KC Engineering / Capabilities
An integrated technical capability.
Three pillars, one practice. End-to-end wind tunnel testing. Modern ML/AI woven into engineering systems. Data-driven and hybrid control on real hardware. Each is led by senior engineering.
Wind tunnel testing, delivered end-to-end.
We treat a test campaign as a single accountable program, not a procurement sequence. The firm owns design of experiments, model design and instrumentation, fabrication, tunnel execution, data acquisition, reduction, uncertainty quantification, and archival reporting.
Operational experience across subsonic, supersonic, and water tunnels. Closed-loop and cyber-physical test architectures when the question is dynamic rather than static.
A / Design
Test plan & DoE
Objective-driven design of experiments, screening sweeps, factor sensitivity, and budget-aware test matrices.
B / Model
Model & instrumentation
CAD, aerodynamic and structural sizing, fabrication path, sensor selection, and integrated DAQ layout.
C / Execute
Tunnel execution
Subsonic, transonic, supersonic, and water-tunnel runs. SPIV, PSP, DIC, schlieren, force/moment, pressure.
D / Report
Reduction & reporting
Validated reductions, uncertainty budgets, archival reports, journal-grade artifacts, deliverable packaging.
Anchored in
- 2022Experiments on Vortical Interactions on Generic Multi-Swept Wing Configurations
- 2023Nonlinear, rate-dependent evaluation of flight damping derivatives of the Orion Crew Module
- 2016Cyber-physical flexible wing for aeroelastic investigation of stall and classical flutter
- 2017Developing a reduced order model from structural kinematic measurements of a flexible finite span wing in stall flutter
Modern ML / AI, fluent in real engineering.
We bring neural networks, time-series models, and operator-theoretic methods into the instruments, control loops, and decision pipelines our clients already operate. The goal is a deployable, edge-ready capability, not a notebook demo.
Production examples: a PyTorch + Streamlit wave-elevation estimator using LSTM and Space-Time POD for an ocean-energy client, neural-network surrogates for unsteady aerodynamic measurement, and controller-structure discovery via linear-genetic programming.
A / Surrogates
Neural surrogates
Feedforward, LSTM, and GRU surrogates replacing expensive measurement or simulation chains in production.
B / Reduction
POD · DMD · SINDy
Space-time POD, dynamic mode decomposition, and sparse identification, built for control-readiness.
C / Tooling
Engineering apps
Streamlit, PyTorch, and hyperparameter-sweep pipelines delivered to in-house engineering teams.
D / Deploy
Real-time inference
Model export to LabVIEW RT and FPGA targets, embedded at the edge, never blocking on the cloud.
Anchored in
- 2026Wave Estimation Tool v1.1 - LSTM + Space-Time POD
- 2017Developing a reduced order model from structural kinematic measurements of a flexible finite span wing in stall flutter
- 2024Leading-edge vortex dynamics on finite-aspect-ratio swept wings exhibiting large-amplitude oscillations
- 2008Low-dimensional modelling of a transient cylinder wake using double proper orthogonal decomposition
Controllers that earn their place on the hardware.
Controllers that respect the physics and exploit the data: adaptive, model-reference, reduced-order, and hybrid physics+ML architectures, identified from real measurements and deployed onto real hardware at real time scales.
Demonstrated across flow control, aero-elastic stabilization, vortex-shedding regulation, and ocean wave-energy. Synthesized, validated, and signed off in-house.
A / Identify
System identification
Black-box, grey-box, and physics-informed identification from rich experimental data.
B / Reduce
ROM-based control
Controllers synthesized from low-dimensional models: fast, interpretable, and stable.
C / Hybridize
Physics + ML
Hybrid controllers using physics priors with ML residuals: robust and data-efficient.
D / Deploy
Real-time hardware
LabVIEW RT, FPGA, and embedded targets, from synthesis through hardware-in-the-loop.
Anchored in
- 2014Closed-loop flow control of a forebody at a high incidence angle
- 2018Feedback flow control: a heuristic approach
- 2017Robust LQR control for stall flutter suppression: a polytopic approach
- 2022Efficiency analysis of the cycloidal wave-energy converter under real-time dynamic control using a 3D radiation model
[ 04 ] Underlying disciplines
The technical bench under the practice.
Eight foundational disciplines. Every engagement draws from this bench.
- 001 Tunnel
Experimental Aerodynamics
Subsonic, supersonic, and water-tunnel campaigns: SPIV/PTV, PSP, DIC, force & moment, pressure rake, high-speed schlieren.
SPIV / PTVPSPDICSchlieren
- 002 Tunnel
Computational Fluid Dynamics
Industrial RANS, hybrid RANS/LES, and DES: CAD-to-mesh through validated unsteady physics extraction on DoD HPCMP resources.
RANS / DES / LESDoD HPCMPMesh design
- 003 ML
Machine Learning & Surrogates
PyTorch-based feedforward, LSTM, and GRU models for engineering inference. Streamlit tooling, hyper-sweeps, export pipelines.
PyTorchLSTM / GRUStreamlit
- 004 MLControl
Reduced-Order Modeling
POD, DMD, SINDy, and Koopman-operator methods. Control-ready, low-dimensional representations of dominant nonlinear physics.
POD / DMDSINDyKoopman
- 005 TunnelControl
Fluid-Structure Interaction
Dynamic stall hysteresis, leading-edge vortex stabilization, gust response, flutter mitigation for flexible aerostructures.
Aero-elasticityGust responseDynamic stall
- 006 Control
Closed-Loop Control
System ID, state estimation, and feedback synthesis on real-time hardware: PID, MRAC, adaptive, and linear-genetic above 1 kHz.
System IDMRACLabVIEW RT / FPGA
- 007 Control
Active Flow Control
Synthetic jets, fluidic oscillators, plasma, and high-rate blowing: separation, lift, drag, and shock manipulation.
Synthetic jetsOscillatorsPlasma
- 008 TunnelMLControl
Technical Advisory
Independent technical review for program offices, R&D leadership, and acquisition teams. De-risk and surface the right questions.
Program reviewRFP shapingR&D strategy
[ 05 ] Engagement modes
How we work with clients.
Three modes of engagement that map cleanly onto how real technical programs run.
-
A - Discovery
Scoped technical study
Fixed-scope investigation. Trade studies, feasibility analyses, design-of-experiments, ML proof-of-capability, or sensitivity sweeps.
Typical 6 - 16 weeks
-
B - Program
Cooperative R&D
Multi-year cooperative agreements as prime or sub. Wind-tunnel campaigns, hardware, controller deployment, archival record.
Typical 12 - 48 months
-
C - Advisory
Strategic advisory
On-retainer advisory for program offices and R&D leadership. Independent review, proposal evaluation, on-call subject-matter access.
Retainer, monthly
Bring us a hard problem.
If your problem is straightforward, you have other options. If it isn’t, this is the right firm.