Im heutigen Produktionsumfeld erhöhen steigender Wettbewerbsdruck sowie Kundenanforderungen die Notwendigkeit, Terminpläne genau einzuhalten, speziell in der Phase vor dem Start der Produktion (SOP). Aus diesem Grund werden automatisierte Produktionssysteme in verschiedensten Industrien mittels virtueller Inbetriebnahme umfassend getestet, um Fehler bereits vor der realen Inbetriebnahme zu erkennen und somit Verzögerungen während der realen Inbetriebnahme zu vermeiden. Die Inbetriebnahmezeit kann dadurch um bis zu 30 % verkürzt werden. Grundlage der virtuellen Inbetriebnahme ist das virtuelle Inbetriebnahme-Modell. Während der realen Inbetriebnahme werden Geometrien, Schaltpläne und speicherprogrammierbare Steuerungen (SPS) durch Optimierungen verändert. Diese Änderungen fließen aber meist nicht in die Simulation ein, sodass nach erfolgter Inbetriebnahme die Simulation nicht mehr der realen Anlage entspricht. Infolgedessen bleibt das mögliche Potential des virtuellen Modells in der Betriebsphase von Anlagen ungenutzt. Das Ziel der vorliegenden Dissertation ist es, eine Vorgehensweise zu erarbeiten, mit der das virtuelle Inbetriebnahme-Modell zu einem prozessaktuellen virtuellen Maschinenzwilling weiterentwickelt werden kann. Dieser soll während der Betriebsphase dazu dienen, zukünftige Maschinenzustände nachzubilden und somit z. B. Ausbringungsreduktionen vorherzusagen. Dazu wird eine Methode entwickelt, die es ermöglicht, Alterungsphänomene von Teilprozessen im Maschinenzwilling zu simulieren, deren Auswirkungen darzustellen und daraus Erkenntnisse für Produktion und Instandhaltung abzuleiten. Die Kombination von realen und simulierten Erkenntnissen erzeugt einen hybriden Vorhersageansatz als Werkzeug für die Instandhaltung. Anschließend wird die Vorgehensweise zur Weiterentwicklung sowie die Methode zur Simulation von Alterungseffekten an einem realen Anwendungsfall demonstriert und validiert. Dabei handelt es sich um einen Engpassprozess in der Produktion der neuesten Generation von Elektromotoren der BMW AG. Unvorhergesehene Ausbringungsverringerungen werden auf diese Weise minimiert und die Effektivität der Gesamtanlage bleibt erhalten. Die Produktion bleibt damit in der Lage, ihre Produkte zur richtigen Zeit, in der richtigen Qualität und Quantität, und dem richtigen Preis an den Kunden auszuliefern. / In today’s production environment, increased competition, and growing customer demands have made it ever more important to meet schedules, especially in the phase before the start of production (SOP). For this reason, automated production systems in various industries are thoroughly tested by using virtual commissioning in order to detect and eliminate errors at an early stage – before the real commissioning on site. This helps to avoid delays and can reduce commissioning times by up to 30 %. The basis of the virtual commissioning is the virtual model. During real commissioning, geometries, electrical plans (E-plans) and programmable logic controllers (PLC) are optimized and modified. However, these changes usually remain unconsidered as not being updated in the simulation, resulting in a model that no longer reflects the real plant. As a consequence, the virtual model is no longer used in the production phase. The aim and objective of this dissertation is to create a general procedure that allows the further development of the virtual commissioning model into a process-actual virtual machine twin. This virtual machine twin will serve during production to emulate future machine states in order to predict for example output reductions. To this end, a method-ology will be defined to simulate aging effects of critical processes and to determine the impact on the overall process. The results are used to derive recommendations for production and maintenance. The combination of real and simulated outcomes establishes a so-called 'Hybrid Predictive Maintenance System'. Ultimately, the general procedure for the further development and the methodology for simulating aging effects will be demonstrated and validated on a critical process in the electric engines production of BMW AG. As a result, unexpected downtimes are minimized, and the overall equipment effectiveness (OEE) is ensured. This enables the production to deliver products at the right time, in the right quality and quantity, and at the right price.

Выпускная квалификационная работа на тему «Разработка интеллектуального цифрового двойника для видеоигр на примере использования Unity ML-Agent» посвящена созданию цифрового двойника для видеоигр с использованием технологии Unity ML-Agents. Основная цель работы — сократить время и повысить качество процесса тестирования видеоигр путем использования моделей машинного обучения для создания интеллектуальных двойников.Работа состоит из трех основных разделов: 1. Анализ применения машинного обучения и искусственного интеллекта в видеоиграх В этом разделе рассматриваются основные методы и инструменты, используемые для создания ИИ в видеоиграх, а также анализируются существующие решения и их практическое применение. 2. Практическая реализация создания цифрового двойника для игрового симулятора. Проводится анализ инструмента Unity ML-Agents, описываются его ключевые компоненты и методы обучения агентов. Также описывается процесс разработки прототипа цифрового двойника, и предлагается методология оценки его эффективности. 3. Изучение особенностей внедрения разработанного цифрового двойника и оценка его эффективности для автоматизации процесса тестирования. В этом разделе описывается эксперимент по использованию цифрового двойника для автоматизации тестирования, оценивается его эффективность и рассматриваются условия и ограничения применения данного инструмента. Научная новизна работы заключается в предложении применения методов машинного обучения для автоматизации ручного труда в процессе создания и тестирования видеоигр. В результате исследования было разработано и протестировано решение, позволяющее значительно сократить трудозатраты на тестирование и повысить его качество. / The graduation qualification work on the topic "Development of an Intelligent Digital Twin for Video Games using Unity ML-Agent" is dedicated to creating a digital twin for video games using Unity ML-Agents technology. The main goal of the work is to reduce the time and improve the quality of the video game testing process by using machine learning models to create intelligent twins. The work consists of three main sections: 1. Analysis of the Application of Machine Learning and Artificial Intelligence in Video Games. This section examines the main methods and tools used to create AI in video games, as well as analyzes existing solutions and their practical applications. 2. Practical Implementation of Creating a Digital Twin for a Game Simulator. This section analyzes the Unity ML-Agents tool, describes its key components and methods for training agents. It also outlines the process of developing a digital twin prototype and proposes a methodology for evaluating its effectiveness. 3. Study of the Features of Implementing the Developed Digital Twin and Assessing its Effectiveness for Automating the Testing Process. This section describes an experiment using the digital twin for automated testing, evaluates its effectiveness, and considers the conditions and limitations of applying this tool. The scientific novelty of the work lies in proposing the use of machine learning methods to automate manual labor in the process of creating and testing video games. As a result of the research, a solution was developed and tested, which significantly reduces labor costs for testing and improves its quality.

Die Nutzung von Augmented Reality (AR) stellt eine Vielzahl von Lösungsansätzen für Herausforderungen unterschiedlicher Bereiche, z.B. der Industrie, Medizin, Unterhaltungsmedien und Bildung, zur Verfügung. Speziell werden hier die Möglichkeiten von AR in der Fertigungsindustrie, genauer bei der Abnahme elektrischer Antriebe der Siemens AG, betrachtet. Dabei werden unter anderem messbare Größen wie Drehmoment, Ankerstrom, -spannung, Drehzahl, Schwingungen, Fertigungstoleranzen etc. gemessen, aber auch nicht quantifizierbare Größen wie Sichtprüfungen von Schweißnähten und Lackierung in Betracht gezogen. Angestrebt wird eine Methode, welche es zukünftig ermöglicht, die Inbetriebnahme ortsunabhängig zu verfolgen und Inspektionen aus der Ferne durchzuführen. Die Herausforderung liegt dabei darin, eine zumindest gleichwertige, wenn nicht sogar verbesserte Erfahrung und Prüfergebnisvalidität für die Beteiligten gegenüber der vor Ort Abnahme zu erzeugen. Im Rahmen des Forschungsvorhabens wurde in interdisziplinärer Zusammenarbeit zwischen Ingenieur/-innen und Sozialwissenschaftler/-innen ein Demonstrator-Konzept für eine ortsunabhängige Maschinenabnahme entwickelt.

The Master’s thesis deals with design of digital twins of machine tools and application of designed procedure on a case study – digital twin of the machine tool MCV 754 Quick. In the first part are described current roles of digital twins in the field of machine tools and also hardware and software options for creating them. Software and hardware tools by Siemens (Mechatronics Concept Designer and SIMIT) are used for the case study. The complex procedure of creating the digital twin is designed in the second part, starting with preparation of a model, ending with the virtual commissioning of the machine tool MCV 754 Quick. The procedure is described in detail, including preparation of 3D model of the machine tool, assignment of physical properties and joints, preparation of PLC, connection all the elements of the whole system: Sinumerik – SIMIT – Mechatronics Concept Designer and controlling the twin via CNC automation system. Output of the thesis is the virtually commissioned machine tool, capable of being controlled by Sinumerik 840D sl. This is the first step in development of the full-fledged digital twin of machine tool, which can be used for testing the functionality and capability of this new technology in industry.

Den här rapporten behandlar ett projekt utfört i kandidatkursen TDDD96 - Kandidatprojekt i programvaruutveckling. Projektet är utfört av åtta studenter från datateknik- ochmjukvaruteknikprogrammen på Linköpings universitet (LiU) för kunden från institutionen för medicinsk teknik (IMT) på LiU. Syftet med projektet var att ta fram en prototypmed ett grafiskt gränssnitt för uppvisning och simulering av hur kroppen påverkas av t.ex.träning, sömn och kostintag. Modellerna som simulerats är framtagna av forskningsgrup-pen integrativ system biologi för institutionen för medicinsk teknik LiU och resultatet är i form av dataändringar och grafer i programmets gränssnit.

Digital twins for industrial control systems have gained significant interest over recent years. This attention is mainly because of the advanced capabilities offered by digital twins in the areas of simulation, optimization, and predictive maintenance. Some recent studies discuss the possibility of using digital twins for intrusion detection in industrial control systems. To this end, this thesis aims to propose a security framework for industrial control systems including its digital twin for security monitoring and a machine learning-based intrusion detection system for real-time intrusion detection. The digital twin solution used in this study is a standalone simulation of an industrial filling plant available as open-source. After thoroughly evaluating the implementation aspects of the existing knowledge-driven open-source digital twin solutions of industrial control systems, this solution is chosen. The cybersecurity analysis approach utilizes this digital twin to model and execute different realistic process-aware attack scenarios and generate a training dataset reflecting the process measurements under normal operations and attack scenarios. A total of 23 attack scenarios are modelled and executed in the digital twin and these scenarios belong to four different attack types, naming command injection, network DoS, calculated measurement injection, and naive measurement injection. Furthermore, the proposed framework also includes a machine learning-based intrusion detection system. This intrusion detection system is designed in two stages. The first stage involves an offline evaluation of the performance of eight different supervised machine learning algorithms on the labelled dataset. In the second stage, a stacked ensemble classifier model that combines the best performing supervised algorithms on different training dataset labels is modelled as the final machine learning model. This stacked ensemble model is trained offline using the labelled dataset and then used for classifying the incoming data samples from the digital twin during the live operation of the system. The results show that the designed intrusion detection system is capable of detecting and classifying intrusions in near real-time (0.1 seconds). The practicality and benefits of the proposed digital twin-based security framework are also discussed in this work. / Digitala tvillingar för industriella styrsystem har fått ett betydande intresse under de senaste åren. Denna uppmärksamhet beror främst på de avancerade möjligheter som digitala tvillingar erbjuder inom simulering, optimering och förutsägbart underhåll. Några färska studier diskuterar möjligheten att använda digitala tvillingar för intrångsdetektering i industriella styrsystem. För detta ändamål syftar denna avhandling till att föreslå ett säkerhetsramverk för industriella styrsystem inklusive dess digitala tvilling för säkerhetsövervakning och ett maskininlärningsbaserat intrångsdetekteringssystem för intrångsdetektering i realtid. Den digitala tvillinglösningen som används i denna studie är en fristående simulering av en industriell fyllningsanläggning som finns tillgänglig som öppen källkod. Efter noggrann utvärdering av implementeringsaspekterna för de befintliga kunskapsdrivna digitala tvillinglösningarna med öppen källkod för industriella styrsystem, väljs denna lösning. Cybersäkerhetsanalysmetoden använder denna digitala tvilling för att modellera och exekvera olika realistiska processmedvetna attackscenarier och generera en utbildningsdataset som återspeglar processmätningarna under normala operationer och attackscenarier. Totalt 23 angreppsscenarier modelleras och utförs i den digitala tvillingen och dessa scenarier tillhör fyra olika angreppstyper, namnskommandoinjektion, nätverks -DoS, beräknad mätinjektion och naiv mätinjektion. Dessutom innehåller det föreslagna ramverket också ett maskininlärningsbaserat system för intrångsdetektering. Detta intrångsdetekteringssystem är utformat i två steg. Det första steget innebär en offline -utvärdering av prestanda för åtta olika algoritmer för maskininlärning övervakad på den märkta datauppsättningen. I det andra steget modelleras en staplad ensemble -klassificerarmodell som kombinerar de bäst presterande övervakade algoritmerna på olika etiketter för utbildningsdataset som den slutliga modellen för maskininlärning. Denna staplade ensemblemodell tränas offline med hjälp av den märkta datauppsättningen och används sedan för att klassificera inkommande dataprover från den digitala tvillingen under systemets levande drift. Resultaten visar att det konstruerade intrångsdetekteringssystemet kan upptäcka och klassificera intrång i nära realtid (0,1 sekunder). Det praktiska och fördelarna med den föreslagna digitala tvillingbaserade säkerhetsramen diskuteras också i detta arbete.

Virtualization of products means the representation of some of their properties by models. In a stronger digitalized world, these models will gain a much broader use than models had in engineering so far. Even for one modelling aspect different models of the same product will be used, depending on the specific need of the model user. That need may change in the course of product life, between first product concepts till over the different phases of development, to product use, maintenance, or even recycling. Since a digitalized world use of these diverse models will not be limited to experts model consistency will play a much stronger role. Model hierarchies will play a stronger role and can serve also as means for teaching product users a deeper understanding of product properties. A consistent model hierarchy leading from a simple to a more advanced property representation can support this learning process. In this paper perturbation methods are analyzed as a means for setting up model hierarchies in a consistent manner. This is studied by models for the behavior of a electrohydraulic drive, which consists of a variable speed motor, a pump, a double stroke cylinder and a counterbalance valve. Model hierarchy is achieved by model reduction in the sense of perturbation theory. The use of these different models for different questions in a system design context and their interrelations are exemplified.

Over the last few decades, the complexity and variety of automation systems have increased dramatically. Commissioning has grown more and more critical for the entire industry. Conventional commissioning is time-consuming and expensive. It's always been a challenge in manufacturing to put new designs into production or implement new technologies, control codes, or tactics. In Virtual Commissioning (VC), control programs of the physical system's Digital Twin (DT) can be validated in Software-in-the-Loop (SIL) before the actual commissioning. The emergence of new VC tools and methods has become a tremendous advantage, bringing the values of shorter duration, flexibility, and lower risks to the commissioning process. In this thesis, advanced process control was developed using the software Matlab and Simulink in conjunction with the engineering tools S7-PLCSIM Advanced and STEP 7 TIA Portal to conduct VC. A VC approach with four key steps is taken to evaluate the possibility of validating advanced process control. The steps are modeling DT of a rolling mill, model-based control design, simulation model development in Simulink, communication between the simulation model and the PLC program using S-7 TIA Portal, and PLCSIM Advanced. Also, a simulated Human-Machine Interface was designed to operate and visualize the process. VC of the rolling mill process was verified and validated by Model-in-the-Loop (MIL) and SIL simulation. The simulation gives satisfactory results as both MIL and SIL show identical outputs of the process.

Background: The Digital Twin is an increasingly trending technology that utilizes many of the key technologies relevant to the digitalization of industries and Industry 4.0, such as AI, Big Data, and IoT, to bring a virtual asset and a physical asset together to perform analysis and execute real-time decision making backed up by data (Grieves & Vickers, 2017; Rasheed et al., 2020). Current research on the benefits and challenges of Digital Twins in process industries is sparse and under-developed, and the significance of Digital Twins in their operational lifecycle phase requires investigation (Perno et al., 2020; Schleich et al., 2019; Sjödin, 2013). Research Questions: RQ1: What are the benefits of industrial process plant Digital Twins for companies in process industries? RQ2: What are the challenges that companies will face with industrial process plant Digital Twins in process industries? Literature Review: A theoretical framework was developed based on the applications, challenges, and enablers researched in other industries which acted as a foundation for the data collection process. Research Method: The thesis follows a qualitative research approach. A single-case study was employed with a large equipment supplier in the mineral processing industry, where data was collected through 17 semi-structured interviews with people possessing in-depth knowledge about the needs of the mineral processing industry. Findings: Five main benefits of industrial process plant Digital Twins in process industries were identified, Process performance, Monitoring and control, Predictive maintenance and scheduling, Business opportunities, and Safety. Correspondingly, four main challenges were identified, Perception and presentation, Privacy and security, Data management and performance, and Mutual scope and focus. Conclusions: The thesis project's managerial and theoretical implications include providing equipment suppliers in process industries with what benefits there lie with industrial process plant Digital Twins and which challenges must be overcome. By providing these benefits and challenges to equipment suppliers, they can prioritize which of these are most important to consider in their situation. The thesis contributes to the research of Digital Twins and adds to the sparse existing knowledge of what the benefits and challenges are of industrial process plant Digital Twins in process industries. The main limitation is the absence of process plant companies during the data collection phase.

Our society moves towards digitalization and the industry is not an exception. Siemenshas developed a wireless condition monitoring device called Simotics Connect in order tohelp them to move forward in the world of digitalization. The Simotics Connect has threeinbuilt sensors. One for temperature, one for vibrations and one for magnetic flux density,a product that is new in the market. This master thesis has investigated its usability forthird party motors, which has not been done.Four areas were investigated, the status in the current market, creating a motorgeometry estimation based on nameplate data, presenting a temperature model to calculatea motor’s cross section temperature and, finally, proposed a stator current model using themagnetic field measurement.Market research has shown that a space for the Simotics Connect to thrive in mostdefinitely exists.The motor geometry estimation, that is based on preliminary electromagnetic sizing,creates a digital twin for the motor that has sufficient accuracy as a tool when calculatinge.g. temperature calculations but lacks accuracy for more advanced and sensitivecalculations e.g for magnetic flux density measurement usability.The temperature model that is presented shows great accuracy when calculating thecross section temperature in the stator but the accuracy decreases for the cross sectiontemperature in the rotor.A stator current model is proposed using a proportional relationship between themagnetic flux density and stator current. The results indicates a linear relationship, thoughusing the digital twin to calculate the proportional constant were concluded to not beaccurate enough. / Sammhället rör sig idag mot digitalisering och industrin är ej ett undantag. Siemens harutvecklat en trådlös underhållsmätare kallad Simotics Connect för att hjälpa dem strävamot en värld inom digitalisering. Simotics Connect hat tre inbyggda sensorer. En för temperatur,en för vibrationer och en för magnetisk flödestäthet, vilket är nytt på marknaden.Detta masterprojekt har undersökt användningen av Simotics Connect för tredjepartsmotorer,vilket ej har gjorts tidigare.Fyra områden undersöktes, statusen på den nuvarande marknaden, en motorgeometriuppskattningmodellbaserad på namnskylsdata, en temperaturmodell för att beräknamotorns tvärsnittstemperatur och, slutligen, en statorströmmodell som använder sig avmagnetiska flödestäthetsmätningen.Marknadsundersökningen har visat att det finns ett utrymme för Simotics Connectatt blomstra inom på den nuvarande marknaden.Motorns geometriska uppskattning, som är baserad i preliminär elektromagnetiskgeometribestämning, skapar en digital tvilling av motorn som är tillräckligt noggrann föratt aggera som ett verktyg vid t.ex. temperatursberäkningar men saknar noggrannhet förmer avancerade och känsliga beräkningar, t ex för användbarhet inom magnetisk flödestäthetsberäkningar.Temperaturmodellen som presenteras visar stor noggrannhet vid beräkning av statornstvärsnittstemperatur, men noggrannheten minskar för rotorns tvärsnittstemperatur.En statorströmmodell föreslås med ett proportionellt förhållande mellan magnetflödesdensitetenoch statorströmmen. Resultaten indikerar ett linjärt förhållande, men användandetav den digitala tvillingen för att beräkna proportionell konstant konstateras attinte vara tillräckligt noggrann metod.