Alarm management practices in natural gas processing plants

Soares, Vinícius Barroso; Pinto, José Carlos; Souza, Maurício B. de

doi:10.1016/j.conengprac.2016.07.004

Cited by 29 publications

(6 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, these approaches [5][6][7] [13] are limited to a particular domain and are difficult to apply in other industries. In addition to this, several machine learning techniques are also employed for the generation of alarms based on the historical trends [20][21] [23]. However, the reliability of such approaches is questionable, therefore, their actual application in industries is difficult.…”

Section: Figure 2 Architecture Of Alarm System In Industrial Controlmentioning

confidence: 99%

“…Furthermore, the proposed approach is not fault tolerant. In another study [21], the authors primarily studied the correlation variables among the alarms that effect the occurrence of alarm or cause abnormalities. They took the alarms logs of past three years from three identical Natural Gas Processing Plants and applied correlation methods i.e.…”

Section: ) Prediction Of Alarms Using Machine Learning Techniquesmentioning

confidence: 99%

“…Aforementioned approaches [20][21][22][23][24][25][26] certainly automate the different aspects of alarms systems without human intervention. However, such approaches are only meant for academia.…”

Section: ) Prediction Of Alarms Using Machine Learning Techniquesmentioning

confidence: 99%

See 2 more Smart Citations

A Model-Driven Alarms Framework (MAF) With Mobile Clients Support for Wide-Ranging Industrial Control Systems

et al. 2020

View full text Add to dashboard Cite

An efficient and robust alarms management is an extremely desirable feature to prevent critical failures in modern Industrial Control Systems. Generally, alarm systems are categorized into two major components; i.e., Alarm Server and Alarm Client. Alarm server is responsible to process alarms on real time values, while alarm clients display the generated alarms. With the advent of modern technologies, like, Internet of Thing (IoT), the need for mobile alarm clients has significantly been increased, as alarms need to be displayed on different gadgets instead of traditional big screens. Owing to significance of alarms in the industry, several state-of-the-art approaches have been introduced, suggesting various improvements. However, such approaches are only meant for specific industry and cannot be applied across-the-board. Furthermore, full support for mobile clients is not available. On the other hand, certain industrial software products like, SIMATIC WinCC, Genesis64 provide complete alarm solution including mobile client features, as well. However, such solutions are proprietary with higher costs, i.e., pay per tag model. Considering such limitations, this article proposes a novel Model-driven Alarms Framework (MAF) with mobile clients (Android and iOS) support. Particularly, MAF comprises an Alarm Profile for Industrial Control Systems (APICS) for the modeling of server, mobile clients and configuration requirements of alarms with remarkable simplicity. Furthermore, a complete open source Alarms Profile Transformation Engine (APTE) has been implemented to automatically generate alarm server (Kotlin), native Android (Kotlin) and iOS (Swift) alarm clients from APICS-compliant models. The feasibility of proposed framework is demonstrated through two case studies (flour mill and home automation). The results prove that MAF not only provides complete alarm server solution to generate bulk of alarms simultaneously, but also supports android and iOS alarm clients for efficient visualization. Besides that, MAF is easy to use, cost-effective and can be applied in wide-ranging industrial applications.

show abstract

Section: Figure 2 Architecture Of Alarm System In Industrial Controlmentioning

confidence: 99%

Section: ) Prediction Of Alarms Using Machine Learning Techniquesmentioning

confidence: 99%

See 1 more Smart Citation

A Model-Driven Alarms Framework (MAF) With Mobile Clients Support for Wide-Ranging Industrial Control Systems

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Also closed-related to plant processes, these records are bond to plant operation monitoring, and commonly are brought to the attention of operators. According to (Soares et al (2016)), in an industrial process monitoring, there is a correlation between variables during an abnormal situation and that can affect other variables simultaneously.…”

Section: Study Environment and Situationsmentioning

confidence: 99%

Towards the use of LSTM-based Neural Network for Industrial Alarm Systems

Silva

Pereira

2020

Anais Do Congresso Brasileiro De Automática 2020

View full text Add to dashboard Cite

With the coming of the complex industrial plants, the majority of control rooms sharing a common scenario involve operators faced amount of alarms during an overload. Besides, were often unable to determine which were important in an alarm widespread flooding scenario is an increasingly frequent situation. To solve this issue, proper handling of industrial processes requires designing, implementing and updating Process Malfunction Prediction (PMP) systemsable to give advice to the operators being a recommendation to make necessary adjustments in operating variables. From the wide range of models, we apply Recurrent Neural Networks (RNNs) using Long Short-Term Memory (LSTM) units, built was a regressor trained to predict the behavior of failures in an industrial process. The activity values of LSTM units can give recommendations for the monitoring of malfunctions. To this end, data mining and machinelearning techniques are used, which allow the implementation of a regression. The distinctive feature of PMP is that dynamically provides information using the data process. Further, proposed approach was evaluated in a simulated industrial process case study scenario. Lastly, the evaluation of the experimental results demonstrate the contribution of this work.

show abstract

“…For example, a statistical approach is proposed in Chen (2010) to reduce false alarms in multivariate statistical process control by focusing on those induced by random effects and using Bernoulli and Binomial distributions. A long-term application of an alarm management protocol is reported in Soares et al (2016) based on alarm prioritization to three large natural gas processing plants with a study of the possible advantages of employing many correlation analyses techniques in reducing the number of alarms. A new resilience measure is suggested in Yoon et al (2017), which considers false alarm rates and reliability and formulated in a probabilistic framework.…”

Section: Introductionmentioning

confidence: 99%

Decision making under uncertainty in the alarm systems response

Mohammed

Smail

Chebila

2019

IJQRM

View full text Add to dashboard Cite

Purpose The purpose of this paper is to develop an advanced decision-making support for the appropriate responding to critical alarms in the hazardous industrial facilities. Design/methodology/approach A fuzzy analytical hierarchy process is suggested by considering three alternatives and four criteria using triangular fuzzy numbers to handle the associated uncertainty. A logarithmic fuzzy preference programming (LFPP)-based nonlinear priority method is employed to analyze the suggested model. Findings A quantitative decision-making support is not only a necessity in responding to critical alarms but also easy to implement even in a relatively short reaction time. Confirmation may not be the appropriate option to deal with a critical alarm, even with the availability of the needed resources. Practical implications A situation related to a flammable gas alarm in a gas plant is treated using the developed model showing its practical efficiency and practicality. Originality/value The proposed model provides a rational, simple and holistic fuzzy multi criteria tool with a refined number of criteria and alternatives using an LFPP method to handle process alarms.

show abstract

Alarm management practices in natural gas processing plants

Cited by 29 publications

References 20 publications

A Model-Driven Alarms Framework (MAF) With Mobile Clients Support for Wide-Ranging Industrial Control Systems

A Model-Driven Alarms Framework (MAF) With Mobile Clients Support for Wide-Ranging Industrial Control Systems

Towards the use of LSTM-based Neural Network for Industrial Alarm Systems

Decision making under uncertainty in the alarm systems response

Contact Info

Product

Resources

About