This contribution introduces a framework for the fault detection and healing of chemical processes over wireless sensor networks. The approach considers the development of a hybrid system which consists of a fault detection method based on machine learning, a wireless communication model and an ontology-based multi-agent system with a cooperative control for the process monitoring.
Industry 4.0 is transforming chemical processes into complex, smart cyber-physical systems, by the addition of elements such as smart sensors, Internet of Things, big data analytics or cloud computing. Modern engineering systems and manufacturing processes are operating in highly dynamic environments, and exhibiting scale, structure and behaviour complexity. Under these conditions, plant operators find it extremely difficult to manage all the information available, infer the desired conditions of the plant and take timely decisions to handle abnormal operation1. Human beings acquire information from the surroundings through sensory receptors for vision, sound, smell, touch, and taste, the Five Senses. The sensory stimulus is converted to electrical signals as nerve impulse data communicated with the brain. When one or more senses fail, the humans are able to re-establish communication and improve the other senses to protect from incoming dangers. Furthermore, a mechanism of ‘reasoning’ has been developed during evolution, which enable analysis of present data and generation of a vision of the future, which might be called the Sixth Sense. As industrial processes are already equipped with five senses: ‘hearing’ from acoustic sensors, ‘smelling’ from gas and liquid sensors, ‘seeing’ from camera, ‘touching’ from vibration sensors and ‘tasting’ from composition monitors, the Sixth Sense could be achieved by forming a sensing network which is self-adaptive and self-repairing, carrying out deep-thinking analysis with even limited data, and predicting the sequence of events via integrated system modelling.
This contribution introduces the development of an intelligent monitoring and control framework for chemical process, integrating the advantages of Industry 4.0 technologies, cooperative control and fault detection via wireless sensor networks.
This paper investigates the challenging fault prediction problem in process industries that adopt autonomous and intelligent cyber-physical systems (CPS), which is in line with the emerging developments of industrial internet of things (IIoT) and Industry 4.0. Particularly, we developed an end-to-end deep learning approach based on a large volume of real-time sensory data collected from a chemical plant equipped with wireless sensors. Firstly, a novel recursive architecture with multi-lookback inputs is proposed to perform autoregression on imbalanced time-series data as a preliminary prediction. In this process, a novel learning algorithm named recursive gradient descent (RGD) is developed for the proposed architecture to reduce cumulative prediction uncertainties. Subsequently, a classification model based on temporal convolutions over multiple channels with decay effect is proposed to perform multi-class classification for fault root cause identification and localization. The overall network is named the cumulative uncertainty reduction network (CURNet), for its superior capacity in reducing prediction uncertainties accumulated over multiple prediction steps. Performance evaluations show that CURNet is able to achieve superior performance especially in terms of fault prediction recall and fault type classification accuracy, compared to the existing techniques.
Industry 4.0 aims to transform chemical and biochemical processes into intelligent systems via the integration of digital components with the actual physical units involved. This process can be thought of as the addition of a central nervous system with a sensing and control monitoring of components and regulating the performance of the individual physical assets (processes, units, etc.) involved. Established technologies central to the digital integrating components are smart sensing, mobile communication, Internet of Things, modelling and simulation, advanced data processing, storage and analysis, advanced process control, artificial intelligence and machine learning, cloud computing, and virtual and augmented reality. An essential element to this transformation is the exploitation of large amounts of historical process data and large volumes of data generated in real-time by smart sensors widely used in industry. Exploitation of the information contained in these data requires the use of advanced machine learning and artificial intelligence technologies integrated with more traditional modelling techniques. The purpose of this paper is twofold: a) to present the state-of-the-art of the aforementioned technologies, and b) to present a strategic plan for their integration toward the goal of an autonomous smart plant capable of self-adaption and self-regulation for short- and long-term production management.
Recently, the fifth-generation (5G) cellular system has been standardised. As opposed to legacy cellular systems geared towards broadband services, the 5G system identifies key use cases for ultra-reliable and low latency communications (URLLC) and massive machine-type communications (mMTC). These intrinsic 5G capabilities enable promising sensor-based vertical applications and services such as industrial process automation. The latter includes autonomous fault detection and prediction, optimised operations and proactive control. Such applications enable equipping industrial plants with a sixth sense (6S) for optimised operations and fault avoidance. In this direction, we introduce an inter-disciplinary approach integrating wireless sensor networks with machine learning-enabled industrial plants to build a step towards developing this 6S technology. We develop a modular-based system that can be adapted to the vertical-specific elements. Without loss of generalisation, exemplary use cases are developed and presented including a fault detection/prediction scheme, and a sensor density-based boundary between orthogonal and non-orthogonal transmissions. The proposed schemes and modelling approach are implemented in a real chemical plant for testing purposes, and a high fault detection and prediction accuracy is achieved coupled with optimised sensor density analysis.