This talk summarizes important experiences we cultivated in several projects where we developed AI methods for industrial customers like chemical production plants or gas fired power plants. One might think that apply...
详细信息
ISBN:
(纸本)9798400705915
This talk summarizes important experiences we cultivated in several projects where we developed AI methods for industrial customers like chemical production plants or gas fired power plants. One might think that applying leading edge AI methods to large quantities of industrial data will automatically yield valuable results. In our experience there are many "traditional" obstacles that need to be removed, before the magic can happen [2]. For many use cases, data from various sources need to be integrated. For example, time series data stored in industrial data Historians need to be labeled according to quality data stored in Laboratory Information Management systems (LIMS). The process data like setpoints and raw material characteristics need to be contextualized using the production data, e.g. production orders, raw material lots, from ERP systems. Although many companies are currently building up enterprise data lakes, the reality in many plants still is that data is locked in separate silos and a real-time data integration requires significant efforts. Building, automating and operating a large industrial plant requires major (traditional) engineering efforts and expertise. data scientists usually have different backgrounds and often lack knowledge, for example in control systemsengineering and process engineering. In addition, many important engineering artifacts like Piping and Instrumentation Diagrams (P&ID) and process design document are either not up to date or even unavailable at all. However, in order to define relevant use cases and design powerful AI solutions, quite a high level of both traditional engineering and data science know-how is required. This means that traditional plant experts need to closely cooperate with data science experts. In the long run, software tools like ABB Ability (TM) BatchInsight will be able to offer easy to use solutions for a broad class of use-cases, so that plant experts can use AI solutions without the need for additi
With the development of intelligent manufacturing, reconfigurable manufacturing has become one of the current application characteristics, but the actual landing of reconfigurable system in manufacturing workshop stil...
详细信息
In the current technological landscape, the transmis-sion of data through communication channels has experienced significant advancements. The rapid growth in communication technology has brought about improved outcom...
详细信息
This study investigates the usage of water in urban areas, with particular attention to location, age, water quality, and bathing habits. We examined the data using machine learning, more especially a RandomForestClas...
详细信息
Industries are embracing the implementation of digital twin to improve reliability and efficiency of their systems. Even though the concept of digital twin is simple, its implementation is challenging due to various i...
详细信息
ISBN:
(纸本)9798350374995;9798350374988
Industries are embracing the implementation of digital twin to improve reliability and efficiency of their systems. Even though the concept of digital twin is simple, its implementation is challenging due to various issues including the detailed framework which can vary depending on the type of industry. Power transformers are one of the greatest engineering inventions in the 19(th) century that helped achieve today's highly efficient electricity transport process. They are indispensable for the reliable operation of the power network, and managing these assets to their optimal lifetime without unexpected in-service failures is vital. While various modelling approaches have been utilized in design, operation and condition monitoring of power transformers, it is expected that implementation of digital twin will incorporate various physics/knowledge based, data driven models into a single platform where data are of high fidelity, live and interchanged among models and support the best utilization of transformers with high reliability. This paper provides a demonstration of implementing a digital twin platform for power transformers through MATLAB software.
The advent of artificial intelligence within the manufacturing industry has been a major swift and is the philosophy of machines to reason, conduct, and achieve either the same output as or comparable to mankind. Nume...
详细信息
Outliers are the data points that vary significantly from the primary distribution of data. In data mining, determining outliers is an essential task for establishing the data quality, decision-making, and models'...
详细信息
Monitoring and identification of defects during additive manufacturing is mostly done by bespoke optical or acoustic measurement systems. These in-situ monitoring technologies are either intrusive or sensitive to nois...
详细信息
ISBN:
(纸本)9783031382406;9783031382413
Monitoring and identification of defects during additive manufacturing is mostly done by bespoke optical or acoustic measurement systems. These in-situ monitoring technologies are either intrusive or sensitive to noisy manufacturing environments. We propose a movement tracking-based in-situ monitoring system for additive manufacturing, which is non-intrusive, less sensitive to environmental factors, and easier to operate and maintain. It evaluates the hypothesis that extruder nozzle temperature can be predicted from printer head movement, since temperature and acceleration are correlated due to the printers control unit. Subsequently, this provides an indication of print quality as the extruder temperature plays a vital role. We collected data from experiments using the MakerBot Replicator to examine the hypothesis. Results show that a Random Forest algorithm is more accurate in predicting the temperature variation using head acceleration and time lag temperature data as input parameters, and outperforms a k-Nearest Neighbors and a Vector Autoregression algorithm.
The 4.0 revolution is leading to increasingly automated, flexible, and intelligent manufacturingsystems that require greater complexity to manage during maintenance and process control. In this context the optimizati...
详细信息
ISBN:
(纸本)9783031381645;9783031381652
The 4.0 revolution is leading to increasingly automated, flexible, and intelligent manufacturingsystems that require greater complexity to manage during maintenance and process control. In this context the optimization of the human machine interaction plays a crucial important role in the design of modern industrial systems. Virtual Reality (VR) offers realistic simulation environments where users can be involved to replicate specific human tasks, detecting and solving problems before they occur. The paper proposes a human-centric digital design methodology that integrates VR technologies with human data analysis tools to support the design or redesign of complex industrial systems. Different wearable devices have been used to collect data about physical and mental user conditions to provide an early assessment of the operators' workload, while comparing different design solutions into the virtual space. An industrial use case related to the redesign of packaging automated machines was used to validate the proposed method and tools: a preliminary correlation between physiological parameters and machines interactions was found.
Due to the complex geometric shapes, high precision design dimensions, and specific material characteristics of automobile wheel hubs, there are stringent requirements for processing procedures, efficiency, and stabil...
详细信息
暂无评论