Machine Learning Techniques for Extending BIT Coverage to Sensor Faults
In this paper, we describe the problem of developing sensor fault detection within airborne instrumentation systems, and solutions based upon machine-learning techniques. We conclude with a report on our proof-of-concept demonstrator and outline the next steps towards the implementation of an autonomous self-diagnostic sensor solution.
Good data is key to the success of an instrumentation program, and modern data acquisition systems allow for reliable, high-fidelity data capture. Unfortunately, instrumentation programs are often hindered by undetected sensor and wiring problems that can lead to invalid data and inconclusive analysis. Traditionally BIT (built-in test) coverage is limited to the electronic acquisition units and does not extend to sensors and associated wiring.
Many authors have identified sensors and wiring as the weakest link in an entire instrumentation system (e.g. , ), where the transducer and consistency of the transducer/structure interface can “make or break” a system. Choosing long-life sensors with lifetimes in excess of assets under test is one approach for addressing this problem; however, these high cost and high specification sensors are rarely economically viable. Traditionally for airborne measurement programs, there is a realistic expectation that sensors will be replaced over time, and that dedicated data analysts will be available to spot subtle signs within data, which indicates the onset of sensor/wiring faults. This approach does not scale well for large fleet deployments and does not allow for robust automation.
Figure 1: Confusion Matrix Results
Login and download the white paper to learn more.
- Machine Learning
- Sensor faults
- Built-in Test
Login to download this resource