AUTOMOTIVE ENGINE MISFIRE DETECTION SYSTEM INCLUDING BIT-SERIAL BASED RECURRENT NEUROPROCESSOR

• Main Authors: JAMES JOHN VICTOR, FELDKAMP LEE ALBERT, TAWEL RAOUL, MARKO KENNETH ANDREW, ARANKI NAZEEH, PUSKORIUS GINTARAS VINCENT
• Format: Patent
• Language: English
• Published: 25.08.2000
• Edition: 7
• Subjects: BLASTING; CALCULATING; COMBUSTION ENGINES; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; CONTROL OR REGULATING SYSTEMS IN GENERAL; CONTROLLING; CONTROLLING COMBUSTION ENGINES; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; HEATING; HOT-GAS OR COMBUSTION-PRODUCT ENGINE PLANTS; LIGHTING; MEASURING; MECHANICAL ENGINEERING; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS ORELEMENTS; PHYSICS; REGULATING; TESTING; TESTING STATIC OR DYNAMIC BALANCE OF MACHINES ORSTRUCTURES; TESTING STRUCTURES OR APPARATUS NOT OTHERWISE PROVIDED FOR; WEAPONS

PURPOSE: An automotive engine misfire detection system including a bit-serial based recurrent neuroprocessor is provided to improve a detecting engine misfires on board the vehicle in real time and in a cost effective manner by using a crankshaft based engine diagnostic system including a neuroprocessor for performing a required computations. CONSTITUTION: A crankshaft based engine diagnostic system includes a neuroprocessor for performing the required computations for detecting engine misfires on board the vehicle in real time and in a cost effective manner. The neuroprocessor achieves cost effectiveness and design compactness by employing a combination of bit-serial and bit-parallel techniques for the implementation of the neurons of the neuroprocessor. The number of neurons required to perform the task is significantly reduced by time multiplexing groups of neurons from a fixed pool of neurons to implement the successive hidden layers of a recurrent neural network topology.