Du Danfeng

Application of PNN to Fault Diagnosis of IC Engine

In order to simplify data stream of automobile diagnosing instruments, A fault diagnostic method for internal combustion(IC) engine based on probability neural network (PNN) was presented. At first a PNN model was established, and then based on the sample of Jetta ATK engine, the model was trained and simulated by a number of sample sets of symptoms and troubles. Simultaneously, the comparison has been done between PNN and backpropagation(BP) network.The simulation experimental results demonstrated that PNN model is more feasible and successful than BP network model and could make data stream of diagnosing instruments easier.

Research on a Forestation Hole Digging Robot

In order to save time of digging hole when trees are planted, a robot that has six degrees of freedom is developed. In this paper the structure and function of a forestation hole digging robot is introduced and analyzed. At the same time, by the experiment it can be proved that the efficiency of the robot is very high and the ground surface destroyed by the robot is very small. The maximum diameter of the digging hole is 0.55m, a seedling tree can be planted in the hole. Furthermore, it is also very useful to eradicate stumps in forest area.

Theoretical and Experimental Research on the Characteristics of Micro-Length Fiber Carbonized Wood DPF

The Diesel powered vehicles produce a considerable amount of particulate laden exhaust gases, which is thought to be the main source of air pollution and also cause lung cancer. The diesel particulate filter (DPF) is one of the leading technologies for reducing diesel particulate matter (PM). In this paper, the method of utilizing micro-length fiber carbonized wood (MLFCW) as filter material of DPF is presented for the first time. Theory and test prove that the filter efficiency of the MLFCW DPF is high, the exhaust back pressure low and service life long, which can successively open up a new direction for reducing diesel PM.

Research on the Exhaust Pressure Drop Gradient of Micro-Length Fiber Carbonized Wood DPF

The large amount of particulate matter (PM) and nitrogen oxides (NOx) in diesel exhaust have brought critical environmental and health issues, resulting in the implementation of limits for diesel exhaust PM and NOx emissions in some countries. The diesel particulate filter (DPF) is one of the leading technologies for reducing diesel PM. In this paper, the method of utilizing micro-length fiber carbonized wood (MLFCW) as filter material of DPF is presented and a corresponding mathematic model is built for the first time. Theory and test prove that the filter efficiency of the MLFCW DPF is high, the exhaust pressure drop gradient low and service life long, which can successively open up a new direction for reducing diesel PM.

A Fault Diagnostic Method for EFI Engine Based on MATLAB Software Package

At present, the diagnostic instruments used widely here and abroad is not entire, which can not diagnose the mechanical fault without fault code. In order to solve this problem, this paper presents a method for fault diagnosis of electronic fuel injection (EFI) engine using radial basis function (RBF) neural network. By connecting MATLAB software package and ACCESS database, a fault diagnosis program is set up and a fault without code can be found. Meanwhile, the comparison has been done between RBF network and back propagation (BP) network. The simulation experimental results show that the RBF model is more feasible and successful than BP model and makes fault diagnosis easier.