A Model-Driven Framework for Process Control Alarms with Mobile Client Support

Abstract

In the past decade, the industrial control systems have experienced rapid changes due to the continuous advancements of software technology. The domain of alarm systems is undergoing deep interest in the research community recently. It is pondered interdisciplinary, as it crosses several areas, like industry automation systems, process control systems (PCS), smart homes, context awareness and involves human factor. Human operators oversee the processes involved in the industries and with a large amount of information displayed on their system, it is easy for an operator to miss important information needed in making critical control decisions and actions. An Alarm system is considered an acute member of PCS. It notifies the human operator with a meaningful message, in time, to take reactive measures successfully. The integrity of an alarm system is entailed for the performance and safety of a process control system. Ergonomically designed visualization of an alarm is a key factor for an operator to perform efficiently. And this along with the human operator are the two strong pillars of an alarm system. Most of the available alarm systems are designed using object-oriented or process-oriented way that leads to poor maintenance and makes it hard to scale at large. Currently, there is no opensource system available that is compliant with major OPC-compatible data acquisition approaches, provide a complete server implementation along with the Android Client Support. We have proposed a Model-Driven Approach to develop Alarm System for Process Control Systems (MDAASPCS). Our framework enables a user to easily design an alarm system for PCS. It provides simple, open-source, reusable, easy to integrate with native mobile client solution for an Alarm system. In the first step, a Unified Modeling Language (UML) Profile (i.e. Alarm for Process Control (APCS) Profile) is developed by extending UML-metamodel to adapt to the concepts of an Alarm system. Secondly, a transformation engine, named Model-Driven Alarms Server and Mobile Clients for Process Control System (MDASMCPCS), is developed using Acceleo. MDASMCPCS performs Model-to-Text (M2T) transformation to automatically generate the alarms server, system translation artifact and native mobile clients (i.e. iOS and Android both). Finally, we validated our proposed framework against two extensive case studies i.e. Home Automation and Flour Mill System. The experimental results prove that our proposed framework provides complete HMI development solution at higher level of abstraction by reducing the complexity and cost of development.