The virtual sensor technology developed by Vir2sense provides the ability to reliably monitor difficult to measure quantities for emission monitoring and control, sensor enhancement and replacement, onboard identification of fuel characteristics and engine condition monitoring. The technology has been used for the development of specialist fuel engine development and simulation model plugins.

Emission Monitoring and Control

The Vir2sense NOx PVS is a physically-assisted virtual NOx sensor. The PVS continuously predicts the emissions based on inputs from robust sensors typically available on the ECU (engine control unit) using efficient combustion and emission models. In order to ensure the accuracy of the measurement over the engine lifetime, a direct emission measurement system is used only intermittently, which ensures its durability, especially in the harsh conditions of the engine exhaust. Compared to traditional emission measurement systems, the PVS ensures a significantly longer operating time between required repair intervals, while limiting the operating cost and complexity to the bare minimum. The Vir2sense PVS-based system is being developed to be used in within “intelligent” NOx SCR systems, with the combination allowing a 30% reduction in SCR catalyst size with equal or improved performance.

Sensor Enhancement & Replacement

Vir2sense builds model-based virtual sensors for physical sensor diagnostics, backup or replacement. An example is the Vir2sense VPS, a virtual cylinder pressure sensor which uses real-time ECU (engine control unit) data to calculate the expected pressure trace. The VPS can be used to identify sensor faults through plausibility checks, and act as a sensor backup to allow continuous engine operation even during physical sensor failure.

Onboard Fuel Characterization

Vir2sense is developing technology for onboard fuel characterisation and fuel-specific engine control, based on readily available onboard engine data. By determining a “fuel fingerprint” pertaining to relevant combustion and emission characteristics, the engine can be controlled to an optimum operating point independent of fuel used, allowing the introduction of alternative fuels in electronically controlled engines.

Engine Condition Monitoring

Vir2sense is using physical sensor data in combination with virtual sensors and machine learning for anomaly identification during engine operation. The system is aiming at identifying abnormal operation early enough to enable repairs prior to major damages or emergency shutdowns, increasing engine availability and reducing maintenance costs.

Engine Simulation Plug-ins

Vir2sense has developed high-performance combustion and emission models to be used as simulation plug-ins. Particular focus has been placed on modelling of combustion of different fuels, with the resulting models being used in the design and optimization of engines running on alternative fuels.