Time series data processing method and device

Abstract

The invention discloses a time series data processing method and device, the method comprising: acquiring time series data, each data in the time series is a value of a time point; sequentially calculating the fluctuation energy E(n )=V(n)‑V(n‑1), where E(n) is the fluctuation energy at the current time point, V(n) is the value at the current time point, and V(n‑1) is the previous time point The value of ; loop through each time point in the time series, determine the state of the time point according to the fluctuation energy of each time point, the state includes: a temporary state and a stable state; the stable state of the time point As the state mode at the time point; the state mode includes: a steady mode, a rising mode, and a falling mode. By using the invention, the segmentation of time series data and the identification of state patterns can be realized simply and accurately.

Description

technical field [0001] The invention relates to the field of data processing, in particular to a time series data processing method and device. Background technique [0002] Time-series data refers to data collected in chronological order. This type of data reflects the state or degree of a certain thing, phenomenon, etc. over time. It can be intuitively expressed as a time-series waveform in a plane, such as an electrocardiogram (ECG ), electroencephalogram (EEG), current and voltage signals in manufacturing, K-lines for stock trading, time-domain waveforms of speech signals, etc. In the analysis and application of such data, it is usually necessary to segment the waveform and classify the segmented waveforms to provide corresponding information for various applications. [0003] In the existing technology, there are many ways to segment time series data, but most of the existing technologies have various limitations or defects, such as: high computational complexity, inac...

AI Technical Summary

Problems solved by technology

[0003] In the existing technology, there are many ways to segment time series data, but most of the existing technologies have various limitations or defects, such as: high computational complexity, inaccurate segmentation in complex scenes, and inapplicability to Real-time computing, highly affected by noisy data, strong assumptions about the distribution or source of data, supervised learning requires training data and labels, or is only applicable to a signal with a specific graphic pattern, etc.

Images