Weidmüller (Germany)

Abstract The ability of industrial X-ray computed tomography (CT) to scan an object with several internal and external features at once causes increasing adoption in dimensional metrology. In order to evaluate the quality of a measurement value, the task-specific measurement uncertainty has to be determined. Currently, VDI/VDE 2630 part 2.1 gives a guideline to determine the uncertainty of CT measurements experimentally by conducting repeated measurements. This is costly and time-consuming. Thus, the aim is to determine the task-specific measurement uncertainty numerically by simulations (e. g. according to the guide to expression of uncertainty in measurement (GUM) Supplement 1). To achieve that, a digital twin is necessary. This contribution presents a simple first approach how a digital twin can be built. In order to evaluate this approach, a study comparing measurements and simulations of different real CT systems was carried out by determining the differences between the measurement results of the digital twin and of the measurement results of the real-world CT systems. The results have shown a moderate agreement between real and simulated data. To improve on this aspect, a standardized method to characterize CT systems and methods to implement CT parameters into the simulation with sufficient accuracy will be developed.

This paper describes the first techno-economical study for the deployment of very dense wireless networks for new Internet of things (IoT) applications. We propose mobile optical wireless communication (OWC) also known as light fidelity (LiFi) because it offers unique advantages for the IoT such as robustness through exclusive channel access in the unlicensed optical spectrum, lower latency, and enhanced security. We investigate the deployment of LiFi in an industrial scenario, as currently specified by ITU-T recommendation G.9991, including the required fixed backbone and assuming full coverage. We propose six different LiFi topologies and compare them in different industrial scenarios. The comparison is performed in terms of Bill of Material, and in terms of costs. We show that a fully wired topology requires higher installation costs but incurs in lower operational costs, resulting in lower total costs of ownership (TCO) than other wireless topologies. Furthermore, power consumption has been identified as the cost driver for all topologies (more than 200% of the initial investments), triggering the need for power-saving techniques. Last but not least, the separation of Optical Front Ends (OFEs) has been identified as a critical design parameter, as shorter separation requires higher costs but reduces the cost per delivered bitrate. We discuss promising ways for the deployment of very dense wireless networks as key to bringing LiFi into future IoT applications.

Abstract The integration of Open Platform Communications Unified Architecture into the field level holds potential to reduce manual configuration efforts by consistent information modelling and a uniform communication protocol. This paper discusses initiatives that are active in this context and reflects the status of the current specification work. The review focuses on the achievability of Quality of Service requirements, the adaption of established automation system concepts and the introduction of new concepts for automation systems at the field level. Using a remote input/output application as an example, we investigate the impact of new concepts for low-level system engineering. It is shown that besides a possible reduction of configuration steps, a loose coupling between the automation application, the used network infrastructure and the used devices can be achieved.

The Internet of Things currently is predominantly narrowband and cannot always guarantee high reliability and low latency. Future IoT applications such as flexible manufacturing, augmented reality and self-driving vehicles need sophisticated real-time processing units in the cloud to which mobile IoT devices are connected. These high-capacity links meet the requirements of the upcoming 6G systems and cannot be facilitated by the current mobile communication infrastructure. Light communication, which is also denoted as LiFi, offers huge amounts of spectrum, extra security and interference-free transmission. We present the current state-of-the-art of LiFi systems and introduce new features needed for future IoT applications. We propose a distributed Multiple-Input Multiple-Output topology with a fronthaul using plastic optical fiber. Such a system offers seamless mobility between the light access points and also to 5G, besides low latency and integrated positioning. Future LiFi development, implementation and efforts towards standardization are addressed in the EU ELIoT project which is presented here.

Deep Neural Networks (DNNs) have been recently in the focus of interest owing to their high-quality results in various application domains of big data analytics. The lack of end-to-end toolchains that can automatically optimize and translate a DNN algorithm, usually implemented in high-level languages, on FPGA-based platform hinders software developers to exploit such platforms. A few available toolchains, namely Xilinx DPU, have been optimized for computational expensive applications, e.g., real-time image and video processing, and therefore they introduce large performance overhead on lightweight applications, e.g., sensor-driven decision analytics. In this paper, we introduce the FLight framework to fully automatize the acceleration of lightweight DNN algorithms on FPGA-based embedded platforms. The framework takes a trained model in TensorFlow/Keras as input and generates an optimized synthesizable C++ version and maps it on the target platform using HLS tools. FLight is an easy-to-use framework and does not require deep knowledge of embedded system design since it automatically performs all the required mapping steps. FLight supports the acceleration of various deep learning algorithms, e.g., Feed-Forward Neural networks (FFNNs), Convolutional Neural Networks (CNNs), and sequential models like Recurrent Neural Networks (RNNs). We evaluated the applicability of FLight with various DNN models on both academic and industrial datasets. As a case study, we exploited the FLight framework for wind turbine condition monitoring using an industrial dataset provided by Weidmüller Monitoring Systems GmbH. The experimental results revealed the 40 times speedup compared to the Xilinx DPU framework.

The Industry 4.0 sector is evolving in a tremendous pace by introducing a set of industrial automation mechanisms tightly coupled with the exploitation of Internet of Things (IoT), 5G and Artificial Intelligence (AI) technologies. By combining such emerging technologies, interconnected sensors, instruments, and other industrial devices are networked together with industrial applications, formulating the Industrial IoT (IIoT) and aiming to improve the efficiency and reliability of the deployed applications and provide Quality of Service (QoS) guarantees. However, in a 5G era, efficient, reliable and highly performant applications' provision has to be combined with exploitation of capabilities offered by 5G networks. Optimal usage of the available resources has to be realised, while guaranteeing strict QoS requirements such as high data rates, ultra-low latency and jitter. The first step towards this direction is based on the accurate profiling of vertical industries' applications in terms of resources usage, capacity limits and reliability characteristics. To achieve so, in this paper we provide an integrated methodology and approach for benchmarking and profiling 5G vertical industries' applications. This approach covers the realisation of benchmarking experiments and the extraction of insights based on the analysis of the collected data. Such insights are considered the cornerstones for the development of AI models that can lead to optimal infrastructure usage along with assurance of high QoS provision. The detailed approach is applied in a real IIoT use case, leading to profiling of a set of 5G network functions.

5G and network function virtualization (NFV) have the potential to make factory networks more flexible and scalable. For this purpose, we defined use cases in previous work and demonstrated them using emulation-based prototype deployments. In this paper, we extend our previous work and deploy our developed vertical-specific network services and cloud-native virtualized network functions (VNFs) in a physical NFV testbed that is located on the shop floor to analyze the potential of NFV in real-life environments. We use the 5GTANGO service platform to deploy them on Kubernetes. Both are running on top of OpenStack.

In this contribution, we give an insight in our experiences in the technical and organizational realization of industrial analytics. We address challenges in implementing industrial analytics in real-world applications and discuss aspects to consider when designing a machine learning solution for production. We focus on technical and organizational aspects to make industrial analytics work for real-world applications in factory automation. As an example, we consider a machine learning use case in the area of industry compressors. We discuss the importance of scalability and reusability of data analytics pipelines and present a container-based system architecture.

We are addressing two related applications for the analysis of data from wind turbines. First, we consider time series data arising from virtual sensors in numerical simulations as employed during product development, and, second, we investigate sensor data from condition monitoring systems of installed wind turbines. For each application we propose a data analysis procedure based on dimensionality reduction. In the case of virtual product development we develop tools to assist the engineer in the process of analyzing the time series data from large bundles of numerical simulations in regard to similarities or anomalies. For condition monitoring we develop a procedure which detects damages early in the sensor data stream.