Diese Seite auf Deutsch ansehen

Tariq Hussain Magsi

B.Sc.
Faculty of Process Engineering, Energy and Mechanical Systems

Institute of Building Services Engineering

Tariq Hussain Magsi
  • Phone: +49 221-8275-2103

Positions

  • Research Associate

Research fields

  • Smart Building and Digital Twins
  • Information modeling and information technology standardization of buildings using NLP
  • Chatbot application for building automation
  • Optimization of engineering processes for municipal building automation based on standardized system types and information models (OptGA4.0).
    Currently, engineering processes for building services engineering (BSE) in terms of planning, execution, commissioning, and operation are characterized by high manual effort, low performance, and frequent disruptions with significant information loss between process phases. This leads to significant quality deficits in the construction and use of building services engineering. During operation, deficits in the implementation of automation functions and in facility-related facility management (FM) become apparent, resulting in suboptimal operating processes and unnecessary energy consumption. The fundamental objective of the research project is to develop an information model that describes the information balance of building services engineering (BSE) in a uniform, lifecycle-consistent, and machine-readable manner in building services applications. Based on this model, BSE types are semantically defined, taking existing descriptions into account. The focus of this information technology standardization is on the use cases of process control, quality assurance of automation functions, and FM. Building services systems, which in practice are often described in a semantically inconsistent manner, are mapped to the developed information standard using AI applications, creating a homogeneous language space (syntax and semantics) as the basis for implementing rule-based automation. This forms the basis for the development of SelfX FM applications, e.g., asset and energy monitoring, which avoid manual effort for information engineering during initial setup and throughout the lifecycle. SelfX here refers to the ability of systems to independently explore their environment and realize automated, i.e., engineering-free, interactions with other systems. Project partners: Building Management of the City of Cologne, School Construction Hamburg, Kieback&Peter, VDI, building Smart, VDMA - Association for Automation and Management for Houses and Buildings, AMEV, eTASK, Faculty of Architecture and Civil Engineering, University of Wuppertal Funding body: Federal Ministry of Education and Research - FH-Kooperativ Duration: January 1, 2024 to December 31, 2027.
    Optimization of engineering processes for municipal building automation based on standardized system types and information models (OptGA4.0).
  • EcoTwin - Digital Twin of Urban Green Spaces
    Two global megatrends – increasing urbanization and climate change – are leading to a mutually reinforcing stress situation. Both the increase in heat waves and extreme droughts, as well as the dangers posed by heavy and persistent rainfall, cannot be averted without sustainable land management. Green spaces and unsealed soils not only have a cooling function and store water, they also fulfill an important role for biodiversity and human well-being. The extent to which the functions of urban green spaces (ecosystem services) can be provided depends largely on the site characteristics of the areas. In addition to a multitude of freely accessible environmental databases, cost-effective and powerful environmental sensors have been developed in recent years that collect site-specific and up-to-date data. Therefore, there is no shortage of data or technologies to collect relevant environmental data, but rather a lack of data processing, interpretation, and visualization to implement concrete climate adaptation measures on site. The aim of the project is therefore to link available environmental data and make it available to responsible decision-makers in the form of a "digital twin" of urban green spaces, so that climate and disaster risk reduction measures can be effectively implemented at specific locations. Project partners: Bonn-Rhein-Sieg University of Applied Sciences (H-BRS) – International Center for Sustainable Development (IZNE), RF-Frontend GmbH, GIQS. Funding: ERDF/JTF Program NRW 2021-2027. Duration: February 1, 2024, to January 31, 2026.
    EcoTwin - Digital Twin of Urban Green Spaces
  • Optimized GA through the design of modular and continuously learning AI models (modAI)
    AI applications are currently spreading in many areas of everyday life (chatbots, translators, autonomous driving, etc.). Across the lifecycle of residential and non-residential buildings, applications of machine learning (ML), natural language processing (NLP), and reinforcement learning (RL) promise great potential for optimizing building technology processes. ML, NLP, and RL methods are used to learn patterns or behavior of complex systems, which can be used to recognize, predict, or influence their behavior. Within the framework of the research project, this method will be transferred to building automation (BA) applications. The research project analyzes and evaluates fields of application of AI applications for socially, ecologically, and economically optimized building processes. In addition to supporting applications for planning processes (e.g., NLP-based automation for creating function lists, etc.), the potential of recurring applications based on learned building or system dynamics will be analyzed. A key research goal is interoperable modularization and reusability of the AI ​​applications assessed as potentially significant. To this end, base models (e.g., RL agents for optimizing building target values) must be designed, which are trained for the specific use case during the application's runtime. In addition, system architecture solutions are being developed that enable both the operation of the AI ​​applications and their further training. Such modular solutions are to be integrated into the current building automation system architecture (e.g., as standardized functional modules). In the research project, prototype AI applications are being developed in cooperation with an automation system manufacturer and two municipal building automation users. These demonstrators are intended to validate modular AI applications in terms of functionality and efficiency. Project partners: - Funding provider: Zukunft Bau Duration: 2024-2027
    Optimized GA through the design of modular and continuously learning AI models (modAI)

Up
M
M