than the first preset time length; and determining whether the time sequence is abnormal according to the statistical values of the data points to be detected and the target time sequence data in the processing result, wherein the K-sigma method statistics is carried out on the target time sequence data, and then the mean value and the variance are calculated to obtain the statistical value of the target time sequence data.
9. An electronic device, comprising: processor, memory storing a computer program which, when executed by the processor, performs the time series anomaly detection method according to any one of claims 1 to 7.
10. A computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the time series anomaly detection method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110853832.8A CN113536042B (en) | 2021-07-28 | 2021-07-28 | Time series abnormity detection method, device and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110853832.8A CN113536042B (en) | 2021-07-28 | 2021-07-28 | Time series abnormity detection method, device and equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113536042A CN113536042A (en) | 2021-10-22 |
| CN113536042B true CN113536042B (en) | 2022-06-21 |
Family
ID=78121096
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110853832.8A Active CN113536042B (en) | 2021-07-28 | 2021-07-28 | Time series abnormity detection method, device and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113536042B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115310049B (en) * | 2022-10-09 | 2023-10-27 | 云智慧（北京）科技有限公司 | Method, device and equipment for detecting period of time sequence data |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109697207A (en) * | 2018-12-25 | 2019-04-30 | 苏州思必驰信息科技有限公司 | The abnormality monitoring method and system of time series data |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015095974A1 (en) * | 2013-12-27 | 2015-07-02 | Metafor Software Inc. | System and method for anomaly detection in information technology operations |
| US10628435B2 (en) * | 2017-11-06 | 2020-04-21 | Adobe Inc. | Extracting seasonal, level, and spike components from a time series of metrics data |
| CN110032670B (en) * | 2019-04-17 | 2022-11-29 | 腾讯科技（深圳）有限公司 | Method, device and equipment for detecting abnormity of time sequence data and storage medium |
| CN110213125A (en) * | 2019-05-23 | 2019-09-06 | 南京维拓科技股份有限公司 | Abnormality detection system based on time series data under a kind of cloud environment |
| US11132342B2 (en) * | 2019-12-02 | 2021-09-28 | Alibaba Group Holding Limited | Periodicity detection and period length estimation in time series |
| CN111145895B (en) * | 2019-12-24 | 2023-10-20 | 中国科学院深圳先进技术研究院 | Abnormal data detection method and terminal equipment |
-
2021
- 2021-07-28 CN CN202110853832.8A patent/CN113536042B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109697207A (en) * | 2018-12-25 | 2019-04-30 | 苏州思必驰信息科技有限公司 | The abnormality monitoring method and system of time series data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113536042A (en) | 2021-10-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2021164267A1 (en) | Anomaly detection method and apparatus, and terminal device and storage medium | |
| WO2016155241A1 (en) | Method, system and computer device for capacity prediction based on kalman filter | |
| CN113536042B (en) | Time series abnormity detection method, device and equipment | |
| CN112506752A (en) | Operation and maintenance method, device and equipment based on time series data trend prediction | |
| CN115453356B (en) | Power equipment operation state monitoring and analyzing method, system, terminal and medium | |
| CN112631881A (en) | Operation and maintenance method, device and equipment based on time series data trend prediction | |
| CN115122155A (en) | Machine tool remote diagnosis method and system based on industrial internet big data | |
| CN115373861B (en) | GPU resource scheduling method and device, electronic equipment and storage medium | |
| CN108723895B (en) | Signal segmentation method for real-time monitoring of drilling machining state | |
| CN113297195B (en) | Time series abnormity detection method, device and equipment | |
| CN109142979B (en) | Method and device for detecting abnormal state of power distribution network | |
| CN104794548A (en) | ARIMA (Autoregressive integrated moving average) model load prediction based parallelization computing method | |
| CN117596191A (en) | Power Internet of things abnormality detection method, device and storage medium | |
| CN112906967A (en) | Desulfurization system slurry circulating pump performance prediction method and device | |
| Fu et al. | SPC methods for nonstationary correlated count data with application to network surveillance | |
| CN114172708A (en) | Method for identifying network flow abnormity | |
| CN114818813A (en) | Cutter wear value prediction algorithm based on self-attention convolution neural network | |
| CN114692987A (en) | Time sequence data analysis method, device, equipment and storage medium | |
| CN115358314A (en) | Method, device and equipment for training wind power prediction model and predicting power | |
| CN112162528B (en) | Fault diagnosis method, device, equipment and storage medium of numerical control machine tool | |
| CN109298999B (en) | Core software testing method and device based on data distribution characteristics | |
| CN110488772B (en) | DCS centralized monitoring method and device and centralized monitoring terminal | |
| CN114121711A (en) | Detection method, device, medium and electronic equipment for wafer production process | |
| TWI720388B (en) | Apparatus, method and computer program for processing a piecewise-smooth signal | |
| US10929177B2 (en) | Managing resources for multiple trial distributed processing tasks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |