📢 #highlycited paper 📚 A WENO-Based Upwind Rotated Lattice Boltzmann Flux Solver with Lower Numerical Dissipation for Simulating Compressible Flows with Contact Discontinuities and Strong Shock Waves 🔗 mdpi.com/2076-3417/14/23/114… 👨‍🔬 by Yunhao Wang et al. 🏫 Nanjing University of Aeronautics and Astronautics #CFD #numericalmethods #simulation

📡 New Special Issue in Symmetry! "Symmetry in #NumericalMethods" 📅 Deadline: 31 October 2026 🔗 brnw.ch/21x3ag7 #Symmetry

A new Special Issue opens for submission! Title: Symmetry in #NumericalMethods Editor: Dušan Nikezić and Nikola Mirkov Details: brnw.ch/21x3ag7 #callforpapers #mdpisymmetry #computationalfluiddynamics #timeseriesforecasting @Univerzitet_BG @ComSciMath_Mdpi

📢 #highlycited paper 📚 #SonicCrystal #NoiseBarrier with Resonant Cavities for Train Brake Noise Mitigation 🔗 mdpi.com/2076-3417/14/7/2753 👨‍🔬 by David Ramírez-Solana et al. 🏫 Universitat Politècnica de València (UPV)/Politecnico di Bari/D’Annunzio University of Chieti—Pescara #localresonances #numericalmethods

University of Surrey | PhD in Numerical Methods & PDEs ⚙️📐 Passionate about mathematics, HPC, and scientific computing? Check out this fully funded PhD at University of Surrey 🇬🇧 📌 Position: PhD – Parallel Methods for Nonlinear Oscillatory PDEs 🏫 Department: Mathematics and Statistics 📍 Location: Surrey, United Kingdom 👨‍🏫 Supervisor: Bin Cheng ⏳ Deadline: May 4, 2026 📅 Start Date: October 1, 2026 🕒 Duration: 3.5 years 💡 Project Focus Develop advanced numerical methods for solving complex nonlinear PDEs using modern parallel computing architectures ⚡ 🔬 What You’ll Work On • 📐 Oscillatory PDE models in fluid & climate systems • ⚙️ Parallel-in-time algorithms (HPC) • 🧠 Near-resonance theory in nonlinear dynamics • 💻 Efficient CPU-based large-scale simulations • 📊 Performance & energy efficiency analysis 🌍 Research Impact Applications in: • 🌦️ Weather prediction • 🌊 Fluid dynamics • 🌍 Climate modelling 🎯 Ideal Candidate • Strong background in Mathematics / Applied Math • Interest in PDEs, numerical analysis, or HPC • Programming & computational skills (plus) • Passion for theoretical applied research 🎁 What You Get • 💰 Fully funded (≈ £21,805/year stipend) • 🎓 Covers tuition (UK & eligible international) • ✈️ Travel & research funding • 🌐 Access to cutting-edge HPC research 🌆 Why UK? Home to world-class research universities and strong academic networks 🇬🇧 🔗 More Info phdscanner.com/opportunities… 🚨 Deadline: May 4, 2026 #PhD #Mathematics #HPC #NumericalMethods #UK #ResearchJobs #FullyFunded #PDE #PhDOpportunity

How we chose our ODE solver for PK/PD simulation — a journey through 4 methods Building a 3-compartment PK model simulator (remimazolam/propofol), we tested every major ODE solver. Here's what happened: 1. Euler (1st order) Simple: y(n 1) = y(n) h*f(t,y) Fast but inaccurate — error ~0.1-1%. Not acceptable for clinical simulation. 2. Dormand-Prince RK45 (5th order, adaptive) Theoretically the best — adaptive step size, error control, 5th-order accuracy. Reality: PK systems are mildly stiff (ke0 timescale vs clearance). RK45 entered step-rejection cascades, step size collapsed below minimum, solver crashed. No stiffness detection = no rescue. 3. LSODA (auto stiff/non-stiff switching) The gold standard in pharma — auto-detects stiffness, switches Adams/BDF. Problem: Requires external JS library. When unavailable, fell back to RK4 silently. Unpredictable behavior in PWA offline mode. 4. RK4 (4th order, fixed step) "Boring" but bulletproof. - Error < 0.000001% at dt=0.1 min (validated) - No external dependencies - No stiffness issues at physiological ke0 ranges - 15-25ms for 60-min simulation - Predictable, debuggable, maintainable Winner: RK4. Sometimes the most sophisticated algorithm isn't the best engineering choice. RK4's O(dt^4) convergence is overkill for clinical precision — and that's exactly why it's reliable. All solver code is still in the repo if you want to compare: github.com/ysuzuki1978/remim… #NumericalMethods #Pharmacokinetics #ODE #RK4 #TIVA #TCI #SoftwareEngineering

Julia’s GPU-accelerated ODE solvers have shown 20–100× speedups over JAX and PyTorch—and it’s not magic. In this talk, Dr. Chris Rackauckas explains the architectural differences behind Julia’s approach to #GPU solvers and why they unlock performance gains across real scientific and engineering workloads. youtu.be/ZSFfv2cckx0?si=nhcT… #JuliaLang #GPUComputing #ScientificComputing #HPC #SciML #DifferentialEquations #MachineLearning #PerformanceEngineering #NumericalMethods #AIEngineering

When we move into Itô Calculus you have to give up the fantasy that you’ll evaluate stochastic integrals in closed form. Forget it! 😏 The notation is compact, but the objects are brutal. Even something that looks harmless like Xₜ = X₀ ∫₀ᵗ (sin(Xₛ³) s² e^{Xₛ}) ds ∫₀ᵗ (1 Xₛ²) dBₛ already couples the noise term to the entire random path Xₛ in a nonlinear way. And a 2D system like dXₜ = (Xₜ − Xₜ³ Yₜ²) dt (1 Xₜ²) dBₜ dYₜ = (−Yₜ³ sin Xₜ) dt (1 Xₜ² Yₜ²) dBₜ is basically simulation or nothing in practice. In stochastic calculus, solving the SDE usually means a good numerical scheme like Euler-Maruyama, not a neat closed-form formula. #ItôCalculus #StochasticCalculus #SDEs #EulerMaruyama #NumericalMethods #ProbabilityTheory

Did you know? Henry Briggs devised a clever log-approximation to build the first base-10 tables In the early 1600s, Henry Briggs (the “father” of common logarithms) used this practical approximation to compute log. **Why it’s brilliant** * Works well for large (m,n). * Briggs picked (m) and (n) as powers of two, so finding (a^{1/n}) and (10^{1/m}) reduced to repeated square-root extractions—pencil-and-paper friendly. * This method powered the accurate log tables that turbocharged astronomy, navigation, and engineering for centuries. From hand roots to slide rules to silicon—numerical ingenuity never goes out of style. #mathhistory #numericalmethods #innovation #logarithms #Briggs

What if ODE solvers weren’t just numerical black boxes? In this talk, Dr. Chris Rackauckas breaks convention with Julia by integrating symbolic-numeric techniques into stiff ODE/DAE solvers. Discover how tools like #ModelingToolkit and codegen transform solver performance—boosting speed and robustness in ways traditional methods can’t match. A must-watch for computational scientists and engineers. youtu.be/NjcrCY0iqGA?si=hvJp… #JuliaLang #ModelingToolkit #ScientificComputing #NumericalMethods #SymbolicComputing #ODESolvers #DAESolvers #HighPerformanceComputing #TechnicalComputing #DifferentialEquations

Computer Oriented Numerical Methods Complete Course 🚀 We’re excited to announce a brand new course on BCA PATHSHALA - @JOINWEBSYRO Academy 👉 Start Learning Now: academy.websyro.com/courses/… #BCAPATHSHALA #WebsyroAcademy #LearnToCode #NumericalMethods #Programming

Julia and ModelingToolkit.jl are a powerful combination for engineers. This video explores how to model and analyze both steady-state and dynamic systems. You'll see how to efficiently set up, solve, and interpret system responses, using a hydraulic actuator model as a demonstration. It's a great workflow for engineers looking to enhance their #modeling and #simulation capabilities. juliahub.com/videos/julia-fo… #JuliaLang #ModelingToolkit #SciML #Engineering #ModelingAndSimulation #DynamicSystems #SystemsEngineering #OpenSource #NumericalMethods

Major news for engineers and researchers! A recent talk on ModelingToolkit.jl highlighted new features that simplify workflows and boost performance. MTKParameters enables parameter #optimization, while improved initialization allows for a wider range of initial conditions. The package also now supports dynamic optimization and new polynomial solvers. It's a game-changer for #modeling and #simulation. juliahub.com/blog/what-s-new… #JuliaLang #ModelingToolkit #SciML #OpenSource #Tech #SymbolicAI #ScientificComputing #NumericalMethods #Engineering

📊Digging into market volatility: Here's a look at average daily and real ranges across different periods (9-day to 100-day). Interesting to see how these metrics evolve. #QuantAnalysis #Stats #DataScience #FinancialModeling #AlgorithmicTrading #MachineLearning #MarketData #FinancialEngineering #Econometrics #TimeSeries #QuantitativeFinance #BigData #PythonForFinance #RStats #NumericalMethods #TradingStrategy #RiskManagement #MarketAnalytics #PredictiveModeling #ComputationalFinance

Excited to dive deep into Core of Neural Nets now🚀. Spent the last week solidifying my understanding of PINNs for the 1D Heat Equation. #PINNs #PhysicsInformedNeuralNetworks #MachineLearning #DeepLearning #AI #NumericalMethods #PartialDifferentialEquations

Python Programming and Numerical Methods: A Guide for Engineers and Scientists Covers Python fundamentals powerful numerical methods. Explore it free 👉 pythonnumericalmethods.berke… #DataScience #Python #MachineLearning #NumericalMethods