BEGIN:VCALENDAR VERSION:2.0 PRODID:-//CERN//INDICO//EN BEGIN:VEVENT SUMMARY:Hands-On Practical Introduction to Quantum Computing DTSTART;VALUE=DATE-TIME:20241201T113000Z DTEND;VALUE=DATE-TIME:20241201T130000Z DTSTAMP;VALUE=DATE-TIME:20260814T183409Z UID:indico-contribution-658-2474@events.chpc.ac.za DESCRIPTION:Speakers: Francesco Petruccione (UKZN)\nThis practical introdu ction to quantum computing aims to offer a foundational understanding of k ey quantum computing concepts\, algorithms\, and practical applications. T he lectures will cover the basics of quantum computing including qubits\, entanglement\, and quantum gates\, as well as an introduction to quantum c ircuits. As an example\, we will explore in the tutorial the quantum dynam ics of a spin system on quantum computer.\n\n**Target Audience:** Students \, academics and industry representatives interested in a practical introd uction to quantum computing.\n\n**Prerequisites:** Basic Python knowledge\ n\nhttps://events.chpc.ac.za/event/139/contributions/2474/ LOCATION:Boardwalk Convention Centre BICC.G-D1 - D1 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2474/ END:VEVENT BEGIN:VEVENT SUMMARY:MODELLING CLIMATE CHANCE IMPACTS ON FUTURE OFFSHORE WIND ENERGY US ING GEOSPATIAL INTELLINECE DTSTART;VALUE=DATE-TIME:20241201T070000Z DTEND;VALUE=DATE-TIME:20241201T083000Z DTSTAMP;VALUE=DATE-TIME:20260814T183409Z UID:indico-contribution-658-2403@events.chpc.ac.za DESCRIPTION:Speakers: Kgabo Humphrey T hamaga ()\nOffshore wind energy is the most commercially and technologically developed marine renewable energ y. Conversely\, the potential impacts of climate change and variability on future wind energy remains poorly understood\, including shifts and varia tions in the general wind pattern. Therefore\, this project seeks to model the impacts of climate change scenarios on the potential offshore wind en ergy using geospatial intelligence (Satellite data and deep learning). The set objectives of the project include: (i) to understand historical clima te change and wind trend for the past thirty (30) years using meteorologic al\, satellite datasets and deep learning approaches\, (ii) build predicti ve model for future offshore wind energy potential under different climate change scenarios. This project will use comprehensive geospatial data to develop and predict the spatial and temporal variability of climate impact s (i.e.\, wind speed\, wave height\, sea level rise\, and ocean currents) on future offshore wind energy production under different climate change s cenarios. Wind speed and wind density retrieved using SAR data were used t o predict future offshore wind energy. The results will contribute to a be tter understanding of how climate change may affect offshore wind producti on and ensure long term sustainability. By integrating geospatial intellig ence and advanced modeling techniques\, this project will provide critica l insights into the resilience and adaptability of offshore wind energy sy stems to the challenges posed by climate change. \nKey words: Climate chan ge\, Offshore wind energy\, Geospatial intelligence\n\nhttps://events.chpc .ac.za/event/139/contributions/2403/ LOCATION:Boardwalk Convention Centre BICC.G-W - Wood Rooms URL:https://events.chpc.ac.za/event/139/contributions/2403/ END:VEVENT BEGIN:VEVENT SUMMARY:Exploiting exascale HPC for materials modelling\; UK's Excalibur P roject DTSTART;VALUE=DATE-TIME:20241201T113000Z DTEND;VALUE=DATE-TIME:20241201T130000Z DTSTAMP;VALUE=DATE-TIME:20260814T183409Z UID:indico-contribution-658-2407@events.chpc.ac.za DESCRIPTION:Speakers: Scott Woodley (UCL)\, Thomas Keal (STFC)\, Marcello Puligheddu (STFC)\, Matt Smith (University of York)\, Alexey Sokol (UCL)\, Rajany KV (STFC)\nGULP is one of the leading materials software for model ling materials using the method of interatomic potentials\, likewise CASTE P and CP2K are two of the leading electronic structure codes for modelling materials. In this workshop the theory and a practical guide to how these can be employed to exploit HPC will be taught by developers of these code s. \n\nThe ChemShell computational chemistry environment will be introduce d with a focus on multiscale quantum mechanical/molecular mechanical (QM/M M) modelling of materials systems. The course will cover how to set up QM/ MM models for a range of materials chemistry problems and running calculat ions through interfaces to QM and MM codes including CASTEP and GULP. Rece nt developments in ChemShell for materials modelling carried out under the PAX-HPC project will be explored.\n\nhttps://events.chpc.ac.za/event/139/ contributions/2407/ LOCATION:Boardwalk Convention Centre BICC.G-D2 - D2 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2407/ END:VEVENT BEGIN:VEVENT SUMMARY:Different Techniques of Force-field Derivation and Setting Up Mole cular Dynamics (MD) Calculations at CHPC Using DL_POLY Code DTSTART;VALUE=DATE-TIME:20241201T070000Z DTEND;VALUE=DATE-TIME:20241201T083000Z DTSTAMP;VALUE=DATE-TIME:20260814T183409Z UID:indico-contribution-658-2408@events.chpc.ac.za DESCRIPTION:Speakers: Cliffton Masedi (University of Limpopo)\nMolecular d ynamics (MD) is a computer simulation method for studying the physical mov ements of atoms and molecules. The MD method can assist one in obtaining t he static quantities and dynamic quantities. This method gives a route to dynamical properties of the system: transport coefficients\, time-dependen t responses to perturbations\, rheological properties and spectra. The ato ms and molecules are allowed to interact for a fixed period of time\, givi ng a view of the dynamic evolution of the system. The DL_POLY Code paralle l molecular dynamics simulation package will be utilised for exploration o f such properties of molecular systems.\n\nhttps://events.chpc.ac.za/event /139/contributions/2408/ LOCATION:Boardwalk Convention Centre BICC.G-B2 - B2 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2408/ END:VEVENT BEGIN:VEVENT SUMMARY:A deep learning-based wind energy turbines suitability location an alysis and transferability DTSTART;VALUE=DATE-TIME:20241201T160000Z DTEND;VALUE=DATE-TIME:20241201T173000Z DTSTAMP;VALUE=DATE-TIME:20260814T183409Z UID:indico-contribution-658-2406@events.chpc.ac.za DESCRIPTION:Speakers: Phila Sibandze ()\nDespite significant advancements in the energy industry over the past decade\, most global regions still f ace challenges in ensuring the security and supply of fossil fuels. The ch allenge in wind power usage is identifying the optimal location for turbin e installation to maximize energy generation while minimizing environmenta l and socioeconomic impacts. This study aims to explore a data-driven deep learning-based modeling framework that predicts land suitability for larg e-scale wind energy development by inventorying current wind farms and usi ng spatial decision criteria. The proposed framework will use recurrent ne ural and convolutional neural networks to simulate intricate interactions between meteorological\, environmental\, and infrastructure-related spatia l variables influencing wind energy potential. The model will use various spatial datasets\, including wind speed data\, topography\, and environmen tal constraints\, to assess its transferability to different wind regimes\ , environmental conditions\, and infrastructure challenges across various geographic regions\, Furthermore\, the offshore and inland regions will be utilized to identify wind potential locations using LiDAR and SAR data fr om Sentinel-1 satellites for suitable evaluation and detection. The result s of this study will be used to translate renewable energy sources and red uce climate change by improving wind energy potential evaluation accuracy. \n\nKeywords: Deep learning\, Inland\, Transferability\, Turbines suitabil ity\, wind energy\n\nhttps://events.chpc.ac.za/event/139/contributions/240 6/ LOCATION:Boardwalk Convention Centre BICC.G-W - Wood Rooms URL:https://events.chpc.ac.za/event/139/contributions/2406/ END:VEVENT BEGIN:VEVENT SUMMARY:A near real-time spatial monitoring and forecasting of wind energ y using geospatial big data DTSTART;VALUE=DATE-TIME:20241201T113000Z DTEND;VALUE=DATE-TIME:20241201T130000Z DTSTAMP;VALUE=DATE-TIME:20260814T183409Z UID:indico-contribution-658-2405@events.chpc.ac.za DESCRIPTION:Speakers: Phila Sibandze\, ()\, Kgabo Humphrey Thamaga ()\nTh is workshop will focus on the integration of geospatial datasets\, and dee p\nlearning algorithms for real-time monitoring and forecasting of offshor e wind\nenergy. The session will cover the framework for data retrieval\, pre-processing\,\nintegration of remotely sensed datasets and the developm ent of predictive models\nto optimize wind turbine performance. After deve loping the predictive models\, we\nwill integrate climate scenarios to for ecast the state of wind in real-time\nmonitoring and near-future predictio n. Participants will learn how to integrate\ncutting-edge geospatial\, and meteorological datasets with deep learning\nalgorithms to predict energy production. The workshop will provide valuable\ninsights for renewable ene rgy-related professionals and stakeholders on how\ngeospatia\n\nThe global demand for renewable energy\, particularly wind energy\, is escalating du e to the urgent need to combat climate change and decrease reliance on fos sil fuels. The study aims to develop a system using geospatial data and de ep learning techniques for monitoring and forecasting wind energy. The stu dy seeks to answer three key questions: (i) to develop a framework for int egrating and processing geospatial big data for wind energy monitoring. (i i) implement a near-real-time data acquisition pipeline for continuous mon itoring\, and (iii) develop a predictive model using deep learning algorit hms and statistical methods. The use of geospatial and meteorological data sets\, turbine performance data (wind speed and direction\, theoretical po wer and active power) and Recurring Neural Network -Long Short-Term Memory will enable near-real-time monitoring and prediction of wind energy. The model performance will be evaluated using statistical indicators like stab ility tests and forecast accuracy metrics like MAE and RMSE\, to measure i ts stability under different conditions. The proposed model will be used t o provide accurate wind patterns and energy potential insights\, thereby o ptimizing wind turbine performance and energy production through the integ ration of various datasets. The study’s results will enhance wind energy predictions\, aid in better grid planning\, decrease fossil fuel reliance \, and enhance grid stability.\n\n**Keywords:** Deep Learning\, geospatial Big data\, remote Sensing\, wind energy\n\nhttps://events.chpc.ac.za/even t/139/contributions/2405/ LOCATION:Boardwalk Convention Centre BICC.G-W - Wood Rooms URL:https://events.chpc.ac.za/event/139/contributions/2405/ END:VEVENT BEGIN:VEVENT SUMMARY:Windows Subsystem for Linux with GPU support for molecular dynamic s simulations DTSTART;VALUE=DATE-TIME:20241201T113000Z DTEND;VALUE=DATE-TIME:20241201T130000Z DTSTAMP;VALUE=DATE-TIME:20260814T183409Z UID:indico-contribution-658-2402@events.chpc.ac.za DESCRIPTION:Speakers: Krishna Govender (University of Johannesburg)\nWindo ws Subsystem for Linux (WSL) is a feature of the Windows operating system that enables you to run a Linux file system\, along with Linux command-lin e tools and graphical user interface (GUI) applications\, directly on Wind ows. Unlike conventional virtual machines such as those run with Oracle Vi rtualbox or VMWare\, WSL requires fewer resources (CPU\, memory and storag e) and can access all the hardware components of your machine (including t he graphical processing unit (GPU)) that Windows has access to. \n\nThere are several Centre for High Performance Computing (CHPC) users that still make use of the compute resources for both testing and production\, especi ally when it comes to molecular dynamics codes such as AMBER. This is not an ideal situation for testing purposes as it would mean that individuals need to queue\, sometimes for long periods of time\, before discovering so mething might be wrong with their setup. Recently\, AMBER has become opens ource for non-commercial use\, which means that users no longer need to do their testing on the CHPC as they can test on their local laptop/desktop prior to submitting simulations to the queuing system at the CHPC. \n\nIn this workshop we will:\n• Setup WSL on a Windows machine.\n• Install a version of Ubuntu using WSL. \n• Install essential libraries needed to compile AMBER in Ubuntu.\n• Install CUDA Toolkit in Ubuntu for GPU suppo rt. \n• Compile the serial\, parallel and GPU versions of AMBER.\n• Ex port the above WSL instance so that it can be deployed onto other laptops/ desktops. \n\nThis workshop will be ideal for researchers\, scientists and students that make use of the CHPC resources for their computational chem istry research. \n\n**Prerequisite:** \nLaptop with Windows 10 or 11 (If y ou have Linux\, you will still be able to learn how to compile the softwar e package/s).\nPreferably 8GB RAM or more (not a must\, but things can be slow with less RAM).\nNvidia GPU (not a must as you can still get the CPU version of the code compiled and running).\n\n\n**Note:** We can consider other applications should there be time as it is possible to install opens ource electronic structure codes within WSL.\n\nhttps://events.chpc.ac.za/ event/139/contributions/2402/ LOCATION:Boardwalk Convention Centre BICC.G-B2 - B2 Tsitsikamma URL:https://events.chpc.ac.za/event/139/contributions/2402/ END:VEVENT END:VCALENDAR