Recursive Correlative Statistical Analysis Method With Sliding Windows for Incipient Fault Detection

Qin, Yihao; Yan, Yayun; Ji, Hongquan; Wang, Youqing

doi:10.1109/tie.2021.3070521

Cited by 63 publications

(27 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Foi utilizada então uma janela deslizante de dois dias para o cálculo da média de cada resíduo, que avança sobre os dados com 90% de sobreposição das amostras. Procedimento similar foi usado em Qin et al (2021) para detectar falhas incipientes. O resultado dessa análise foi utilizado de duas formas: plotando o resultado ao longo do tempo, e treinando um algoritmo para detectar a eminência de falhas.…”

Section: Análise Via Variáveis Externasunclassified

Detecção de anomalias em ventiladores industriais via sinais de vibração

Zuqui¹,

Rauber²,

Assis

et al. 2021

Procedings Do XV Simpósio Brasileiro De Automação Inteligente

View full text Add to dashboard Cite

Pelletizing furnaces require the use of high power fans to circulate the air in the different regions of the furnace, ensuring the efficiency of the pellet burning process and the use of energy. The failure of one of these fans results in a stop of the production process and economic losses. This work presents a contribution to support predictive maintenance routinely performed on this equipment. Vibration signals in rotating machinery are sensitive to many faults and are therefore widely used in diagnostic systems. From the speed and torque information of the fans and the vibration signals in the bearings collected at low sampling rates, it is possible to generate alarms that guide investigative actions, specifically in the equipment that exhibit anomalous behavior. Graphs with sliding windows allow checking trends in vibration increase, and are used to predict alarms in advance. The monitoring of two fans for four months, compared to inspection reports carried out in the same period showed the contribution that the proposed system can bring to reduce downtime and more rationally use the work of the maintenance teams. Resumo: Fornos de pelotização requerem o uso de ventiladores de grande potência para fazer circular o ar nas diferentes regiões do forno, garantindo a eficiência do processo de queima das pelotas e do uso de energia. A falha em um desses ventiladores implica na parada do processo de produção e perdas econômicas. Neste trabalho é apresentada uma contribuição para apoiar manutenção preditiva realizada rotineiramente nesses equipamentos. Os sinais de vibração em máquinas rotativas são sensíveis a muitas falhas, sendo por isso muito utilizados em sistemas de diagnóstico. A partir das informações de velocidade e torque dos ventiladores e dos sinais de vibração nos mancais coletados a baixas taxas de amostragem, consegue-se gerar alarmes que orientam ações de investigação especificamente nos equipamentos que apresentam comportamentos anômalos. Gráficos com janelas deslizantes permitem verificar tendências de aumento de vibração, e são usados para prever os alarmes com antecedência. O monitoramento de dois ventiladores durante quatro meses, comparado aos laudos de inspeção realizados no período, mostrou a contribuição que o sistema proposto pode trazer para reduzir paradas e usar mais racionalmente o trabalho das equipes de manutenção.

show abstract

Section: Análise Via Variáveis Externasunclassified

Detecção de anomalias em ventiladores industriais via sinais de vibração

Zuqui¹,

Rauber²,

Assis

et al. 2021

Procedings Do XV Simpósio Brasileiro De Automação Inteligente

View full text Add to dashboard Cite

show abstract

“…The ultra-supercritical thermal power plants have made great contributions to the development of society and still play a pivotal role in the current power system [27]. Efficient process monitoring is the foundation of continuous and stable operation for power plants.…”

Section: Fan System Of the Power Plantmentioning

confidence: 99%

“…The Mahalanobis distance (MD) can also be directly used for process monitoring [18]. As understanding of the fault initiation becomes more and more thorough, the moving window methods also be proposed for incipient fault detection [17,29,27]. Considering the practical applicability in industrial processes, a large number of improved methods have been developed for multimode and nonstationary monitoring [37,39,16,34].…”

Section: Introductionmentioning

confidence: 99%

Hybrid variable monitoring: An unsupervised process monitoring framework with binary and continuous variables

Wang¹,

Zhou²,

Chen³

2021

Preprint

View full text Add to dashboard Cite

Traditional process monitoring methods, such as PCA, PLS, ICA, MD et al., are strongly dependent on continuous variables because most of them inevitably involve Euclidean or Mahalanobis distance. With industrial processes becoming more and more complex and integrated, binary variables also appear in monitoring variables besides continuous variables, which makes process monitoring more challenging. The aforementioned traditional approaches are incompetent to mine the information of binary variables, so that the useful information contained in them is usually discarded during the data preprocessing. To solve the problem, this paper focuses on the issue of hybrid variable monitoring (HVM) and proposes a novel unsupervised framework of process monitoring with hybrid variables. HVM is addressed in the probabilistic framework, which can effectively exploit the process information implicit in both continuous and binary variables at the same time. In HVM, the statistics and the monitoring strategy suitable for hybrid variables with only healthy state data are defined and the physical explanation behind the framework is elaborated. In addition, the estimation of parameters required in HVM is derived in detail and the detectable condition of the proposed method is analyzed. Finally, the superiority of HVM is fully demonstrated first on a numerical simulation and then on an actual case of a thermal power plant.

show abstract

“…Multivariate statistical-based process monitoring (MSPM) methods 1 – 4 , e.g., principal component analysis (PCA) 5 , 6 , partial least squares (PLS) 7 , 8 , and canonical correlation analysis (CCA) 9 , 10 , are effective data-driven approaches for monitoring large-scale industrial processes. The main idea of MSPM is analyzing the correlation between process variables and extracting the feature components for the construction of statistical indices.…”

Section: Introductionmentioning

confidence: 99%

Minimalist module analysis for fault detection and localization

Lou

Wang

et al. 2021

Sci Rep

Self Cite

View full text Add to dashboard Cite

Traditional multivariate statistical-based process monitoring (MSPM) methods are effective data-driven approaches for monitoring large-scale industrial processes, but have a shortcoming in handling the redundant correlations between process variables. To address this shortcoming, this study proposes a new MSPM method called minimalist module analysis (MMA). MMA divides process data into several different minimalist modules and one more independent module. All variables in the minimalist module are strongly correlated, and no redundant variables exist; therefore, the extracted feature components in one minimalist module will not be disturbed by noise from the other modules. This study also proposes new monitoring indices and a fault localization strategy for MMA, and simulation tests demonstrate that MMA achieves superior performance in fault detection and localization.

show abstract

Recursive Correlative Statistical Analysis Method With Sliding Windows for Incipient Fault Detection

Cited by 63 publications

References 26 publications

Detecção de anomalias em ventiladores industriais via sinais de vibração

Detecção de anomalias em ventiladores industriais via sinais de vibração

Hybrid variable monitoring: An unsupervised process monitoring framework with binary and continuous variables

Minimalist module analysis for fault detection and localization

Contact Info

Product

Resources

About