A new cable fault location method based on
wavelet reconstruction is proposed. In this method the
difference between the currents of faulty phase and sound
phase under the high voltage pulse excitation is used as the
measured signal and is decomposed in multi-scale by wavelet
transform, then reconstructed in single scale. Comparing with
traditional fault location method by travelling wave, the
presented method will not be interfered by the reflected wave
from the branch joint of cables or from other positions where
the impedances are not matched and not be influenced by fault
types, otherwise, the reflected waves can be recognized even
the faulty position is near to the measuring terminal, at the
same time, the influence of the wave speed uncertainty can be
reduced. The correctness of the proposed method is proved by
simulation results.
幾篇最經(jīng)典的運(yùn)動(dòng)搜索算法的IEEE論文：A1. Enhanced Hexagonal Search for Fast Block Motion Estimation；A2. Hexagon-based search pattern for fast block motion estimation；A3. Predictive Motion Vector Field Adaptive Search Technique (PMVFAST)等
How the K-mean Cluster work
Step 1. Begin with a decision the value of k = number of clusters
Step 2. Put any initial partition that classifies the data into k clusters. You may assign the training samples randomly, or systematically as the following:
Take the first k training sample as single-element clusters
Assign each of the remaining (N-k) training sample to the cluster with the nearest centroid. After each assignment, recomputed the centroid of the gaining cluster.
Step 3 . Take each sample in sequence and compute its distance from the centroid of each of the clusters. If a sample is not currently in the cluster with the closest centroid, switch this sample to that cluster and update the centroid of the cluster gaining the new sample and the cluster losing the sample.
Step 4 . Repeat step 3 until convergence is achieved, that is until a pass through the training sample causes no new assignments.
