DWSIM brings modern AI directly into the simulation workflow — not as a gimmick, but as a genuine productivity multiplier for engineers. Ask questions, generate flowsheets, diagnose convergence issues, and accelerate your work with the built-in AI Assistant.

An LLM-powered assistant integrated right inside the simulator. Ask questions in plain English about your flowsheet, the thermodynamics you should be using, how to set up a unit operation, or why a calculation isn't converging — and get informed, context-aware answers without leaving DWSIM.

• Natural-language help for every unit operation and property package
• Flowsheet interpretation — explain what a simulation is doing and highlight issues
• Script generation — produce Python and VB.NET snippets for automation tasks
• Connect to your preferred LLM provider, including local models via an LLM server

One-click AI analysis of your entire flowsheet — or of any individual unit operation. DWSIM reads the current solution, equipment specifications, and stream data, then returns a plain-English report you can act on:

• Flowsheet-level insights — Overall process summary, mass & energy balance sanity checks, bottleneck detection, efficiency observations, and suggested improvements
• Unit operation insights — Per-equipment diagnostics (distillation column profile analysis, heat exchanger pinch evaluation, reactor conversion review, compressor performance, etc.) with targeted recommendations
• Design & troubleshooting hints — Highlight specifications that are pushing equipment limits, flag unusual conditions, and propose next steps
• Report-ready output — Generate executive summaries and engineering notes you can paste straight into reports, reviews, or class assignments

Go from idea to flowsheet in minutes. Describe the process you want — "a three-column separation train for a crude benzene-toluene-xylene mixture" — and let the AI Design Mode propose, place, and connect the unit operations for you. Review, tweak, and simulate.

The AI-Assisted Convergence Enhancer (ACE) is a machine-learning extension that improves the robustness and speed of thermodynamic calculations throughout the DWSIM flowsheet solver. The underlying technology is a set of fully connected feedforward artificial neural networks (ANNs) trained on convergence data collected during ordinary simulations. The trained networks are used to provide either improved initial estimates or complete fallback solutions for flash calculations and reactor equilibria that are difficult to converge by conventional iterative methods.

Built-in tools for ADM1 anaerobic digester parameter regression and other data-fitting workflows, combining classical optimization with modern ML-backed heuristics for faster, more reliable convergence on complex nonlinear models.

DWSIM integrates Life Cycle Assessment (LCA) and Techno-Economic Analysis (TEA) directly into the simulation environment — and AI Insights turns those numbers into actionable narratives.

• LCA insights — Automatic interpretation of environmental impact categories (GWP, acidification, eutrophication, water use, etc.), identification of the biggest contributors in your process, and suggestions for greener alternatives
• TEA insights — Plain-English reading of CAPEX, OPEX, NPV, IRR, and payback metrics, with sensitivity commentary highlighting which process decisions most affect profitability
• Scenario comparison — Ask the AI to compare two flowsheet variants side-by-side on both economic and environmental fronts
• Executive-ready reports — Generate polished summaries suitable for management presentations, technical papers, or grant applications

• Crude oil distillation (atmospheric and vacuum towers)
• Natural gas processing, sweetening (amine units), and dehydration (glycol units)
• LNG liquefaction and regasification
• Fluid catalytic cracking feed conditioning, hydrotreating, reforming support calculations
• Pipeline hydraulics, two-phase and multiphase flow
• Gas compression and expansion trains
• Petroleum assay characterization and blending

• Ethylene, propylene, and olefins separation trains
• Methanol, ammonia, and urea synthesis loops
• Aromatic extraction and BTX separation
• Chlor-alkali and electrolyte processes
• Polymerization reactor feed conditioning
• Acid gas treatment and sulfur recovery

• Steam cycles (Rankine, reheat, regenerative)
• Gas turbine and combined-cycle performance
• Heat pumps, refrigeration cycles, and ORC systems
• CO2 capture (amine, physical solvent, membrane)
• Hydrogen production and purification
• Geothermal and waste-heat recovery

• Batch reactor design and scale-up
• Solvent recovery and recycling
• Crystallization and solids separation
• Extraction and purification trains

• Classroom flowsheets for unit operations courses
• Thermodynamics and phase-equilibria teaching
• Process control and dynamics labs
• Research into novel process intensification, green chemistry, and electrification

• Bioethanol and biodiesel production
• Anaerobic digestion and biogas upgrading (ADM1-based)
• Fermentation and bioreactor design
• Downstream processing: chromatography, ultrafiltration, crystallization
• Biorefinery flowsheets (lignocellulosic, algal, waste-to-energy)

• Desalination (MSF, MED, RO via crossflow ultrafiltration models)
• Wastewater anaerobic treatment (ADM1)
• Evaporation and crystallization for brine management
• VOC absorption and stripping

• Distillation and rectification for spirits
• Evaporation, drying, and concentration
• Flavor and essential oil recovery

• Python — Built-in IronPython console plus Python script unit operations
• Excel & Office — Native Excel add-in, property interface, and export to Office formats
• CAPE-OPEN 1.0 / 1.1 — Use DWSIM models in third-party simulators, or plug third-party CAPE-OPEN unit operations and property packages into DWSIM
• Python, .NET & COM Automation API — Drive simulations programmatically from Python, C#, VB.NET, or any COM-capable language
• TCP/IP Server — Run DWSIM headlessly as a simulation backend for web apps, digital twins, or optimization loops