Analysis of smearing-out in contribution plot based fault isolation for Statistical Process Control
This paper studies the smearing effect encountered in contribution plot based fault isolation, i.e., the influence of faulty variables on the contributions of non-faulty variables. Since the generation of contribution plots requires no a priori information about the detected disturbance (e.g., histo...
Ausführliche Beschreibung
Gespeichert in:
| Autor*in: |
Van den Kerkhof, Pieter [verfasserIn] |
|---|
| Format: |
E-Artikel |
|---|---|
| Sprache: |
Englisch |
| Erschienen: |
2013transfer abstract |
|---|
| Schlagwörter: |
|---|
| Umfang: |
9 |
|---|
| Übergeordnetes Werk: |
Enthalten in: Plasticity in responses to dimensional variations of soil space in 19 grassland plant species - Dong, Ran ELSEVIER, 2022, Amsterdam [u.a.] |
|---|---|
| Übergeordnetes Werk: |
volume:104 ; year:2013 ; day:18 ; month:12 ; pages:285-293 ; extent:9 |
| Links: |
|---|
| DOI / URN: |
10.1016/j.ces.2013.08.007 |
|---|
| Katalog-ID: |
ELV027491331 |
|---|
| LEADER | 01000caa a22002652 4500 | ||
|---|---|---|---|
| 001 | ELV027491331 | ||
| 003 | DE-627 | ||
| 005 | 20230625151959.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 180603s2013 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1016/j.ces.2013.08.007 |2 doi | |
| 028 | 5 | 2 | |a GBVA2013019000009.pica |
| 035 | |a (DE-627)ELV027491331 | ||
| 035 | |a (ELSEVIER)S0009-2509(13)00550-2 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 082 | 0 | |a 660 | |
| 082 | 0 | 4 | |a 660 |q DE-600 |
| 082 | 0 | 4 | |a 570 |a 630 |q VZ |
| 084 | |a BIODIV |q DE-30 |2 fid | ||
| 100 | 1 | |a Van den Kerkhof, Pieter |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a Analysis of smearing-out in contribution plot based fault isolation for Statistical Process Control |
| 264 | 1 | |c 2013transfer abstract | |
| 300 | |a 9 | ||
| 336 | |a nicht spezifiziert |b zzz |2 rdacontent | ||
| 337 | |a nicht spezifiziert |b z |2 rdamedia | ||
| 338 | |a nicht spezifiziert |b zu |2 rdacarrier | ||
| 520 | |a This paper studies the smearing effect encountered in contribution plot based fault isolation, i.e., the influence of faulty variables on the contributions of non-faulty variables. Since the generation of contribution plots requires no a priori information about the detected disturbance (e.g., historical faulty data), it is a popular fault isolation technique in Statistical Process Control (SPC). However, Westerhuis et al. (2000) demonstrated that contributions suffer from fault smearing. As a consequence, variables unaffected by the fault may be highlighted and faulty variables obscured during the contribution analysis. This paper presents a thorough analysis of the smearing effect for three general contribution computation methods: complete decomposition, partial decomposition and reconstruction-based contributions. The analysis shows that (i) smearing is present in all three methods, (ii) smearing depends on the chosen number of principal components of the underlying PCA or PLS model and (iii) the extent of smearing increases for variables correlated in the training data for a well-chosen model order. The effect of smearing on the isolation performance of single and multiple sensor faults of various magnitudes is studied and illustrated using a simulation case study. The results indicate that correct isolation with contribution plots is not guaranteed for multiple sensor faults. Furthermore, contribution plots only outperform univariate fault isolation for single sensor faults with small magnitudes. For multiple sensor faults, univariate fault isolation exhibits a significantly larger correct fault isolation rate. Based on the smearing analysis and the specific results for sensor faults, the authors advise to use contributions only if a sound physical interpretation of the principal components is available. Otherwise multivariate detection followed by univariate fault isolation is recommended. | ||
| 520 | |a This paper studies the smearing effect encountered in contribution plot based fault isolation, i.e., the influence of faulty variables on the contributions of non-faulty variables. Since the generation of contribution plots requires no a priori information about the detected disturbance (e.g., historical faulty data), it is a popular fault isolation technique in Statistical Process Control (SPC). However, Westerhuis et al. (2000) demonstrated that contributions suffer from fault smearing. As a consequence, variables unaffected by the fault may be highlighted and faulty variables obscured during the contribution analysis. This paper presents a thorough analysis of the smearing effect for three general contribution computation methods: complete decomposition, partial decomposition and reconstruction-based contributions. The analysis shows that (i) smearing is present in all three methods, (ii) smearing depends on the chosen number of principal components of the underlying PCA or PLS model and (iii) the extent of smearing increases for variables correlated in the training data for a well-chosen model order. The effect of smearing on the isolation performance of single and multiple sensor faults of various magnitudes is studied and illustrated using a simulation case study. The results indicate that correct isolation with contribution plots is not guaranteed for multiple sensor faults. Furthermore, contribution plots only outperform univariate fault isolation for single sensor faults with small magnitudes. For multiple sensor faults, univariate fault isolation exhibits a significantly larger correct fault isolation rate. Based on the smearing analysis and the specific results for sensor faults, the authors advise to use contributions only if a sound physical interpretation of the principal components is available. Otherwise multivariate detection followed by univariate fault isolation is recommended. | ||
| 650 | 7 | |a Process control |2 Elsevier | |
| 650 | 7 | |a Fault detection/isolation |2 Elsevier | |
| 650 | 7 | |a Mathematical modelling |2 Elsevier | |
| 650 | 7 | |a Contribution plots |2 Elsevier | |
| 650 | 7 | |a Chemical processes |2 Elsevier | |
| 650 | 7 | |a Remediation |2 Elsevier | |
| 700 | 1 | |a Vanlaer, Jef |4 oth | |
| 700 | 1 | |a Gins, Geert |4 oth | |
| 700 | 1 | |a Van Impe, Jan F.M. |4 oth | |
| 773 | 0 | 8 | |i Enthalten in |n Elsevier Science |a Dong, Ran ELSEVIER |t Plasticity in responses to dimensional variations of soil space in 19 grassland plant species |d 2022 |g Amsterdam [u.a.] |w (DE-627)ELV008347182 |
| 773 | 1 | 8 | |g volume:104 |g year:2013 |g day:18 |g month:12 |g pages:285-293 |g extent:9 |
| 856 | 4 | 0 | |u https://doi.org/10.1016/j.ces.2013.08.007 |3 Volltext |
| 912 | |a GBV_USEFLAG_U | ||
| 912 | |a GBV_ELV | ||
| 912 | |a SYSFLAG_U | ||
| 912 | |a FID-BIODIV | ||
| 912 | |a SSG-OLC-PHA | ||
| 951 | |a AR | ||
| 952 | |d 104 |j 2013 |b 18 |c 1218 |h 285-293 |g 9 | ||
| 953 | |2 045F |a 660 | ||
| author_variant |
d k p v dkp dkpv |
|---|---|
| matchkey_str |
vandenkerkhofpietervanlaerjefginsgeertva:2013----:nlssfmaiguicnrbtopobsdalioainos |
| hierarchy_sort_str |
2013transfer abstract |
| publishDate |
2013 |
| allfields |
10.1016/j.ces.2013.08.007 doi GBVA2013019000009.pica (DE-627)ELV027491331 (ELSEVIER)S0009-2509(13)00550-2 DE-627 ger DE-627 rakwb eng 660 660 DE-600 570 630 VZ BIODIV DE-30 fid Van den Kerkhof, Pieter verfasserin aut Analysis of smearing-out in contribution plot based fault isolation for Statistical Process Control 2013transfer abstract 9 nicht spezifiziert zzz rdacontent nicht spezifiziert z rdamedia nicht spezifiziert zu rdacarrier This paper studies the smearing effect encountered in contribution plot based fault isolation, i.e., the influence of faulty variables on the contributions of non-faulty variables. Since the generation of contribution plots requires no a priori information about the detected disturbance (e.g., historical faulty data), it is a popular fault isolation technique in Statistical Process Control (SPC). However, Westerhuis et al. (2000) demonstrated that contributions suffer from fault smearing. As a consequence, variables unaffected by the fault may be highlighted and faulty variables obscured during the contribution analysis. This paper presents a thorough analysis of the smearing effect for three general contribution computation methods: complete decomposition, partial decomposition and reconstruction-based contributions. The analysis shows that (i) smearing is present in all three methods, (ii) smearing depends on the chosen number of principal components of the underlying PCA or PLS model and (iii) the extent of smearing increases for variables correlated in the training data for a well-chosen model order. The effect of smearing on the isolation performance of single and multiple sensor faults of various magnitudes is studied and illustrated using a simulation case study. The results indicate that correct isolation with contribution plots is not guaranteed for multiple sensor faults. Furthermore, contribution plots only outperform univariate fault isolation for single sensor faults with small magnitudes. For multiple sensor faults, univariate fault isolation exhibits a significantly larger correct fault isolation rate. Based on the smearing analysis and the specific results for sensor faults, the authors advise to use contributions only if a sound physical interpretation of the principal components is available. Otherwise multivariate detection followed by univariate fault isolation is recommended. This paper studies the smearing effect encountered in contribution plot based fault isolation, i.e., the influence of faulty variables on the contributions of non-faulty variables. Since the generation of contribution plots requires no a priori information about the detected disturbance (e.g., historical faulty data), it is a popular fault isolation technique in Statistical Process Control (SPC). However, Westerhuis et al. (2000) demonstrated that contributions suffer from fault smearing. As a consequence, variables unaffected by the fault may be highlighted and faulty variables obscured during the contribution analysis. This paper presents a thorough analysis of the smearing effect for three general contribution computation methods: complete decomposition, partial decomposition and reconstruction-based contributions. The analysis shows that (i) smearing is present in all three methods, (ii) smearing depends on the chosen number of principal components of the underlying PCA or PLS model and (iii) the extent of smearing increases for variables correlated in the training data for a well-chosen model order. The effect of smearing on the isolation performance of single and multiple sensor faults of various magnitudes is studied and illustrated using a simulation case study. The results indicate that correct isolation with contribution plots is not guaranteed for multiple sensor faults. Furthermore, contribution plots only outperform univariate fault isolation for single sensor faults with small magnitudes. For multiple sensor faults, univariate fault isolation exhibits a significantly larger correct fault isolation rate. Based on the smearing analysis and the specific results for sensor faults, the authors advise to use contributions only if a sound physical interpretation of the principal components is available. Otherwise multivariate detection followed by univariate fault isolation is recommended. Process control Elsevier Fault detection/isolation Elsevier Mathematical modelling Elsevier Contribution plots Elsevier Chemical processes Elsevier Remediation Elsevier Vanlaer, Jef oth Gins, Geert oth Van Impe, Jan F.M. oth Enthalten in Elsevier Science Dong, Ran ELSEVIER Plasticity in responses to dimensional variations of soil space in 19 grassland plant species 2022 Amsterdam [u.a.] (DE-627)ELV008347182 volume:104 year:2013 day:18 month:12 pages:285-293 extent:9 https://doi.org/10.1016/j.ces.2013.08.007 Volltext GBV_USEFLAG_U GBV_ELV SYSFLAG_U FID-BIODIV SSG-OLC-PHA AR 104 2013 18 1218 285-293 9 045F 660 |
| spelling |
10.1016/j.ces.2013.08.007 doi GBVA2013019000009.pica (DE-627)ELV027491331 (ELSEVIER)S0009-2509(13)00550-2 DE-627 ger DE-627 rakwb eng 660 660 DE-600 570 630 VZ BIODIV DE-30 fid Van den Kerkhof, Pieter verfasserin aut Analysis of smearing-out in contribution plot based fault isolation for Statistical Process Control 2013transfer abstract 9 nicht spezifiziert zzz rdacontent nicht spezifiziert z rdamedia nicht spezifiziert zu rdacarrier This paper studies the smearing effect encountered in contribution plot based fault isolation, i.e., the influence of faulty variables on the contributions of non-faulty variables. Since the generation of contribution plots requires no a priori information about the detected disturbance (e.g., historical faulty data), it is a popular fault isolation technique in Statistical Process Control (SPC). However, Westerhuis et al. (2000) demonstrated that contributions suffer from fault smearing. As a consequence, variables unaffected by the fault may be highlighted and faulty variables obscured during the contribution analysis. This paper presents a thorough analysis of the smearing effect for three general contribution computation methods: complete decomposition, partial decomposition and reconstruction-based contributions. The analysis shows that (i) smearing is present in all three methods, (ii) smearing depends on the chosen number of principal components of the underlying PCA or PLS model and (iii) the extent of smearing increases for variables correlated in the training data for a well-chosen model order. The effect of smearing on the isolation performance of single and multiple sensor faults of various magnitudes is studied and illustrated using a simulation case study. The results indicate that correct isolation with contribution plots is not guaranteed for multiple sensor faults. Furthermore, contribution plots only outperform univariate fault isolation for single sensor faults with small magnitudes. For multiple sensor faults, univariate fault isolation exhibits a significantly larger correct fault isolation rate. Based on the smearing analysis and the specific results for sensor faults, the authors advise to use contributions only if a sound physical interpretation of the principal components is available. Otherwise multivariate detection followed by univariate fault isolation is recommended. This paper studies the smearing effect encountered in contribution plot based fault isolation, i.e., the influence of faulty variables on the contributions of non-faulty variables. Since the generation of contribution plots requires no a priori information about the detected disturbance (e.g., historical faulty data), it is a popular fault isolation technique in Statistical Process Control (SPC). However, Westerhuis et al. (2000) demonstrated that contributions suffer from fault smearing. As a consequence, variables unaffected by the fault may be highlighted and faulty variables obscured during the contribution analysis. This paper presents a thorough analysis of the smearing effect for three general contribution computation methods: complete decomposition, partial decomposition and reconstruction-based contributions. The analysis shows that (i) smearing is present in all three methods, (ii) smearing depends on the chosen number of principal components of the underlying PCA or PLS model and (iii) the extent of smearing increases for variables correlated in the training data for a well-chosen model order. The effect of smearing on the isolation performance of single and multiple sensor faults of various magnitudes is studied and illustrated using a simulation case study. The results indicate that correct isolation with contribution plots is not guaranteed for multiple sensor faults. Furthermore, contribution plots only outperform univariate fault isolation for single sensor faults with small magnitudes. For multiple sensor faults, univariate fault isolation exhibits a significantly larger correct fault isolation rate. Based on the smearing analysis and the specific results for sensor faults, the authors advise to use contributions only if a sound physical interpretation of the principal components is available. Otherwise multivariate detection followed by univariate fault isolation is recommended. Process control Elsevier Fault detection/isolation Elsevier Mathematical modelling Elsevier Contribution plots Elsevier Chemical processes Elsevier Remediation Elsevier Vanlaer, Jef oth Gins, Geert oth Van Impe, Jan F.M. oth Enthalten in Elsevier Science Dong, Ran ELSEVIER Plasticity in responses to dimensional variations of soil space in 19 grassland plant species 2022 Amsterdam [u.a.] (DE-627)ELV008347182 volume:104 year:2013 day:18 month:12 pages:285-293 extent:9 https://doi.org/10.1016/j.ces.2013.08.007 Volltext GBV_USEFLAG_U GBV_ELV SYSFLAG_U FID-BIODIV SSG-OLC-PHA AR 104 2013 18 1218 285-293 9 045F 660 |
| allfields_unstemmed |
10.1016/j.ces.2013.08.007 doi GBVA2013019000009.pica (DE-627)ELV027491331 (ELSEVIER)S0009-2509(13)00550-2 DE-627 ger DE-627 rakwb eng 660 660 DE-600 570 630 VZ BIODIV DE-30 fid Van den Kerkhof, Pieter verfasserin aut Analysis of smearing-out in contribution plot based fault isolation for Statistical Process Control 2013transfer abstract 9 nicht spezifiziert zzz rdacontent nicht spezifiziert z rdamedia nicht spezifiziert zu rdacarrier This paper studies the smearing effect encountered in contribution plot based fault isolation, i.e., the influence of faulty variables on the contributions of non-faulty variables. Since the generation of contribution plots requires no a priori information about the detected disturbance (e.g., historical faulty data), it is a popular fault isolation technique in Statistical Process Control (SPC). However, Westerhuis et al. (2000) demonstrated that contributions suffer from fault smearing. As a consequence, variables unaffected by the fault may be highlighted and faulty variables obscured during the contribution analysis. This paper presents a thorough analysis of the smearing effect for three general contribution computation methods: complete decomposition, partial decomposition and reconstruction-based contributions. The analysis shows that (i) smearing is present in all three methods, (ii) smearing depends on the chosen number of principal components of the underlying PCA or PLS model and (iii) the extent of smearing increases for variables correlated in the training data for a well-chosen model order. The effect of smearing on the isolation performance of single and multiple sensor faults of various magnitudes is studied and illustrated using a simulation case study. The results indicate that correct isolation with contribution plots is not guaranteed for multiple sensor faults. Furthermore, contribution plots only outperform univariate fault isolation for single sensor faults with small magnitudes. For multiple sensor faults, univariate fault isolation exhibits a significantly larger correct fault isolation rate. Based on the smearing analysis and the specific results for sensor faults, the authors advise to use contributions only if a sound physical interpretation of the principal components is available. Otherwise multivariate detection followed by univariate fault isolation is recommended. This paper studies the smearing effect encountered in contribution plot based fault isolation, i.e., the influence of faulty variables on the contributions of non-faulty variables. Since the generation of contribution plots requires no a priori information about the detected disturbance (e.g., historical faulty data), it is a popular fault isolation technique in Statistical Process Control (SPC). However, Westerhuis et al. (2000) demonstrated that contributions suffer from fault smearing. As a consequence, variables unaffected by the fault may be highlighted and faulty variables obscured during the contribution analysis. This paper presents a thorough analysis of the smearing effect for three general contribution computation methods: complete decomposition, partial decomposition and reconstruction-based contributions. The analysis shows that (i) smearing is present in all three methods, (ii) smearing depends on the chosen number of principal components of the underlying PCA or PLS model and (iii) the extent of smearing increases for variables correlated in the training data for a well-chosen model order. The effect of smearing on the isolation performance of single and multiple sensor faults of various magnitudes is studied and illustrated using a simulation case study. The results indicate that correct isolation with contribution plots is not guaranteed for multiple sensor faults. Furthermore, contribution plots only outperform univariate fault isolation for single sensor faults with small magnitudes. For multiple sensor faults, univariate fault isolation exhibits a significantly larger correct fault isolation rate. Based on the smearing analysis and the specific results for sensor faults, the authors advise to use contributions only if a sound physical interpretation of the principal components is available. Otherwise multivariate detection followed by univariate fault isolation is recommended. Process control Elsevier Fault detection/isolation Elsevier Mathematical modelling Elsevier Contribution plots Elsevier Chemical processes Elsevier Remediation Elsevier Vanlaer, Jef oth Gins, Geert oth Van Impe, Jan F.M. oth Enthalten in Elsevier Science Dong, Ran ELSEVIER Plasticity in responses to dimensional variations of soil space in 19 grassland plant species 2022 Amsterdam [u.a.] (DE-627)ELV008347182 volume:104 year:2013 day:18 month:12 pages:285-293 extent:9 https://doi.org/10.1016/j.ces.2013.08.007 Volltext GBV_USEFLAG_U GBV_ELV SYSFLAG_U FID-BIODIV SSG-OLC-PHA AR 104 2013 18 1218 285-293 9 045F 660 |
| allfieldsGer |
10.1016/j.ces.2013.08.007 doi GBVA2013019000009.pica (DE-627)ELV027491331 (ELSEVIER)S0009-2509(13)00550-2 DE-627 ger DE-627 rakwb eng 660 660 DE-600 570 630 VZ BIODIV DE-30 fid Van den Kerkhof, Pieter verfasserin aut Analysis of smearing-out in contribution plot based fault isolation for Statistical Process Control 2013transfer abstract 9 nicht spezifiziert zzz rdacontent nicht spezifiziert z rdamedia nicht spezifiziert zu rdacarrier This paper studies the smearing effect encountered in contribution plot based fault isolation, i.e., the influence of faulty variables on the contributions of non-faulty variables. Since the generation of contribution plots requires no a priori information about the detected disturbance (e.g., historical faulty data), it is a popular fault isolation technique in Statistical Process Control (SPC). However, Westerhuis et al. (2000) demonstrated that contributions suffer from fault smearing. As a consequence, variables unaffected by the fault may be highlighted and faulty variables obscured during the contribution analysis. This paper presents a thorough analysis of the smearing effect for three general contribution computation methods: complete decomposition, partial decomposition and reconstruction-based contributions. The analysis shows that (i) smearing is present in all three methods, (ii) smearing depends on the chosen number of principal components of the underlying PCA or PLS model and (iii) the extent of smearing increases for variables correlated in the training data for a well-chosen model order. The effect of smearing on the isolation performance of single and multiple sensor faults of various magnitudes is studied and illustrated using a simulation case study. The results indicate that correct isolation with contribution plots is not guaranteed for multiple sensor faults. Furthermore, contribution plots only outperform univariate fault isolation for single sensor faults with small magnitudes. For multiple sensor faults, univariate fault isolation exhibits a significantly larger correct fault isolation rate. Based on the smearing analysis and the specific results for sensor faults, the authors advise to use contributions only if a sound physical interpretation of the principal components is available. Otherwise multivariate detection followed by univariate fault isolation is recommended. This paper studies the smearing effect encountered in contribution plot based fault isolation, i.e., the influence of faulty variables on the contributions of non-faulty variables. Since the generation of contribution plots requires no a priori information about the detected disturbance (e.g., historical faulty data), it is a popular fault isolation technique in Statistical Process Control (SPC). However, Westerhuis et al. (2000) demonstrated that contributions suffer from fault smearing. As a consequence, variables unaffected by the fault may be highlighted and faulty variables obscured during the contribution analysis. This paper presents a thorough analysis of the smearing effect for three general contribution computation methods: complete decomposition, partial decomposition and reconstruction-based contributions. The analysis shows that (i) smearing is present in all three methods, (ii) smearing depends on the chosen number of principal components of the underlying PCA or PLS model and (iii) the extent of smearing increases for variables correlated in the training data for a well-chosen model order. The effect of smearing on the isolation performance of single and multiple sensor faults of various magnitudes is studied and illustrated using a simulation case study. The results indicate that correct isolation with contribution plots is not guaranteed for multiple sensor faults. Furthermore, contribution plots only outperform univariate fault isolation for single sensor faults with small magnitudes. For multiple sensor faults, univariate fault isolation exhibits a significantly larger correct fault isolation rate. Based on the smearing analysis and the specific results for sensor faults, the authors advise to use contributions only if a sound physical interpretation of the principal components is available. Otherwise multivariate detection followed by univariate fault isolation is recommended. Process control Elsevier Fault detection/isolation Elsevier Mathematical modelling Elsevier Contribution plots Elsevier Chemical processes Elsevier Remediation Elsevier Vanlaer, Jef oth Gins, Geert oth Van Impe, Jan F.M. oth Enthalten in Elsevier Science Dong, Ran ELSEVIER Plasticity in responses to dimensional variations of soil space in 19 grassland plant species 2022 Amsterdam [u.a.] (DE-627)ELV008347182 volume:104 year:2013 day:18 month:12 pages:285-293 extent:9 https://doi.org/10.1016/j.ces.2013.08.007 Volltext GBV_USEFLAG_U GBV_ELV SYSFLAG_U FID-BIODIV SSG-OLC-PHA AR 104 2013 18 1218 285-293 9 045F 660 |
| allfieldsSound |
10.1016/j.ces.2013.08.007 doi GBVA2013019000009.pica (DE-627)ELV027491331 (ELSEVIER)S0009-2509(13)00550-2 DE-627 ger DE-627 rakwb eng 660 660 DE-600 570 630 VZ BIODIV DE-30 fid Van den Kerkhof, Pieter verfasserin aut Analysis of smearing-out in contribution plot based fault isolation for Statistical Process Control 2013transfer abstract 9 nicht spezifiziert zzz rdacontent nicht spezifiziert z rdamedia nicht spezifiziert zu rdacarrier This paper studies the smearing effect encountered in contribution plot based fault isolation, i.e., the influence of faulty variables on the contributions of non-faulty variables. Since the generation of contribution plots requires no a priori information about the detected disturbance (e.g., historical faulty data), it is a popular fault isolation technique in Statistical Process Control (SPC). However, Westerhuis et al. (2000) demonstrated that contributions suffer from fault smearing. As a consequence, variables unaffected by the fault may be highlighted and faulty variables obscured during the contribution analysis. This paper presents a thorough analysis of the smearing effect for three general contribution computation methods: complete decomposition, partial decomposition and reconstruction-based contributions. The analysis shows that (i) smearing is present in all three methods, (ii) smearing depends on the chosen number of principal components of the underlying PCA or PLS model and (iii) the extent of smearing increases for variables correlated in the training data for a well-chosen model order. The effect of smearing on the isolation performance of single and multiple sensor faults of various magnitudes is studied and illustrated using a simulation case study. The results indicate that correct isolation with contribution plots is not guaranteed for multiple sensor faults. Furthermore, contribution plots only outperform univariate fault isolation for single sensor faults with small magnitudes. For multiple sensor faults, univariate fault isolation exhibits a significantly larger correct fault isolation rate. Based on the smearing analysis and the specific results for sensor faults, the authors advise to use contributions only if a sound physical interpretation of the principal components is available. Otherwise multivariate detection followed by univariate fault isolation is recommended. This paper studies the smearing effect encountered in contribution plot based fault isolation, i.e., the influence of faulty variables on the contributions of non-faulty variables. Since the generation of contribution plots requires no a priori information about the detected disturbance (e.g., historical faulty data), it is a popular fault isolation technique in Statistical Process Control (SPC). However, Westerhuis et al. (2000) demonstrated that contributions suffer from fault smearing. As a consequence, variables unaffected by the fault may be highlighted and faulty variables obscured during the contribution analysis. This paper presents a thorough analysis of the smearing effect for three general contribution computation methods: complete decomposition, partial decomposition and reconstruction-based contributions. The analysis shows that (i) smearing is present in all three methods, (ii) smearing depends on the chosen number of principal components of the underlying PCA or PLS model and (iii) the extent of smearing increases for variables correlated in the training data for a well-chosen model order. The effect of smearing on the isolation performance of single and multiple sensor faults of various magnitudes is studied and illustrated using a simulation case study. The results indicate that correct isolation with contribution plots is not guaranteed for multiple sensor faults. Furthermore, contribution plots only outperform univariate fault isolation for single sensor faults with small magnitudes. For multiple sensor faults, univariate fault isolation exhibits a significantly larger correct fault isolation rate. Based on the smearing analysis and the specific results for sensor faults, the authors advise to use contributions only if a sound physical interpretation of the principal components is available. Otherwise multivariate detection followed by univariate fault isolation is recommended. Process control Elsevier Fault detection/isolation Elsevier Mathematical modelling Elsevier Contribution plots Elsevier Chemical processes Elsevier Remediation Elsevier Vanlaer, Jef oth Gins, Geert oth Van Impe, Jan F.M. oth Enthalten in Elsevier Science Dong, Ran ELSEVIER Plasticity in responses to dimensional variations of soil space in 19 grassland plant species 2022 Amsterdam [u.a.] (DE-627)ELV008347182 volume:104 year:2013 day:18 month:12 pages:285-293 extent:9 https://doi.org/10.1016/j.ces.2013.08.007 Volltext GBV_USEFLAG_U GBV_ELV SYSFLAG_U FID-BIODIV SSG-OLC-PHA AR 104 2013 18 1218 285-293 9 045F 660 |
| language |
English |
| source |
Enthalten in Plasticity in responses to dimensional variations of soil space in 19 grassland plant species Amsterdam [u.a.] volume:104 year:2013 day:18 month:12 pages:285-293 extent:9 |
| sourceStr |
Enthalten in Plasticity in responses to dimensional variations of soil space in 19 grassland plant species Amsterdam [u.a.] volume:104 year:2013 day:18 month:12 pages:285-293 extent:9 |
| format_phy_str_mv |
Article |
| institution |
findex.gbv.de |
| topic_facet |
Process control Fault detection/isolation Mathematical modelling Contribution plots Chemical processes Remediation |
| dewey-raw |
660 |
| isfreeaccess_bool |
false |
| container_title |
Plasticity in responses to dimensional variations of soil space in 19 grassland plant species |
| authorswithroles_txt_mv |
Van den Kerkhof, Pieter @@aut@@ Vanlaer, Jef @@oth@@ Gins, Geert @@oth@@ Van Impe, Jan F.M. @@oth@@ |
| publishDateDaySort_date |
2013-01-18T00:00:00Z |
| hierarchy_top_id |
ELV008347182 |
| dewey-sort |
3660 |
| id |
ELV027491331 |
| language_de |
englisch |
| fullrecord |
<?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim"><record><leader>01000caa a22002652 4500</leader><controlfield tag="001">ELV027491331</controlfield><controlfield tag="003">DE-627</controlfield><controlfield tag="005">20230625151959.0</controlfield><controlfield tag="007">cr uuu---uuuuu</controlfield><controlfield tag="008">180603s2013 xx |||||o 00| ||eng c</controlfield><datafield tag="024" ind1="7" ind2=" "><subfield code="a">10.1016/j.ces.2013.08.007</subfield><subfield code="2">doi</subfield></datafield><datafield tag="028" ind1="5" ind2="2"><subfield code="a">GBVA2013019000009.pica</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-627)ELV027491331</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(ELSEVIER)S0009-2509(13)00550-2</subfield></datafield><datafield tag="040" ind1=" " ind2=" "><subfield code="a">DE-627</subfield><subfield code="b">ger</subfield><subfield code="c">DE-627</subfield><subfield code="e">rakwb</subfield></datafield><datafield tag="041" ind1=" " ind2=" "><subfield code="a">eng</subfield></datafield><datafield tag="082" ind1="0" ind2=" "><subfield code="a">660</subfield></datafield><datafield tag="082" ind1="0" ind2="4"><subfield code="a">660</subfield><subfield code="q">DE-600</subfield></datafield><datafield tag="082" ind1="0" ind2="4"><subfield code="a">570</subfield><subfield code="a">630</subfield><subfield code="q">VZ</subfield></datafield><datafield tag="084" ind1=" " ind2=" "><subfield code="a">BIODIV</subfield><subfield code="q">DE-30</subfield><subfield code="2">fid</subfield></datafield><datafield tag="100" ind1="1" ind2=" "><subfield code="a">Van den Kerkhof, Pieter</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="245" ind1="1" ind2="0"><subfield code="a">Analysis of smearing-out in contribution plot based fault isolation for Statistical Process Control</subfield></datafield><datafield tag="264" ind1=" " ind2="1"><subfield code="c">2013transfer abstract</subfield></datafield><datafield tag="300" ind1=" " ind2=" "><subfield code="a">9</subfield></datafield><datafield tag="336" ind1=" " ind2=" "><subfield code="a">nicht spezifiziert</subfield><subfield code="b">zzz</subfield><subfield code="2">rdacontent</subfield></datafield><datafield tag="337" ind1=" " ind2=" "><subfield code="a">nicht spezifiziert</subfield><subfield code="b">z</subfield><subfield code="2">rdamedia</subfield></datafield><datafield tag="338" ind1=" " ind2=" "><subfield code="a">nicht spezifiziert</subfield><subfield code="b">zu</subfield><subfield code="2">rdacarrier</subfield></datafield><datafield tag="520" ind1=" " ind2=" "><subfield code="a">This paper studies the smearing effect encountered in contribution plot based fault isolation, i.e., the influence of faulty variables on the contributions of non-faulty variables. Since the generation of contribution plots requires no a priori information about the detected disturbance (e.g., historical faulty data), it is a popular fault isolation technique in Statistical Process Control (SPC). However, Westerhuis et al. (2000) demonstrated that contributions suffer from fault smearing. As a consequence, variables unaffected by the fault may be highlighted and faulty variables obscured during the contribution analysis. This paper presents a thorough analysis of the smearing effect for three general contribution computation methods: complete decomposition, partial decomposition and reconstruction-based contributions. The analysis shows that (i) smearing is present in all three methods, (ii) smearing depends on the chosen number of principal components of the underlying PCA or PLS model and (iii) the extent of smearing increases for variables correlated in the training data for a well-chosen model order. The effect of smearing on the isolation performance of single and multiple sensor faults of various magnitudes is studied and illustrated using a simulation case study. The results indicate that correct isolation with contribution plots is not guaranteed for multiple sensor faults. Furthermore, contribution plots only outperform univariate fault isolation for single sensor faults with small magnitudes. For multiple sensor faults, univariate fault isolation exhibits a significantly larger correct fault isolation rate. Based on the smearing analysis and the specific results for sensor faults, the authors advise to use contributions only if a sound physical interpretation of the principal components is available. Otherwise multivariate detection followed by univariate fault isolation is recommended.</subfield></datafield><datafield tag="520" ind1=" " ind2=" "><subfield code="a">This paper studies the smearing effect encountered in contribution plot based fault isolation, i.e., the influence of faulty variables on the contributions of non-faulty variables. Since the generation of contribution plots requires no a priori information about the detected disturbance (e.g., historical faulty data), it is a popular fault isolation technique in Statistical Process Control (SPC). However, Westerhuis et al. (2000) demonstrated that contributions suffer from fault smearing. As a consequence, variables unaffected by the fault may be highlighted and faulty variables obscured during the contribution analysis. This paper presents a thorough analysis of the smearing effect for three general contribution computation methods: complete decomposition, partial decomposition and reconstruction-based contributions. The analysis shows that (i) smearing is present in all three methods, (ii) smearing depends on the chosen number of principal components of the underlying PCA or PLS model and (iii) the extent of smearing increases for variables correlated in the training data for a well-chosen model order. The effect of smearing on the isolation performance of single and multiple sensor faults of various magnitudes is studied and illustrated using a simulation case study. The results indicate that correct isolation with contribution plots is not guaranteed for multiple sensor faults. Furthermore, contribution plots only outperform univariate fault isolation for single sensor faults with small magnitudes. For multiple sensor faults, univariate fault isolation exhibits a significantly larger correct fault isolation rate. Based on the smearing analysis and the specific results for sensor faults, the authors advise to use contributions only if a sound physical interpretation of the principal components is available. Otherwise multivariate detection followed by univariate fault isolation is recommended.</subfield></datafield><datafield tag="650" ind1=" " ind2="7"><subfield code="a">Process control</subfield><subfield code="2">Elsevier</subfield></datafield><datafield tag="650" ind1=" " ind2="7"><subfield code="a">Fault detection/isolation</subfield><subfield code="2">Elsevier</subfield></datafield><datafield tag="650" ind1=" " ind2="7"><subfield code="a">Mathematical modelling</subfield><subfield code="2">Elsevier</subfield></datafield><datafield tag="650" ind1=" " ind2="7"><subfield code="a">Contribution plots</subfield><subfield code="2">Elsevier</subfield></datafield><datafield tag="650" ind1=" " ind2="7"><subfield code="a">Chemical processes</subfield><subfield code="2">Elsevier</subfield></datafield><datafield tag="650" ind1=" " ind2="7"><subfield code="a">Remediation</subfield><subfield code="2">Elsevier</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Vanlaer, Jef</subfield><subfield code="4">oth</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Gins, Geert</subfield><subfield code="4">oth</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Van Impe, Jan F.M.</subfield><subfield code="4">oth</subfield></datafield><datafield tag="773" ind1="0" ind2="8"><subfield code="i">Enthalten in</subfield><subfield code="n">Elsevier Science</subfield><subfield code="a">Dong, Ran ELSEVIER</subfield><subfield code="t">Plasticity in responses to dimensional variations of soil space in 19 grassland plant species</subfield><subfield code="d">2022</subfield><subfield code="g">Amsterdam [u.a.]</subfield><subfield code="w">(DE-627)ELV008347182</subfield></datafield><datafield tag="773" ind1="1" ind2="8"><subfield code="g">volume:104</subfield><subfield code="g">year:2013</subfield><subfield code="g">day:18</subfield><subfield code="g">month:12</subfield><subfield code="g">pages:285-293</subfield><subfield code="g">extent:9</subfield></datafield><datafield tag="856" ind1="4" ind2="0"><subfield code="u">https://doi.org/10.1016/j.ces.2013.08.007</subfield><subfield code="3">Volltext</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_USEFLAG_U</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ELV</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SYSFLAG_U</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">FID-BIODIV</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SSG-OLC-PHA</subfield></datafield><datafield tag="951" ind1=" " ind2=" "><subfield code="a">AR</subfield></datafield><datafield tag="952" ind1=" " ind2=" "><subfield code="d">104</subfield><subfield code="j">2013</subfield><subfield code="b">18</subfield><subfield code="c">1218</subfield><subfield code="h">285-293</subfield><subfield code="g">9</subfield></datafield><datafield tag="953" ind1=" " ind2=" "><subfield code="2">045F</subfield><subfield code="a">660</subfield></datafield></record></collection>
|
| author |
Van den Kerkhof, Pieter |
| spellingShingle |
Van den Kerkhof, Pieter ddc 660 ddc 570 fid BIODIV Elsevier Process control Elsevier Fault detection/isolation Elsevier Mathematical modelling Elsevier Contribution plots Elsevier Chemical processes Elsevier Remediation Analysis of smearing-out in contribution plot based fault isolation for Statistical Process Control |
| authorStr |
Van den Kerkhof, Pieter |
| ppnlink_with_tag_str_mv |
@@773@@(DE-627)ELV008347182 |
| format |
electronic Article |
| dewey-ones |
660 - Chemical engineering 570 - Life sciences; biology 630 - Agriculture & related technologies |
| delete_txt_mv |
keep |
| author_role |
aut |
| collection |
elsevier |
| remote_str |
true |
| illustrated |
Not Illustrated |
| topic_title |
660 660 DE-600 570 630 VZ BIODIV DE-30 fid Analysis of smearing-out in contribution plot based fault isolation for Statistical Process Control Process control Elsevier Fault detection/isolation Elsevier Mathematical modelling Elsevier Contribution plots Elsevier Chemical processes Elsevier Remediation Elsevier |
| topic |
ddc 660 ddc 570 fid BIODIV Elsevier Process control Elsevier Fault detection/isolation Elsevier Mathematical modelling Elsevier Contribution plots Elsevier Chemical processes Elsevier Remediation |
| topic_unstemmed |
ddc 660 ddc 570 fid BIODIV Elsevier Process control Elsevier Fault detection/isolation Elsevier Mathematical modelling Elsevier Contribution plots Elsevier Chemical processes Elsevier Remediation |
| topic_browse |
ddc 660 ddc 570 fid BIODIV Elsevier Process control Elsevier Fault detection/isolation Elsevier Mathematical modelling Elsevier Contribution plots Elsevier Chemical processes Elsevier Remediation |
| format_facet |
Elektronische Aufsätze Aufsätze Elektronische Ressource |
| format_main_str_mv |
Text Zeitschrift/Artikel |
| carriertype_str_mv |
zu |
| author2_variant |
j v jv g g gg i j f v ijf ijfv |
| hierarchy_parent_title |
Plasticity in responses to dimensional variations of soil space in 19 grassland plant species |
| hierarchy_parent_id |
ELV008347182 |
| dewey-tens |
660 - Chemical engineering 570 - Life sciences; biology 630 - Agriculture |
| hierarchy_top_title |
Plasticity in responses to dimensional variations of soil space in 19 grassland plant species |
| isfreeaccess_txt |
false |
| familylinks_str_mv |
(DE-627)ELV008347182 |
| title |
Analysis of smearing-out in contribution plot based fault isolation for Statistical Process Control |
| ctrlnum |
(DE-627)ELV027491331 (ELSEVIER)S0009-2509(13)00550-2 |
| title_full |
Analysis of smearing-out in contribution plot based fault isolation for Statistical Process Control |
| author_sort |
Van den Kerkhof, Pieter |
| journal |
Plasticity in responses to dimensional variations of soil space in 19 grassland plant species |
| journalStr |
Plasticity in responses to dimensional variations of soil space in 19 grassland plant species |
| lang_code |
eng |
| isOA_bool |
false |
| dewey-hundreds |
600 - Technology 500 - Science |
| recordtype |
marc |
| publishDateSort |
2013 |
| contenttype_str_mv |
zzz |
| container_start_page |
285 |
| author_browse |
Van den Kerkhof, Pieter |
| container_volume |
104 |
| physical |
9 |
| class |
660 660 DE-600 570 630 VZ BIODIV DE-30 fid |
| format_se |
Elektronische Aufsätze |
| author-letter |
Van den Kerkhof, Pieter |
| doi_str_mv |
10.1016/j.ces.2013.08.007 |
| dewey-full |
660 570 630 |
| title_sort |
analysis of smearing-out in contribution plot based fault isolation for statistical process control |
| title_auth |
Analysis of smearing-out in contribution plot based fault isolation for Statistical Process Control |
| abstract |
This paper studies the smearing effect encountered in contribution plot based fault isolation, i.e., the influence of faulty variables on the contributions of non-faulty variables. Since the generation of contribution plots requires no a priori information about the detected disturbance (e.g., historical faulty data), it is a popular fault isolation technique in Statistical Process Control (SPC). However, Westerhuis et al. (2000) demonstrated that contributions suffer from fault smearing. As a consequence, variables unaffected by the fault may be highlighted and faulty variables obscured during the contribution analysis. This paper presents a thorough analysis of the smearing effect for three general contribution computation methods: complete decomposition, partial decomposition and reconstruction-based contributions. The analysis shows that (i) smearing is present in all three methods, (ii) smearing depends on the chosen number of principal components of the underlying PCA or PLS model and (iii) the extent of smearing increases for variables correlated in the training data for a well-chosen model order. The effect of smearing on the isolation performance of single and multiple sensor faults of various magnitudes is studied and illustrated using a simulation case study. The results indicate that correct isolation with contribution plots is not guaranteed for multiple sensor faults. Furthermore, contribution plots only outperform univariate fault isolation for single sensor faults with small magnitudes. For multiple sensor faults, univariate fault isolation exhibits a significantly larger correct fault isolation rate. Based on the smearing analysis and the specific results for sensor faults, the authors advise to use contributions only if a sound physical interpretation of the principal components is available. Otherwise multivariate detection followed by univariate fault isolation is recommended. |
| abstractGer |
This paper studies the smearing effect encountered in contribution plot based fault isolation, i.e., the influence of faulty variables on the contributions of non-faulty variables. Since the generation of contribution plots requires no a priori information about the detected disturbance (e.g., historical faulty data), it is a popular fault isolation technique in Statistical Process Control (SPC). However, Westerhuis et al. (2000) demonstrated that contributions suffer from fault smearing. As a consequence, variables unaffected by the fault may be highlighted and faulty variables obscured during the contribution analysis. This paper presents a thorough analysis of the smearing effect for three general contribution computation methods: complete decomposition, partial decomposition and reconstruction-based contributions. The analysis shows that (i) smearing is present in all three methods, (ii) smearing depends on the chosen number of principal components of the underlying PCA or PLS model and (iii) the extent of smearing increases for variables correlated in the training data for a well-chosen model order. The effect of smearing on the isolation performance of single and multiple sensor faults of various magnitudes is studied and illustrated using a simulation case study. The results indicate that correct isolation with contribution plots is not guaranteed for multiple sensor faults. Furthermore, contribution plots only outperform univariate fault isolation for single sensor faults with small magnitudes. For multiple sensor faults, univariate fault isolation exhibits a significantly larger correct fault isolation rate. Based on the smearing analysis and the specific results for sensor faults, the authors advise to use contributions only if a sound physical interpretation of the principal components is available. Otherwise multivariate detection followed by univariate fault isolation is recommended. |
| abstract_unstemmed |
This paper studies the smearing effect encountered in contribution plot based fault isolation, i.e., the influence of faulty variables on the contributions of non-faulty variables. Since the generation of contribution plots requires no a priori information about the detected disturbance (e.g., historical faulty data), it is a popular fault isolation technique in Statistical Process Control (SPC). However, Westerhuis et al. (2000) demonstrated that contributions suffer from fault smearing. As a consequence, variables unaffected by the fault may be highlighted and faulty variables obscured during the contribution analysis. This paper presents a thorough analysis of the smearing effect for three general contribution computation methods: complete decomposition, partial decomposition and reconstruction-based contributions. The analysis shows that (i) smearing is present in all three methods, (ii) smearing depends on the chosen number of principal components of the underlying PCA or PLS model and (iii) the extent of smearing increases for variables correlated in the training data for a well-chosen model order. The effect of smearing on the isolation performance of single and multiple sensor faults of various magnitudes is studied and illustrated using a simulation case study. The results indicate that correct isolation with contribution plots is not guaranteed for multiple sensor faults. Furthermore, contribution plots only outperform univariate fault isolation for single sensor faults with small magnitudes. For multiple sensor faults, univariate fault isolation exhibits a significantly larger correct fault isolation rate. Based on the smearing analysis and the specific results for sensor faults, the authors advise to use contributions only if a sound physical interpretation of the principal components is available. Otherwise multivariate detection followed by univariate fault isolation is recommended. |
| collection_details |
GBV_USEFLAG_U GBV_ELV SYSFLAG_U FID-BIODIV SSG-OLC-PHA |
| title_short |
Analysis of smearing-out in contribution plot based fault isolation for Statistical Process Control |
| url |
https://doi.org/10.1016/j.ces.2013.08.007 |
| remote_bool |
true |
| author2 |
Vanlaer, Jef Gins, Geert Van Impe, Jan F.M. |
| author2Str |
Vanlaer, Jef Gins, Geert Van Impe, Jan F.M. |
| ppnlink |
ELV008347182 |
| mediatype_str_mv |
z |
| isOA_txt |
false |
| hochschulschrift_bool |
false |
| author2_role |
oth oth oth |
| doi_str |
10.1016/j.ces.2013.08.007 |
| up_date |
2024-07-06T21:56:34.816Z |
| _version_ |
1803868433093230592 |
| fullrecord_marcxml |
<?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim"><record><leader>01000caa a22002652 4500</leader><controlfield tag="001">ELV027491331</controlfield><controlfield tag="003">DE-627</controlfield><controlfield tag="005">20230625151959.0</controlfield><controlfield tag="007">cr uuu---uuuuu</controlfield><controlfield tag="008">180603s2013 xx |||||o 00| ||eng c</controlfield><datafield tag="024" ind1="7" ind2=" "><subfield code="a">10.1016/j.ces.2013.08.007</subfield><subfield code="2">doi</subfield></datafield><datafield tag="028" ind1="5" ind2="2"><subfield code="a">GBVA2013019000009.pica</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-627)ELV027491331</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(ELSEVIER)S0009-2509(13)00550-2</subfield></datafield><datafield tag="040" ind1=" " ind2=" "><subfield code="a">DE-627</subfield><subfield code="b">ger</subfield><subfield code="c">DE-627</subfield><subfield code="e">rakwb</subfield></datafield><datafield tag="041" ind1=" " ind2=" "><subfield code="a">eng</subfield></datafield><datafield tag="082" ind1="0" ind2=" "><subfield code="a">660</subfield></datafield><datafield tag="082" ind1="0" ind2="4"><subfield code="a">660</subfield><subfield code="q">DE-600</subfield></datafield><datafield tag="082" ind1="0" ind2="4"><subfield code="a">570</subfield><subfield code="a">630</subfield><subfield code="q">VZ</subfield></datafield><datafield tag="084" ind1=" " ind2=" "><subfield code="a">BIODIV</subfield><subfield code="q">DE-30</subfield><subfield code="2">fid</subfield></datafield><datafield tag="100" ind1="1" ind2=" "><subfield code="a">Van den Kerkhof, Pieter</subfield><subfield code="e">verfasserin</subfield><subfield code="4">aut</subfield></datafield><datafield tag="245" ind1="1" ind2="0"><subfield code="a">Analysis of smearing-out in contribution plot based fault isolation for Statistical Process Control</subfield></datafield><datafield tag="264" ind1=" " ind2="1"><subfield code="c">2013transfer abstract</subfield></datafield><datafield tag="300" ind1=" " ind2=" "><subfield code="a">9</subfield></datafield><datafield tag="336" ind1=" " ind2=" "><subfield code="a">nicht spezifiziert</subfield><subfield code="b">zzz</subfield><subfield code="2">rdacontent</subfield></datafield><datafield tag="337" ind1=" " ind2=" "><subfield code="a">nicht spezifiziert</subfield><subfield code="b">z</subfield><subfield code="2">rdamedia</subfield></datafield><datafield tag="338" ind1=" " ind2=" "><subfield code="a">nicht spezifiziert</subfield><subfield code="b">zu</subfield><subfield code="2">rdacarrier</subfield></datafield><datafield tag="520" ind1=" " ind2=" "><subfield code="a">This paper studies the smearing effect encountered in contribution plot based fault isolation, i.e., the influence of faulty variables on the contributions of non-faulty variables. Since the generation of contribution plots requires no a priori information about the detected disturbance (e.g., historical faulty data), it is a popular fault isolation technique in Statistical Process Control (SPC). However, Westerhuis et al. (2000) demonstrated that contributions suffer from fault smearing. As a consequence, variables unaffected by the fault may be highlighted and faulty variables obscured during the contribution analysis. This paper presents a thorough analysis of the smearing effect for three general contribution computation methods: complete decomposition, partial decomposition and reconstruction-based contributions. The analysis shows that (i) smearing is present in all three methods, (ii) smearing depends on the chosen number of principal components of the underlying PCA or PLS model and (iii) the extent of smearing increases for variables correlated in the training data for a well-chosen model order. The effect of smearing on the isolation performance of single and multiple sensor faults of various magnitudes is studied and illustrated using a simulation case study. The results indicate that correct isolation with contribution plots is not guaranteed for multiple sensor faults. Furthermore, contribution plots only outperform univariate fault isolation for single sensor faults with small magnitudes. For multiple sensor faults, univariate fault isolation exhibits a significantly larger correct fault isolation rate. Based on the smearing analysis and the specific results for sensor faults, the authors advise to use contributions only if a sound physical interpretation of the principal components is available. Otherwise multivariate detection followed by univariate fault isolation is recommended.</subfield></datafield><datafield tag="520" ind1=" " ind2=" "><subfield code="a">This paper studies the smearing effect encountered in contribution plot based fault isolation, i.e., the influence of faulty variables on the contributions of non-faulty variables. Since the generation of contribution plots requires no a priori information about the detected disturbance (e.g., historical faulty data), it is a popular fault isolation technique in Statistical Process Control (SPC). However, Westerhuis et al. (2000) demonstrated that contributions suffer from fault smearing. As a consequence, variables unaffected by the fault may be highlighted and faulty variables obscured during the contribution analysis. This paper presents a thorough analysis of the smearing effect for three general contribution computation methods: complete decomposition, partial decomposition and reconstruction-based contributions. The analysis shows that (i) smearing is present in all three methods, (ii) smearing depends on the chosen number of principal components of the underlying PCA or PLS model and (iii) the extent of smearing increases for variables correlated in the training data for a well-chosen model order. The effect of smearing on the isolation performance of single and multiple sensor faults of various magnitudes is studied and illustrated using a simulation case study. The results indicate that correct isolation with contribution plots is not guaranteed for multiple sensor faults. Furthermore, contribution plots only outperform univariate fault isolation for single sensor faults with small magnitudes. For multiple sensor faults, univariate fault isolation exhibits a significantly larger correct fault isolation rate. Based on the smearing analysis and the specific results for sensor faults, the authors advise to use contributions only if a sound physical interpretation of the principal components is available. Otherwise multivariate detection followed by univariate fault isolation is recommended.</subfield></datafield><datafield tag="650" ind1=" " ind2="7"><subfield code="a">Process control</subfield><subfield code="2">Elsevier</subfield></datafield><datafield tag="650" ind1=" " ind2="7"><subfield code="a">Fault detection/isolation</subfield><subfield code="2">Elsevier</subfield></datafield><datafield tag="650" ind1=" " ind2="7"><subfield code="a">Mathematical modelling</subfield><subfield code="2">Elsevier</subfield></datafield><datafield tag="650" ind1=" " ind2="7"><subfield code="a">Contribution plots</subfield><subfield code="2">Elsevier</subfield></datafield><datafield tag="650" ind1=" " ind2="7"><subfield code="a">Chemical processes</subfield><subfield code="2">Elsevier</subfield></datafield><datafield tag="650" ind1=" " ind2="7"><subfield code="a">Remediation</subfield><subfield code="2">Elsevier</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Vanlaer, Jef</subfield><subfield code="4">oth</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Gins, Geert</subfield><subfield code="4">oth</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Van Impe, Jan F.M.</subfield><subfield code="4">oth</subfield></datafield><datafield tag="773" ind1="0" ind2="8"><subfield code="i">Enthalten in</subfield><subfield code="n">Elsevier Science</subfield><subfield code="a">Dong, Ran ELSEVIER</subfield><subfield code="t">Plasticity in responses to dimensional variations of soil space in 19 grassland plant species</subfield><subfield code="d">2022</subfield><subfield code="g">Amsterdam [u.a.]</subfield><subfield code="w">(DE-627)ELV008347182</subfield></datafield><datafield tag="773" ind1="1" ind2="8"><subfield code="g">volume:104</subfield><subfield code="g">year:2013</subfield><subfield code="g">day:18</subfield><subfield code="g">month:12</subfield><subfield code="g">pages:285-293</subfield><subfield code="g">extent:9</subfield></datafield><datafield tag="856" ind1="4" ind2="0"><subfield code="u">https://doi.org/10.1016/j.ces.2013.08.007</subfield><subfield code="3">Volltext</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_USEFLAG_U</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_ELV</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SYSFLAG_U</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">FID-BIODIV</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SSG-OLC-PHA</subfield></datafield><datafield tag="951" ind1=" " ind2=" "><subfield code="a">AR</subfield></datafield><datafield tag="952" ind1=" " ind2=" "><subfield code="d">104</subfield><subfield code="j">2013</subfield><subfield code="b">18</subfield><subfield code="c">1218</subfield><subfield code="h">285-293</subfield><subfield code="g">9</subfield></datafield><datafield tag="953" ind1=" " ind2=" "><subfield code="2">045F</subfield><subfield code="a">660</subfield></datafield></record></collection>
|
| score |
7.401602 |