Research on Pattern Recognition Method of Blockage Signal in Pipeline Based on LMD Information Entropy and ELM

Table 1

Partial feature vector extracted based on LMD information entropy and EMD information entropy.

Blocking types	LMD information entropy				EMD information entropy
	A	B	C	D	A	B	C	D
Health	1.226	0.627	0.485	0.758	1.443	1.118	1.386	0.633
	1.207	0.630	0.498	0.772	1.477	1.157	1.366	0.659
	1.119	0.628	0.565	0.760	1.510	1.038	1.395	0.614
	⋯	⋯	⋯	⋯	⋯	⋯	⋯	⋯
Slight blockage	1.085	0.558	0.436	0.713	1.303	1.177	0.924	0.735
	1.131	0.579	0.429	0.711	1.309	1.023	0.945	0.740
	1.141	0.573	0.436	0.712	1.323	1.027	0.949	0.753
	⋯	⋯	⋯	⋯	⋯	⋯	⋯	⋯
Medium blockage	1.046	0.411	0.386	0.682	1.102	0.985	1.314	0.673
	1.012	0.527	0.362	0.708	1.098	1.101	1.137	0.670
	0.919	0.430	0.393	0.706	1.103	1.053	1.232	0.681
	⋯	⋯	⋯	⋯	⋯	⋯	⋯	⋯
Serious blockage	1.058	0.572	0.313	0.625	1.263	1.105	1.265	0.685
	1.004	0.519	0.295	0.673	1.321	1.128	1.287	0.693
	0.986	0.517	0.332	0.631	1.315	1.160	1.365	0.676
	⋯	⋯	⋯	⋯	⋯	⋯	⋯	⋯