Intelligent Gearbox for Endurance Advanced Rotorcraft

Project Information

iGear

Grant agreement ID: 738144

DOI

10.3030/738144

Project closed

EC signature date 24 February 2017

Start date 1 March 2017

End date 31 May 2021

Funded under

• SOCIETAL CHALLENGES - Smart, Green And Integrated Transport

Total cost

€ 471 855,69

EU contribution

€ 410 044,13

Coordinated by

CRANFIELD UNIVERSITY
United Kingdom

Periodic Reporting for period 4 - iGear (Intelligent Gearbox for Endurance Advanced Rotorcraft)

Reporting period: 2020-10-01 to 2021-05-31

Summary of the context and overall objectives of the project

The ultimate goal of the project iGear (Intelligent Gearbox for Endurance Advanced Rotorcraft) is the development of an on-the-fly health assurance system for gearboxes, in the framework of the compound rotorcraft demonstrator for the RACER Fast Rotorcraft IADP. Challenges include extreme environments (high temperatures and vibration), data acquisition, mining, fault detection, diagnostics and prognostics. The project uses a combination of commercial sensors in a small package, with computed condition indicators. Multiple condition indicators are combined to improve sensitivity and robustness.
It is important that future fast rotorcraft are reliable and sustainable. Users want faultless, trustworthy systems, with the maximum capability and capacity. The rotorcraft and their systems need a long, maintenance-free life. When maintenance is required, it needs to be early and precisely targeted.
The project’s objectives were to assessed technologies for the characterisation of machine health in gearboxes, using sensing and software. Early detection, and subsequent localisation, of gear health degradation, was the main target. The goal was a health monitoring system capable of triggering prompt maintenance or part replacement. It was concluded that very sensitive detection is possible, when the evidence is combined from multiple sensors and algorithms. Multiple sensors can be combined in a small package for installation outside the gearbox with adhesive. Software can give robust diagnosis of healthy conditions, with sensitive early warning of degradation.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

The project started with a state-of-the-art review, and a requirements specification coordinated to stakeholder needs to the specific requirements of a novel helicopter gearbox. The assessment of the available technology reviewed the commercial-off-the-shelf (COTS) products to meet the specification. This tailored the choice of sensors, data acquisition electronics, telemetry, and signal processing, including suitable algorithms and development environment. A report was written to assess available technology, and a plan was developed for less mature options.
Hardware and software were developed for the detection logic and algorithms. A detailed system architecture was created. An electronics system was designed to include power conditioning, sensors signal conditioning, analogue to digital conversion, processing and communications. Software was prepared for data acquisition, processing the information, and controlling the system. A data analysis tool set was prepared, in a portable pre-production format for proof of concept. The architecture used a COTS platform to allow rapid development, but with scope for bespoke algorithms. Simple and mature condition indicators were supplemented by advanced algorithms for detection and diagnosis. Mechanical parts were designed to package the sensors.
Sensors, data acquisition and telemetry were procured to meet the specification. The sensor housing was manufactured. Sensing was assembled on a laboratory gearbox rig test rig at Cranfield. Test rig components were procured from a specialist gear supplier, modified with an agreed set of seeded faults. The sensor pack was tested in the test rig, to meet the requirements. Data acquired was used to trial the acquisition system and algorithms, prior to the test campaign, and a suitable data plan and archive was created.
Gearbox testing and data analysis were completed in the test rig, in a campaign of 13 test series, in which the baseline and seeded fault gears were replaced between every run. The test gears included seeded cracks to the root and web of the gears, and tooth face spalls, each in three degrees of severity. Baseline tests on good gears were repeated between the seeded fault tests, to measure the sensitivity to dis- and re-assembly. Finally the data was analysed through the suite of signal processing, to test and validate the algorithms on the data. Some good separation was observed in the CIs.
Dissemination included communications and engagement, through publications and workshops in conferences, attendance at exhibitions, online articles and videos, social media postings, a leaflet, and Masters student projects and thesis reports.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

Two key areas were developed: multiple parameter sensing, in a small footprint, and data fusion of condition indicators for health monitoring.
- A pack of state-of-the-art sensors was built, with a small footprint, covering a wide range of sensed inputs; for sensitivity and robustness, an array of condition indicators (CIs) was collected and computed. Concerning the sensors, a triaxial accelerometer, acoustic emission sensor, and temperature sensor were combined on one footprint for each measurement point, in a specially prepared sensor stack housing. The small footprint is necessary for the installation on an aircraft gearbox, because gearbox housing are designed to be light and strong, with curved surfaces and webs, with minimal flat areas. To avoid damage, the sensor stack housing is fitted to the surface with an adhesive. A high bandwidth microphone, shaft phase and oil quality sensor were also added. For the CIs, a wide range of features was computed from the sensor data. Sensors sampled at a high bandwidth offered a range of CIs, either individually or used together, in the time and frequency domain, using bulk statistical indicators as well as those combined with engineering knowledge.
- Data fusion: there are several advantages in using condition indicators (CIs) together, or in modified form. Some CIs offer very early detection of signs of degradation events such as sub-surface cracking, while some are better for later confirmation, such as geometrical change. A combination may show progression of the condition or sequence of related conditions. Some CIs rise early in response to transient events, but then may die away. Their cumulative record, by integration, summation, or simple binary latching, ensures that early indication is not lost. Many CIs vary in response to transients. A statistical observation can be matched to the conditions. This is important for helicopters, where climb, hover, and level flight are separated by transient states. The measurement strategy rejects the transients. Overall, the data fusion approach adds confidence to diagnosis, especially in machines with variable speed and load conditions.
The potential socio-economic impact is in future exploitation in machine health and reliability assurance for the RACER Fast Rotorcraft. A supply chain partnership will be needed to develop through to TRL9. An architecture review will lead to systems integration. The wider societal implications of the exploitation include jobs for those involved in hardware and software, future support roles in the remote support for systems tuning and supervision, and the sustainable capability and capacity of rotorcraft systems in the service of public and commercial applications.

Tobias presents his MSc project about data analysis - a great job

iGear architecture

The new gear rig installation is ready to go

The team present at a TESConf workshop, 2018, in Cranfield