Tianjin University

Can Wang

Mingfa Yao

Haifeng Liu

Hu Wang

Zunqing Zheng

Mingchao Liu

Xiaoteng Zhang

A real-time prediction method for an engine emission is provided, including: acquiring multiple known historical test data samples for engine emission, dividing the samples into a training set and a test set to train multiple neural network (NN), calculating mean square errors (MSEs) output by each of the NNs with the different numbers of hidden layer nodes to determine a topological structure of the NNs, optimizing initial weights and initial thresholds for each of the NNs with a mind evolutionary algorithm (MEA), and establishing a real-time engine emission prediction system with an Adaboost algorithm.

Real-time prediction method for engine emission

The present disclosure discloses an intake oxygen concentration control system suitable for an engine with lean NOtrapping technology. The system adopts an exhaust turbocharging device to provide a pressure difference for an oxygen-enriched membrane to generate oxygen-rich and oxygen-deficient gases, and controls the intake oxygen concentration of different cylinders by adjusting the opening of flow control valves to match lean and rich combustion cycles of a lean NOtrapping system. In a lean combustion cycle, all four cylinders are filled with an oxygen-rich gas, which can make the combustion more complete and improve the thermal efficiency and fuel economy. In a rich combustion cycle, one of the four cylinders is filled with an oxygen-deficient gas, and the other three cylinders are filled with air or an oxygen-rich gas with a low concentration, so that less fuel is required to create a reducing atmosphere to realize the release and reduction of NOin a lean NOtrapping device, thereby reducing the fuel consumption and ensuring the output power of the other three cylinders.

Intake oxygen concentration control system suitable for engine with lean NOx trapping technology

Growing community of inventors

Tianjin, China

Can Wang