Adaptive Hierarchical Clustering Method for Satellite Telemetry Data Based on Included Angle dtw Distance
A technology of satellite telemetry data and hierarchical clustering, applied in instruments, character and pattern recognition, computer parts, etc., can solve the problem of inappropriate time series similarity measurement, and achieve the effect of asynchronous measurement
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
specific Embodiment approach 1
[0032] A self-adaptive hierarchical clustering method for satellite telemetry data based on the included angle DTW distance in this embodiment, such as figure 1 Shown in the flowchart, the clustering method is implemented through the following steps:
[0033] Step 1. Segment the historical satellite telemetry data according to the periodic characteristics of the satellite telemetry data, and obtain the original time series X without class labels 0 ={x 01 ,x 02 ,...,x 0n}; where, n is a positive integer greater than 0, indicating the number of time series;
[0034] Step 2. According to the characteristics of the original time series, introduce time variables, re-express the original time series without class labels, and obtain a time series set based on numerical and time representation X={X 1 ,X 2 ,...,X n};in,
[0035] x i ={(x i1 ,t i1 ),(x i2 ,t i2 ),…,(x im ,t im )}, representing the i-th sequence in the time series set X represented by numerical value and ti...
specific Embodiment approach 2
[0044] The difference from Embodiment 1 is that in the method for adaptive hierarchical clustering of satellite telemetry data based on the angle DTW distance in this embodiment, the process of calculating the angle DTW distance between sequence members described in step 4 is that DTW is a Similarity measures for better matching mapping of time series morphology by bending the time axis. It was first applied to processing speech data, and was later used by Berndt and Clifford to measure time series similarity. Since then, DTW has been widely used in the field of time series data mining. DTW in two time series X' i and X' j Find the optimal curved path between them to get the minimum distance metric. Using the angle distance to measure the similarity of time series can effectively reflect the dynamic characteristics of time series. However, if the time series that needs to be measured has a small offset on the time scale, or a certain expansion in the time length, the measu...
specific Embodiment approach 3
[0047] Different from the specific embodiment one or two, in the self-adaptive hierarchical clustering method of satellite telemetry data based on the included angle DTW distance in this embodiment, the number of categories described in step 6 is j', and the weighted inter-class angle DTW distance and The ratio of the weighted intra-class included angle DTW distance, that is, the inter-class intra-class distance ratio R J The process of (j') is, because the calculation of the cluster center is involved in the error sum of squares criterion, and this process cannot be replaced by other means, so it is not suitable for the adaptive method that adopts the similarity measurement method that does not satisfy the triangle inequality Clustering situation; the weighted average square distance and the criterion evaluation intra-class distance, the smaller the value, the higher the clustering quality; the weighted inter-class distance and the criterion evaluation inter-class distance, th...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More