Case Study: SDEC improves DPF soot model calibration with MathWorks

SDEC Automates Calibration of Diesel Particulate Filter Soot Load Models

SDEC, a Shanghai Diesel Engine Co. focused on diesel-engine emissions compliance, needed a better way to calibrate its diesel particulate filter (DPF) soot load model, which acts as a virtual sensor for estimating soot buildup and triggering regeneration. Working with vendor MathWorks, the company sought to reduce the manual effort, cost, and inconsistency of tuning dozens of lookup tables in the model.

Using Simulink, Model-Based Calibration Toolbox, MATLAB, and Parallel Computing Toolbox from MathWorks, SDEC built an automated calibration workflow that cleaned road-test data, optimized lookup tables, and handled large transient datasets more efficiently. The result was a 90% improvement in calibration accuracy, error reduced to below 0.1 g/L, consistent results even from junior engineers, and calibration time cut from weeks to hours.