Industrial Cyber-physical Systems Research Articles

Robust Structure Identification of Industrial Cyber-Physical System From Sparse Data: A Network Science Perspective

Industrial cyber-physical systems (ICPSs) are deployed in many high-value facilities recently, and the monitoring of ICPS is more and more important. However, the prerequisite of ICPS monitoring is how to obtain an accurate network structure. In addition, the structure of ICPS may change over time and the observations data are limited and noisy. These situations make the ICPS network structure identification more difficult. In this article, we proposed the algorithm of temporal network identification from sparse data (ATNISD) to address these two issues simultaneously. First, we established the temporal network analysis model from the aspect of state equation and observation equation. Then, we analyze the characteristics of temporal networks in both time domain and space domain and propose a general framework of temporal networks structure identification, which is a combinatorial optimization problem. To improve the accuracy and alleviate the computational complexity, we decompose the combinatorial problem into small independent simple problems, which can be solved efficiently. The performance of the proposed algorithm is verified on synthetic evolutionary game dynamics on both homogeneous and heterogeneous temporal networks. The experimental results show that the proposed method can efficiently solve the problem of temporal networks structure identification from sparse data. Note to Practitioners—This article addresses the importance of network structure identification in industrial cyber-physical systems (ICPSs). The proposed method can effectively cope with the task of ICPS network structure identification in time-varying environments by exploiting the spatial and temporal features of networks in both time domain and space domain. The proposed algorithm can be implemented in typical slowing changing ICPS with or without observation noise, and it can decompose the combinatorial problem into small independent simple problems to improve the accuracy and release the computational complexity. Extensive simulation experiments demonstrate the accuracy and robustness of the proposed method for solving the structure identification task of temporal networks.

Guest Editorial: Industrial Cyber–Physical Systems—New Trends in Computing and Communications

The papers in this special section focus on industrial cyber-physical systems (CPS), with an emphasis on computing and communications applications. CPS systems are defined by integrating computation and communication facilities on the one hand and the monitoring and control of physical processes on the other hand. Industrial cyber–physical systems (ICPS) refer to the science and art of designing and using cyber–physical systems for industrial and process control applications, for example in smart factories, smart energy grids, smart transportation systems, smart cities, and several other areas. Many of these applications are time- and mission-critical and hence often require low latency and high reliability. In parallel, there has been a strong growth in integrating intelligence, often in the form of machine/deep learning, into applications, which also leads to vastly increasing computational requirements. These innovations then will be integrated into complex systems, which need to be properly engineered to become safe, reliable, trustworthy, and secure while at the same time being cost-efficient.

Permutation entropy based detection scheme of replay attacks in industrial cyber-physical systems

Although data integrity attack detections are critical to cyber-physical systems (CPSs), replay attack detection in industrial CPSs, especially data-based detection methods against replay attacks, has not been well-studied. Because it is difficult to distinguish replayed historical measurements and current measurements, replay attacks are hard to detect. In this paper, we propose a permutation entropy based detection scheme according to the complexity characteristics of sensor measurements. As sensor measurements generated during replay attacks present some sort of regularity, the significant decreasing of permutation entropy indicates the occurrence of replay attacks. Support vector data description (SVDD) is used to classify replay attacks according to permutation entropy of sensor measurements. Considering the manipulative attacker who mixes gaussian white noise or gaussian colored noise with measurements to bypass the entropy detection, we use wavelet analysis to denoise the measurements in advance. Finally, we demonstrate the efficiency of the proposed method through its application to a semi-physical simulation testbed. The experiment results show that our method can detect replay attacks accurately.

Engineering ethical behaviors in autonomous industrial cyber-physical human systems

This paper addresses the engineering of the ethical behaviors of autonomous industrial cyber-physical human systems in the context of Industry 4.0. An ethical controller is proposed to be embedded into these autonomous systems, to enable their successful integration in the society and its norms. This proposed controller that integrates machine ethics is realized through three main strategies that utilize two ethical paradigms, namely deontology, and consequentialism. These strategies are triggered according to the type of event sensed and the state of the autonomous industrial cyber-physical human systems, their combination being potentially unknown or posing ethical dilemmas. Two case studies are investigated, that deal with a fire emergency, and two different contexts i.e. one with an autonomous train, and one with an autonomous industrial plant, are discussed to illustrate the controller utilization. The case studies demonstrate the potential benefits and exemplify the need to integrate ethical behaviors in autonomous industrial cyber-physical human systems already at the design phase. The proposed approach, use cases, and discussions make evident the need to address ethical aspects in new efforts to engineer industrial systems in the context of Industry 4.0.

Use of augmented reality-enabled prototyping of cyber-physical systems for improving cyber-security education

The use of augmented reality-enabled scenarios in cybersecurity teaching is proposed in the article to respond to new requirements for the rapid adoption of new technologies and profound knowledge of cybersecurity issues by professionals. Implementation of project-type activities based on real cybersecurity issues in application fields of cyber-physical systems is suggested to improve the competence forming. A use-case of agricultural cyber-physical system of systems is discussed as a viable example of augmented reality-enabled prototyping of cybersecurity risk-aware architecture. The necessary steps are analysis of general and business-specific tasks on cybersecurity, creation of a list of competencies, formalized in educational standards and curricula, development of gaming scenarios for the formation of hard and soft skills, development of the scenario management system for AR interfaces. The system using AR tools can be easily adapted to different cybersecurity training activities. Industrial cyber-physical systems may be vulnerable due to insecure wireless connectivity, lack of encryption, inadequate access policy. The project-based learning complex is focused on the implementation of a data acquisition, storage and processing platform for new sensor networks and instruments. Representing all the diverse information on different layers will be greatly improved by use of the developed holographic projection AR tools.

Authenticated Key Agreement Scheme With User Anonymity and Untraceability for 5G-Enabled Softwarized Industrial Cyber-Physical Systems

With the tremendous growth of Information and Communications Technology (ICT), Cyber Physical Systems (CPS) have opened the door for many potential applications ranging from smart grids and smart cities to transportation, retail, public safety and networking, healthcare and industrial manufacturing. However, due to communication via public channel occurring among various entities in an industrial CPS (ICPS) with the help of the 5G technology and Software-Defined Networking (SDN), it poses several potential security threats and attacks. To mitigate these issues, we propose a new three-factor user authentication and key agreement scheme (UAKA-5GSICPS) for 5G-enabled SDN based ICPS environment. UAKA-5GSICPS allows an authorized user to access the real-time data directly from some designated Internet of Things (IoT)-based smart devices provided that a successful mutual authentication among them is executed via their controller node in the SDN network. It is shown to be robust against various potential attacks through detailed security analysis including the simulation-based formal security verification. A detailed comparative study with the help of experimental results shows that UAKA-5GSICPS achieves better trade-off among security and functionality features, communication and computation overheads as compared to other existing competing schemes.

BiDrac Industry 4.0 framework: Application to an Automotive Paint Shop Process

To implement an Industry 4.0 Framework in an ongoing industrial manufacturing process to prepare it for complex Artificial Intelligence use cases is not an easy task. The automotive industry is undergoing a large transformation due to a variety of disruptive factors through the introduction of CASE (Connected, Autonomous, Shared, Electrified) technologies. Production improvements will enable future development of CASE capabilities. This paper presents an Industry 4.0 framework, named BiDrac, that integrates computing, communication and control under an Industrial Cyber–Physical​ System (ICPS) ecosystem. It combines Artificial Intelligence and Industrial Internet of Things (IIoT) inside the Industry 4.0 paradigm for the Predictive Maintenance of an Automotive Paint Shop Process inside an Industrial Cloud (IC) as an open platform that provides community of cloud-based solutions to enable the development of new digital solutions. Several real use cases of the proposed platform have been included in the results section to illustrate the potential of the proposed framework.

Artificial Intelligence-Based Sensors for Next Generation IoT Applications: A Review

Sensors play a vital role in our daily lives and are an essential component for Internet of Things (IoT) based systems as they enable the IoT to collect data to take smart and intelligent decisions. Recent advances in IoT systems, applications, and technologies, including industrial Cyber-Physical Systems (CPSs), are being supported by a wide range of different types of sensors based on artificial intelligence (AI). These smart AI-based sensors are typically characterized by onboard intelligence and have the ability to communicate collaboratively or through the Internet. To achieve the high level of automation required in today’s smart IoT applications, sensors incorporated into nodes must be efficient, intelligent, context-aware, reliable, accurate, and connected. Such sensors must also be robust, safety- and privacy-aware for users interacting with them. Sensors leveraging advanced AI technologies, new capabilities have recently emerged which have the potential to detect, identify, and avoid performance degradation and discover new patterns. Along with knowledge from complex sensor datasets, they can promote product innovation, improve operation level, and open up novel business models. We review sensors, smart data processing, communication protocol, and artificial intelligence which will enable the deployment of AI-based sensors for next-generation IoT applications.

A survey on attack detection, estimation and control of industrial cyber–physical systems

Cyber–physical systems (CPSs) are complex systems that involve technologies such as control, communication, and computing. Nowadays, CPSs have a wide range of applications in smart cities, smart grids, smart manufacturing and intelligent transportation. However, with integration of industrial control systems with modern communication technologies, CPSs would be inevitably exposed to increasing security threats, which could lead to severe degradation of the system performance and even destruction of CPSs. This paper presents a survey on recent advances on security issues of industrial cyber–physical systems (ICPSs). We specifically discuss two typical kinds of attacks, i.e., Denial-of-Service (DoS) attack and Deception attack, and present recent results in terms of attack detection, estimation, and control of ICPSs. Classifications of current studies are analyzed and summarized based on different system modeling and analysis methods. In addition, advantages and disadvantage of various methodologies are also discussed. Finally, the paper concludes with some potential future research directions on secure ICPSs.

Siamese Neural Network Based Few-Shot Learning for Anomaly Detection in Industrial Cyber-Physical Systems

With the increasing population of Industry 4.0, both AI and smart techniques have been applied and become hotly discussed topics in industrial cyber-physical systems (CPS). Intelligent anomaly detection for identifying cyber-physical attacks to guarantee the work efficiency and safety is still a challenging issue, especially when dealing with few labeled data for cyber-physical security protection. In this article, we propose a few-shot learning model with Siamese convolutional neural network (FSL-SCNN), to alleviate the over-fitting issue and enhance the accuracy for intelligent anomaly detection in industrial CPS. A Siamese CNN encoding network is constructed to measure distances of input samples based on their optimized feature representations. A robust cost function design including three specific losses is then proposed to enhance the efficiency of training process. An intelligent anomaly detection algorithm is developed finally. Experiment results based on a fully labeled public dataset and a few labeled dataset demonstrate that our proposed FSL-SCNN can significantly improve false alarm rate (FAR) and F1 scores when detecting intrusion signals for industrial CPS security protection.

Softwarized Attention-Based Context-Aware Group Recommendation Technology in Event-Based Industrial Cyber-Physical Systems

Industrial cyber-physical systems are smart systems, which amalgamate the physical processes with computational capabilities to seamlessly capture, monitor and control the entities and scenarios in industrial environments. Among them, event-based industrial cyber-physical systems (EICPSs), such as Meetup and Plancast, have gained rapid developments. EICPSs provide event recommendation service for groups, which alleviates the information overload problem. However, existing group recommendation models in EICPSs focus on how to aggregate the preferences of group members, failing to model the complex and deep influence of contexts on groups. In this article, we propose an attention-based context-aware group event recommendation model (ACGER) in EICPSs. ACGER models the deep, nonlinear influence of contexts on users, groups, and events through multilayer neural networks. Especially, a novel attention mechanism is designed to enable the influence weights of contexts on users/groups change dynamically with the events concerned. Considering that groups may have completely different behavior patterns from group members, we acquire the preference of a group from two perspectives: indirect preference and direct preference. To obtain the indirect preference, we propose a method of aggregating preferences based on attention mechanism. Compared with existing predefined strategies, this method can flexibly adapt the strategy according to the events concerned by the group. To obtain the direct preference, we employ neural networks to learn it from group-event interactions. Furthermore, to make full use of rich user-event interactions in EICPSs, we integrate the context-aware individual recommendation task into ACGER, which enhances the accuracy of learning of user embeddings and event embeddings. Extensive experiments on three real datasets from Meetup and Douban event show that our model ACGER significantly outperforms the state-of-the-art models.

Optimal scheduling for maximizing information freshness and system performance in industrial cyber–physical systems

“Age of Information” is a newly introduced metric, getting vivid attention for measuring the freshness of information in real-time networks. This parameter has evolved to guarantee the timely reception of information from the latest status update received by a user from any real-time application. In this paper, we study a centralized, closed-loop, networked controlled industrial wireless sensor–actuator network for cyber–physical production systems. Here, we jointly address the problem of transmission scheduling of sensor updates and the restoration of an information flow-line after any real-time update, having hard-deadline, drops from it, resulting in a break in the loop. Unlike existing real-time scheduling policies that only ensure timely updates, this work aims to accomplish both the time-sensitivity and data freshness in new and regenerative real-time updates in terms of the ‘age’ of information. Here, the coexistence of both cyber and physical units and their individual requirements for providing the quality of service to the system, as a whole, seems to be one of the major challenges to handle. In this work, minimization of staleness of the time-critical updates to extract maximum utilization out of its information content and its effects on other network performances are thoroughly investigated. A greedy scheduling policy called “Deadline-aware Highest Latency First” has been used to solve this problem; its performance optimality is proved analytically. Finally, our claim is validated by comparing the results obtained by our algorithm with those of other popular scheduling policies through extensive simulations.

Running Industrial Workflow Applications in a Software-Defined Multicloud Environment Using Green Energy Aware Scheduling Algorithm

Industry 4.0 have automated the entire manufacturing sector (including technologies and processes) by adopting Internet of Things and cloud computing. To handle the workflows from Industrial Cyber-Physical systems, more and more data centers have been built across the globe to serve the growing needs of computing and storage. This has led to an enormous increase in energy usage by cloud data centers, which is not only a financial burden but also increases their carbon footprint. The private software defined wide area network (SDWAN) connects a cloud provider's data centers across the planet. This gives the opportunity to develop new scheduling strategies to manage cloud providers workload in a more energy-efficient manner. In this context, this article addresses the problem of scheduling data-driven industrial workflow applications over a set of private SDWAN connected data centers in an energy-efficient manner while managing tradeoff of a cloud provider' revenue. Our proposed algorithm aims to minimize the cloud provider's revenue and the usage of nonrenewable energy by utilizing the real-world electricity prices with the availability of green energy on different cloud data centers, where the energy consumption consists of the usage of running application over multiple data centers and transferring the data among them through SDWAN. The evaluation shows that our proposed method can increase usage of green energy for the execution of industrial workflow up to 3× times with a slight increase in the cost when compared to cost-based workflow scheduling methods.

Describing Structure and Complex Interactions in Multi-Agent-Based Industrial Cyber-Physical Systems

The description of structure and complex interactions in Multi-agent-based Industrial Cyber-physical (MAS-ICPS) systems has been elusively addressed in the literature. Existing works, grounded on model-based engineering, have been successful at characterizing and solving system integration problems. However, they fail to describe accurately the collective and dynamic execution behaviour of large and complex industrial systems, particularly in more discrete production domains, such as: automotive, home appliances, aerospace, food and beverages, etc. In these domains, the execution flow diverts dynamically due to production disturbances, custom orders, fluctuations in demand in mixed model production, faults, quality-control and product rework, etc. These dynamic conditions require re-allocation and reconfiguration of production resources, redirection of production flows, re-scheduling of orders, etc. A meta-model for describing the structure and complex interactions in MAS-ICPS is defined in this paper. This contribution goes beyond the State-Of-The-Art (SOTA) as the proposed meta-model describes structure, as many other literature contributions, but also describes the execution behaviour of arbitrarily complex interactions. The previous is achieved with the introduction of general execution flow control operators in the meta-model. These operators cover, among other aspects, delegation of the execution flow and dynamic decision making. Additionally, the contribution also goes beyond the SOTA by including validation mechanisms for the models generated by the meta-model. Finally, the contribution adds to the current literature by providing a meta-model focusing on production execution and not just on describing the structural connectivity aspects of ICPSs.

Securing industrial communication with software-defined networking.

Industrial Cyber-Physical Systems (CPSs) require flexible and tolerant communication networks to overcome commonly occurring security problems and denial-of-service such as links failure and networks congestion that might be due to direct or indirect network attacks. In this work, we take advantage of Software-defined networking (SDN) as an important networking paradigm that provide real-time fault resilience since it is capable of global network visibility and programmability. We consider OpenFlow as an SDN protocol that enables interaction between the SDN controller and forwarding plane of network devices. We employ multiple machine learning algorithms to enhance the decision making in the SDN controller. Integrating machine learning with network resilience solutions can effectively address the challenge of predicting and classifying network traffic and thus, providing real-time network resilience and higher security level. The aim is to address network resilience by proposing an intelligent recommender system that recommends paths in real-time based on predicting link failures and network congestions. We use statistical data of the network such as link propagation delay, the number of packets/bytes received and transmitted by each OpenFlow switch on a specific port. Different state-of-art machine learning models has been implemented such as logistic regression, K-nearest neighbors, support vector machine, and decision tree to train these models in normal state, links failure and congestion conditions. The models are evaluated on the Mininet emulation testbed and provide accuracies ranging from around 91-99% on the test data. The machine learning model with the highest accuracy is utilized in the intelligent recommender system of the SDN controller which helps in selecting resilient paths to achieve a better security and quality-of-service in the network. This real-time recommender system helps the controller to take reactive measures to improve network resilience and security by avoiding faulty paths during path discovery and establishment.

Модель активного мониторинга как основа управления безопасностью промышленных киберфизических систем

Purpose of the article – creating a model of active security monitoring that meets the current conditions for industrial cyber-physical systems. Methods of the study. The work concretizes the relationship between monitoring and security management processes in the form of a set of monitoring functions for the management implementation. The active monitoring model is defined using a threefold mapping of security goals, mathematical methods, and security object data. Based on this mapping, the paper formulates the reachability and minimality conditions of monitoring technological components (data and mathematical methods) with respect to security purposes and tasks. Results of the study. The article contains a management and control scheme based on the proposed active monitoring model. The workflow includes steps to evaluate, adjust the set of methods used, adjust the data collected, and verify that the security purpose has been achieved. Active monitoring of information security of digitalized objects, including industrial cyber-physical systems, will increase awareness in security management and provide the required level of protection in changing conditions.

Machine learning in SDN networks for secure industrial cyber physical systems: a case of detecting link flooding attack

Machine learning in SDN networks for secure industrial cyber physical systems: a case of detecting link flooding attack

Power Control for Cognitive Users of Perception Layer in Complex Industrial CPS Based on DQN

Wireless communication is a significant auxiliary technology of data transmission for industrial Cyber-Physical system (CPS). While for the complex industrial scenario of coal mine with long and narrow laneway, lifetime of wireless perception nodes is a potential and nonnegligible problem for safety production. In order to deal with this problem, a power control algorithm based on deep Q network (DQN) is adopted to train micro base station (MBS) by two steps so that the MBS can learn an optimal policy to help the cognitive users (CUs) communicate with a proper transmit power. Firstly, the selection range of transmit power for CUs is calculated by the lower bound of Signal-to-Interference plus-Noise Ratio (SINR) to guarantee the transmission condition of both users. Then, the power control problem is modeled as a Markov Decision Process (MDP) with unknown transition function, where the energy consumption is decreased by giving the upper bound of CUs’ SINR or threshold of transmit power in formulation of reward. In modeled MDP, the system state, which collected by primary users (PUs) and fed back to MBS, is reduced dimension by method of principal component analysis and then treated as the input of DQN. After that, DQN is used to train a power control optimal policy by minimizing the loss function. Simulation results demonstrate that the proposed power control algorithm based on DQN has a good performance that the average transition step and energy utility are 3.56 and 1580h, which is better than the existing solutions.

Hyperconnected Architecture for High Cognitive Production Plants

The HyperCOG project addresses the full digital transformation of process industry through an innovative Industrial Cyber-Physical System and Data Analytics. It is based on advanced technologies that enable the development of a hyperconnected network of digital nodes. The nodes can catch outstanding streams of data in real-time, which together with the high computing capabilities, provide sensing, knowledge and cognitive reasoning, making companies robust in the face of variant scenarios. The breaking-edge system proposed in this work is validated on productivity, environmental and replicability aspects on three use cases of three different sectors: steel, cement and chemical.

An Automated Sensor Fusion Approach for the RUL Prediction of Electromagnetic Pumps

The remaining useful life (RUL) prediction of industrial cyber-physical system components demands the use of reliable prognostics parameters and frameworks. Against the traditional use of a single measure of degradation, data from multiple sensors provide abundant characteristic information for modeling, assessing, and extracting useful parameters via appropriate signal processing and sensor fusion methods. This study introduces a multi-sensor prognostics approach which merges highly prognosible statistical features from vibrational and pressure sensor measurements after a multi-level wavelet decomposition of the signals. The prognostic algorithm presented in this work for solenoid pump RUL prediction is a multi-objective genetic algorithm-optimized long short-term memory (MOGA-LSTM) which accepts the fused sensor features as input and returns the RUL of the pump as output. The framework was tested on a run-to-failure experiment on a VSC63A5 Solenoid pump following a significant pump malfunction caused by a clogged suction filter after the test. Using standard prognostic performance evaluation metrics, the performance of the prognostics framework was compared with other reliable state-of-the-art methods with a remarkable comparative advantage in addition to better automation potentials for real-time condition monitoring and RUL prediction.