Virtual Sensing for Low Pressure EGR Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Turbocharged gasoline engines with low pressure exhaust gas recirculation (LPEGR) systems face challenges in accurate control and estimation due to the complexity of gas pressure dynamics, leading to potential engine hardware failure and reduced fuel economy, as conventional systems are not adaptable and require numerous expensive sensors.
Innovation Solution
A control system that includes a differential pressure (dP) valve and sensors to model and adapt gas pressures, using a controller to determine and adjust pressures at various points in the LPEGR system, reducing the need for multiple sensors and mitigating noise and vibration issues, while accurately tracking exhaust gas constituents for precise engine control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If accurate control of EGR and throttle valves is implemented using conventional sensor-based methods, then control precision is improved, but system cost and complexity increase due to requiring a large quantity of sensors
Solution Approach 1:
The patent replaces the conventional mechanical sensor-based pressure measurement system with a computational fluid dynamics (CFD) based virtual sensing system. The CFD model calculates pressures at multiple points in the LPEGR system without requiring physical pressure sensors at each location, thereby reducing hardware complexity while maintaining measurement precision through adaptive modeling and validation against limited sensor data.
Solution Approach 2:
The patent creates virtual copies of physical pressure measurements through CFD-based virtual sensors. Instead of installing physical sensors throughout the system, the model generates virtual pressure readings at critical points by solving the Navier-Stokes equations, effectively copying the measurement function without the physical hardware overhead.
2Measurement precision
If multiple sensors are installed to accurately measure gas pressures at various points in the LPEGR system, then measurement accuracy is improved, but system cost increases
Solution Approach 1:
The CFD-based pressure modeling system serves multiple functions simultaneously: it predicts pressures at all LPEGR system points, validates sensor readings, adapts to component variations, and provides diagnostic information. This single computational system replaces what would otherwise require multiple dedicated physical sensors, achieving universality and reducing overall system cost.
Solution Approach 2:
The patent introduces a CFD-based virtual sensing system as an intermediary between the physical LPEGR system and the control unit. This intermediary processes limited physical sensor data and generates comprehensive pressure information throughout the system, acting as a mediator that reduces the need for direct physical sensor installation at every measurement point.
3Device complexity
If the LPEGR system uses fixed pressure modeling without adaptation, then system simplicity is maintained, but reliability decreases due to inability to account for component variations and aging
Solution Approach 1:
The patent transforms the static, fixed pressure modeling approach into a dynamic adaptive system. The CFD model continuously updates its parameters and predictions based on real-time sensor feedback and operational conditions, allowing the system to adapt to component aging, manufacturing variations, and changing operating conditions, thereby improving reliability without requiring complete system redesign.
Solution Approach 2:
The patent implements feedback mechanisms where actual sensor measurements are compared against CFD model predictions, and the model parameters are adjusted accordingly. This closed-loop feedback ensures the virtual sensing system remains accurate over time and adapts to system changes, significantly improving reliability compared to fixed non-adaptive modeling approaches.
Data Source
AI summary
Systems and methods for a turbocharged engine comprising an exhaust gas recirculation (EGR) valve and an EGR valve differential pressure sensor disposed in a low pressure EGR (LPEGR) system of the engine and a differential pressure (dP) valve that is distinct from a throttle valve and a dP valve outlet pressure sensor disposed in an induction system of the engine utilize a controller configured to, based on the sensed pressures, determine (i) a modeled pressure at the EGR pickup, (ii) a modeled pressure at outlet of an EGR cooler, (iii) a modeled pressure at an outlet of an air filter and (iv) a modeled pressure at the dP valve outlet, and control the dP valve and the EGR valve based on the modeled EGR pickup pressure, the modeled EGR cooler outlet pressure, the modeled air filter outlet pressure, and the modeled dP valve outlet pressure.