Proof of concept for Hardware-in-the-Loop Based Knock Detection Calibration RWTH Aachen University

Format:

Book

Conference/Event

Author/Creator:

Meli, Matteo, author.

Contributor:

Dönitz, Christian

kansagara, Jaykumar

Liberda, Norman

Pischinger, Stefan

Conference Name:

SAE WCX Digital Summit (2021-04-13 : Live Online, Pennsylvania, United States)

Language:

English

Physical Description:

1 online resource cm

Place of Publication:

Warrendale, PA SAE International 2021

Summary:

Knock control is one of the most vital functions for safe andfuel-efficient operation of gasoline engines. However, all knockcontrol strategies rely on accurate knock detection to operate theengine close to the optimal set point. Knock detection is usuallycalibrated on the engine test bench, requiring the engine to runwith knocking combustion in a time-consuming multi-stage campaign.Model-based calibration significantly reduces calibration loops onthe test bench. However, this method requires a large effort inbuilding and validating the model, which is often limited by thelack of function documentation, available measurements or hardwarerepresentation. As the software models are often not available,function structures vary between manufacturers and sub modelfunctions are often documented as black boxes. Hence, using themodel-based approach is not always possible. This article presentsa black box calibration approach for knock detection with minimalsoftware documentation where the Engine Control Unit (ECU) isoperated on a Hardware-in-the-Loop (HiL) rig. For this specificapproach, a playback of the recorded engine knock traces is fedsequentially into the ECU. To achieve fast runtimes, an automatedHiL environment is implemented with the help of MATLAB, ETASINCA-MIP and IPG RealtimeMaker. This setup allows to evaluatevariations of knock window lengths, knock window start angles andfilter combinations. The paper describes the method to calibratethe ECU function, without detailed architectural knowledge of thesoftware, to accurately detect the knocking combustion cycles. Thevalidation of the calibration from the introduced HiL methodreveals an acceptable accuracy (<5% error) and is thereforesuitable for knock detection. This method drastically decreases therequired engine test-bench time and human effort in comparison toconventional approaches and is better suited than the model-basedapproach in some cases

Notes:

Vendor supplied data

Publisher Number:

2021-01-0424

Access Restriction:

Restricted for use by site license