Validation study · Case study
BioSTEAM–DWSIM process parity
A sequence of controlled cross-engine reference cases for mass conservation, matched-efficiency pumping, and ideal water–ethanol flash behavior.
Claims on this page apply only to the stated system boundary, methods, source revision, and evidence status.
- Source revision
- process-parity-v1 · qualified 2026-09-15
- Last verified
- 2026-09-15
Decision question
When model assumptions are aligned and acceptance quantities are explicit, do BioSTEAM and DWSIM reproduce analytically known or independently checked process results for simple unit operations?
System boundary
Three deliberately narrow reference cases: a liquid-water mixer, a pure-water pump, and an ideal 50/50 mol water–ethanol TP flash. Each case has its own acceptance variables; passing one case is not treated as proof of general simulator equivalence.
Functional or reference basis
Mixer: 100 and 50 kg/h water feeds. Pump: 100 kg/h water from 101325 Pa to 1000000 Pa under matched efficiency. Flash: 50/50 mol water–ethanol at 363.15 K and 101325 Pa with explicitly ideal equilibrium assumptions.
Methodology
The mixer was checked against a closed-form 150 kg/h mass balance. For the pump, BioSTEAM selected its native pump efficiency and the same efficiency was supplied to DWSIM before comparing power. The ideal flash was checked against NIST Antoine vapor pressures and an independent binary Rachford–Rice calculation, with conservation evaluated from component molar flows.
Assumptions
Only the named properties and unit-operation conventions were aligned. Diagnostics that the engines formulate differently, such as pump outlet enthalpy treatment, were excluded from parity acceptance rather than forced into a false equivalence.
Results
Mixer outlet flow was 150.0 kg/h in both engines with zero cross-engine mass-flow difference. In the pump case, BioSTEAM power was 0.07113047 kW and DWSIM power was 0.07113165 kW at the matched efficiency, a relative difference of about 1.65e-5; both retained 100 kg/h and reached 1000000 Pa. In the ideal flash, the independent vapor-fraction reference was 0.7329143221, BioSTEAM returned 0.7219168857, and DWSIM returned 0.6882711219; the remaining difference was traced mainly to pure-component vapor-pressure correlations.
Interpretation
Simple cross-engine checks are useful only when the compared quantity and assumptions are explicit. The mixer and pump cases support conservation and power behavior under the stated conditions, while the flash case shows why apparently identical “ideal” setups can still differ because of property correlations.
Limitations
These cases do not validate arbitrary thermodynamic models, recycle convergence, equipment design correlations, scale-up, economics, or plant performance. The flash qualification preserves the original failed tolerance run and the subsequent preregistered acceptance revision.
Reproduction
Rebuild the three frozen cases using the qualified BioSTEAM/ThermoSTEAM and DWSIM 10.2.5 runtimes. Verify the mixer against the 150 kg/h analytical balance, align pump efficiency before comparing power, and recompute the flash reference from Antoine vapor pressures plus Rachford–Rice.
Review the underlying methods.
Use the documentation to inspect the tools and workflows behind this study.