A comparison of statistical learning approaches for engine torque estimation

• Published in: Control engineering practice Vol. 16; no. 1; pp. 43 - 55
• Main Authors: Rakotomamonjy, A., Le Riche, R., Gualandris, D., Harchaoui, Z.
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier Ltd 2008; Elsevier
• Subjects: Applied sciences; Artificial intelligence; Black box; Cognitive science; Computer science; Computer science; control theory; systems; Engine torque; Engines and turbines; Exact sciences and technology; Internal combustion engines: gazoline engine, diesel engines, etc; Learning and adaptive systems; Mechanical engineering. Machine design; Neural networks; Support vector machines; Support vector machines; Black box; Engine torque; Neural networks; Availability; Statistical analysis; Probabilistic approach; Input signal; Least squares problem; Motor torque; Neural network; Experimental study; Gear; Optimization; Mean square error; Statistical method; Filter; Non linear model; Automobile industry; Vector support machine; Internal combustion engine; Linear estimation; Non linear effect; Artificial intelligence
• ISSN: 0967-0661; 1873-6939
• DOI: 10.1016/j.conengprac.2007.03.009

Engine torque estimation has important applications in the automotive industry: for example, automatically setting gears, optimizing engine performance, reducing emissions and designing drivelines. A methodology is described for the on line calculation of torque values from the gear, the accelerator pedal position and the engine rotational speed. It is based on the availability of input-torque experimental signals that are pre-processed (resampled, filtered and segmented) and then learned by a statistical machine learning method. Four methods, spanning the main learning principles, are reviewed in a unified framework and compared using the torque estimation problem: linear least squares, linear and non-linear neural networks and support vector machines. It is found that a non-linear model structure is necessary for accurate torque estimation. The most efficient torque model built is a non-linear neural network that achieves about 2% test normalized mean square error in nominal conditions.