The Digital Design Assistant module consists of many applications software such as Multi- Disciplinary Optimization (MDO), Topology Optimization using Deep Neural Network (ToDNN), Model Based System Engineering (MBSE), Lattice Boltzmann Solver (LBM) and Design of Experiments (DoE) based structural or aerodynamic optimization (DoEOp).
These tools are being developed with deep domain knowledge gathered over years of experience in design and development of aero engines combined with machine learning and artificial intelligence. This module enables the designers to come up with various design configurations and select the best configuration that meets the requirements. It will reduce design cycle time as well as cost significantly to make the time-to-market shorter which is especially very critical for new products in competitive markets.
Aero-engines are made safer by increasing the number of critical control parameters and on-board sensors. As part of engine health monitoring (EHM) development, the vital parameters of aero-engines such as rotor speeds - N1 and N2, Compressor Discharge Temperature, Compressor Discharge Pressure, vibration, oil pressure, oil temperature, Turbine Entry Temperature (TET), Exhaust Gas Temperature (EGT), and fuel flow will be monitored and recorded. A data acquisition system is being developed and will be validated using the data obtained from the test bed.
The acquired data will be processed through artificial intelligent (AI) techniques as shown in the figure below and then maintenance schedule will be recommended. This will avoid high costs of unwarranted scheduled maintenance and maximize the availability of aircraft.
Aerodynamic Performance Deck Generation:
A model-based data-driven completely digital Aerodynamic Performance Deck is being developed along with Control-system Hardware in Loop (HIL) simulation capability which can aid in fully digital evaluation as well as Engine Test-bed run capability. Performance Trend Reports can be visualized for Total Service Life of an individual Engine and compared with fleet trends to detect/predict anomalies using Artificial Intelligence.
Reduced Order Models and Embedded AI onboard Control System for Augmented Performance Monitoring through Companion FADEC