Flow and Fluid Contamination Diagnostics
FRAMER: Fast Reconstruction of Architectural Models from Existing Resources
FTL’s “FRAMER” (Fast Reconstruction of Architectural Models from Existing Resources) system has been developed for the Chemical and Biological Defense (CBD) program of the DoD. It combines specially trained neural networks with advanced 3D processing algorithms and intelligent data exports to enable high quality building model generation from both blueprints and photographs, and parameter exports for accurate transport and dispersion simulation.
FRAMER leverages FTL’s ongoing research utilizing neural networks for image and 3D data processing, object detection, and high-quality 3D model generation for rapid development. This software system provides the highest fidelity 3D building models based on the data used as input and the most accurate possible exports to AR and VR applications, in addition to T&D software such as NIST’s CONTAM.
Blueprints constitute a unique problem for neural networks due to their widely varying range of quality, frequent lack of relevant information, and ambiguous distinctions between room types. FTL overcomes these challenges with a unique synthetic data training step that leverages cutting-edge research with a large dataset of accurate and annotated building models proven to increase neural network accuracy for building part and object detection. This data, which will be extended to include accurate blueprint output, enables new and existing neural networks to be trained easily and repeatedly, increasing the robustness of detection for typical objects such as walls, doors, and windows. The result of this NN-processed blueprint is a data structure containing all the building’s relevant features, from which a 3D building model can be procedurally generated.
FRAMER’s generated 3D models will optionally be segmented and include objects labeled using state of the art neural network research, bringing a richer experience to the existing virtual and augmented reality applications in use at CBD. These key developments also include the use of an additional neural network to automatically augment a 3D indoor scene with new objects and furnishings that match their surroundings.
This exciting research will enable FRAMER to provide true-to-life indoor building areas even when photos or scans of those rooms do not exist. Through FTL’s collaboration with the developers of CONTAM at NIST, FRAMER’s exported building parameters will support high quality physically accurate transport and dispersion modeling. The exported building data will be usable directly in CONTAM through the creation of building templates and automatic editing of project files. Additionally, FTL will leverage its extensive experience with the development of AR and VR applications for FRAMER’s high fidelity 3D building model exports.
OilEye Fuel Monitor
FTL personnel have a long history of fluid monitoring both for DoD and industrial clients, with expertise in flow system design, fluid-dynamic analysis, optical access, and data acquisition. Working for the Navy, Army, and Air Force, FTL has shown in-flow detection and discrimination of sediment, water, and wear particles.
For the OilEye project, FTL designed isokinetic flow cells and custom AI classifier software capable of combining thousands of frames of imagery to extract particle size distribution and other parameters necessary for process monitoring.
FTL uses parallel processing to enable classification of thousands of particles against training sets from tens to hundreds of thousands of test cases in seconds. This allows discrimination of water and sediment particles, as well as identification of biological organisms in fluid flows.
SpOilEye
Flow and Fluid Contamination Diagnostics
The “SpOilEye” system extends FTL Labs Corporation fluid analysis capabilities and establishes the most efficient, cuvette based fluid screening approach, powered by advanced machine learning algorithms. The Fluid Monitor system enables rapid video imaging of contamination droplet and particulate matter in any non-opaque fluid such as fuels, oils, lubes and aqueous solutions, providing consistent and repeatable detection and classification.
The Fluid Monitor uses machine vision and AI learning algorithms with microscope video imagery to determine size, shape, classification, and number of droplets and particles. Automatically generated test reports show actual captured particle and droplet imagery. Unlike particle counters, the Fluid Monitor uses high resolution images for much greater diagnostic accuracy enabling differentiation between droplet and solid contaminants.
The benefits of the Fluid Monitor system include:
- Rapid fluid analysis and reporting of sediment and droplet contaminant levels for system failure prevention.
- Low cost: Comparable to mid-range particle counters.
- Analyzes fuel, oil, or other fluid directly using a standard glass cuvette.
- Provides discrete particle and droplet images for further system diagnostic purposes.
- Uses machine vision and AI learning algorithms to detect droplets and particles in microscope video imagery.
- The Fluid Monitor is the ideal analyzer for a wide range of contamination or process monitoring applications from large hydrant systems to small lab-based quality control testing.
Fluid / Fluid Contamination
Contamination of fuel by engine oils is a significant cost to the Air Force when defueling, refueling, and servicing aircraft. One notable difficulty faced by flight line crews is identification and evaluation of fuel condition during these contamination events. FTL’s “FFC” (fluid / fluid contamination) detection system evaluates fuel on-site using UV-Vis-FTIR spectrometry, coupled with a sophisticated machine learning algorithm to quantify extent of contamination.
FTL’s spectroscopic algorithms have proven to be a powerful tool for analyzing known contaminants, but aircraft fuels and fluids have variations of both fuel/fluid base stock and additive formulations that lead to instability of spectral fingerprints. FTL’s FFC overcomes this with advanced spectrum analysis and AI-based classification software that classifies each contaminant’s spectral fingerprint in terms of prior data, becoming more intelligent as it encounters more real-world variation. FFC is portable, automatic, and quantifies hydraulic oil contamination accurately to 100 ppm contaminant by volume.
Fiber Sensor
FTL’s Fiber Sensor is an optical, field-capable, free-water detection and quantification instrument to be used initially in commercial and military aviation refueling environments. The instrument is expected to be installed on hydrant servicers, hydrant carts and mobile refueling tankers downstream of the final filter-separator or “Monitor” vessels on these vehicles to provide final, real-time fuel-quality inspection before entering the aircraft.
Autogrape
FTL personnel have a long history helping to develop the world’s first optical scattering fuel monitor to be installed aboard Navy aircraft carriers. Built into an 18-inch-long instrumented pipeline spool piece with 2, 4, 6, or 8-inch pipe diameter, the AutoGrape integrates into the data backbone of CVN-68 (Nimitz) and CVN-78 class aircraft carriers and provides instantaneous fuel quality information to maintenance personnel throughout the ship. FTL engineers worked with NAVSEA personnel to develop and execute a First Article Test (FAT) plan. AutoGrape passed all Navy First Article Testing including Shock, Vibration, EMI, Temperature, Humidity, Salt Fog, and Immersion for shipboard use.
BubbleSpoof
FTL has extensive experience developing payload systems for Unmanned Underwater Vehicles (UUV’s) including extensive fluid modeling of the flow and mixed-phase environment surrounding a moving submarine. This has included unique calculations, models, and experimental testing of large bubble plumes in pipes, pools, and open ocean. In particular, FTL has investigated the existence and use of pressure effects from bubble plumes and moving vessels, detected by custom undersea pressure sensor arrays, and verified through computational fluid dynamics calculations.
WHAM Drinking and Wastewater Analysis Module
In addition to fuels and oils, FTL has experience developing sensor systems for classifying drinking water. WHAM (Weighted Heuristic Analyzer Monitor) is a lightweight, cross-platform expert-system for assessment of candidate water sources based on public regulations. Using no reagents or destructive methods, it executes sensor measurements in real-time in an automated flow-through system, and alalyzes those measurements in parallel to provide instantaneous sample assessment.
Modular and flexible in its design, WHAM can be configured to interface with a wide variety of sensors and apply different sets of regulations based on context and need. Sensor readings (input) and analysis results (output) are separately exported as CSV-style spreadsheets as well as stored together, along with configuration settings, in a portable SQLite database. WHAM has been used to characterize numerous water samples, including deionized water as well as local (Amherst, Massachusetts, USA) tap and pond water.
M1 Ingested Dust Sensor
In addition to liquids, FTL has experience developing sensors to quantify contaminants in fast gas flows. The AGT 1500 engine developed by Honeywell for the M1 Ground Combat Vehicle has been observed to suffer premature performance degradation due to ingestion of dust, sand, and FOD (Foreign Object Debris) during operation in difficult environments. FTL personnel worked with the Army at the depot-level to develop and test a novel inductive loop sensor that operated almost ideally in the extreme air intake environment of this unique turbine engine. The resulting sensor detected even small dust ingestion events at very high speed and proved to be highly resistant to fouling.
ElectroSeptic
FTL developed a forward-deployable wastewater solution for the Air Force. FTL’s “ElectroSeptic” technology was aimed at reducing both energy use and footprint for rapid-setup, air-drop water treatment systems.
The core of the ElectroSeptic technology was a novel microbial fuel cell that could break down organic carbon molecules in wastewater using microbial thin films grown on a semi-permeable aerated membrane. Waste particulates in the effluent were able to be digested by “electrogen” microbes with severely reduced O2 input need, thus saving significant air pumping related costs.
Challenges included optimizing aeration bubble size and delivery and minimization of particulate biomatter build-up. This project was a collaboration with municipal and academic wastewater treatment facilities and included optimization of accurate and automated water monitoring.