2020
DOI: 10.1002/cjce.23778
|View full text |Cite
|
Sign up to set email alerts
|

Density peaks clustering‐based steady/transition mode identification and monitoring of multimode processes

Abstract: Multimode is the characteristic of industrial manufacturing processes due to different production strategies and environments. For multimode process monitoring, it is a challenge to identify different steady modes and transition modes. In this paper, a k nearest neighbours (KNN)‐based density peaks clustering (DPC) method is applied to identify different modes. First, the local density of each sample, which is obtained with a KNN constraint and its minimum distance to the higher local density points are calcul… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
4
1

Citation Types

0
6
0

Year Published

2021
2021
2023
2023

Publication Types

Select...
7

Relationship

2
5

Authors

Journals

citations
Cited by 18 publications
(6 citation statements)
references
References 34 publications
0
6
0
Order By: Relevance
“…. , s T N ] T ∈ R N×M is obtained according to equation (2), and it is normalized as S. In the procedure of SVDD modeling, S is firstly mapped from the original space to a higher feature space by a nonlinear transformation function Φ(s i ). For conventional SVDD, all the training data has the same impact on the model construction, which makes it insensible to outliers and data density.…”
Section: Mlrdr Weighted Svddmentioning
confidence: 99%
See 1 more Smart Citation
“…. , s T N ] T ∈ R N×M is obtained according to equation (2), and it is normalized as S. In the procedure of SVDD modeling, S is firstly mapped from the original space to a higher feature space by a nonlinear transformation function Φ(s i ). For conventional SVDD, all the training data has the same impact on the model construction, which makes it insensible to outliers and data density.…”
Section: Mlrdr Weighted Svddmentioning
confidence: 99%
“…e operation conditions of industrial processes will inevitably change with diverse customer requirements, setpoints variation, and different intrinsic features, which leads to multiple modes [1][2][3]. Because of the complexity of multimodal process, it is difficult to obtain satisfactory monitoring results.…”
Section: Introductionmentioning
confidence: 99%
“…[30] A k-nearest neighbours (kNN)-based DPC method was proposed and applied to multimode process monitoring. [31] Most related DPC methods adopted in the process fault detection are used for condition identification and division. In the perspective of data distribution, due to the unique nature of cluster centres selection and clustering manner, DPC can be regarded as a data distribution information acquisition process.…”
Section: Introductionmentioning
confidence: 99%
“…Due to various market demands, drifting disturbance and fluctuating system settings, industrial manufacturing processes are inevitably operated under different working conditions, which are called multimode processes [1,2]. In such processes, there are several stable working points with different statistical characteristics, hence their data do not obey Gaussian distributions, which is the premise of the traditional statistical process monitoring schemes [3].…”
Section: Introductionmentioning
confidence: 99%