Researchers and scientists from the South Ural State University (SUSU) in Russia, in cooperation with Oxford University, are developing pressure sensors with self-validating functionality.
The sensors are part of a new wave of "intelligent sensors", which use modern computational capabilities and more complex data processing algorithms to realise functionality that will enable Industrie 4.0.
The vision of Industrie 4.0 includes the use of systems that are able to self-configure, self-optimise, self-heal and self-protect. Sensors are required to be self-validating or metrological self-checking. This means sensors need to have the in-built capabilities to allow them to detect their most common faults, but also correct their measurement values to take into account the influence of the fault.
A further challenge for self-validating sensors is that the signal has to be processed entirely at the Point of Acquisition, due to communication and computational bandwidth constraints.
Targets for these sensors include the nuclear, aerospace and petrochemical industries. Self-validating sensors would have an additional effect of increasing the intervals between required calibration, which can have major economic implications for the running of facilities and plants.
The research is being led by Oleg Bushuev, in collaboration with Oxford University professor Dr Manus Patrick Henry, a leading specialist in the field of signal processing, computational equipment and self-diagnostics.
Henry authored the concept of intelligent sensors in systematic form during the early 90s. Self-validating sensors are able to check if they have lost their metrology functionality, and should be able to self-diagnose issues with themselves before any problem occurs.
The Russian researchers, in their work on pressure sensors, created a new construction technique which allows for the self-validating functionality.
Within the seemingly ordinary pressure transducer is the means to generate test signals with ultrasonic excitation, in order to evaluate the sensor's structural integrity. The characteristics of these signals are then assessed by a variety of algorithms, including a modified version of Prism signal processing technique proposed by Henry.
[Self-validating pressure sensors. Photo: SUSU]