Development Of Electronic Control System Of Motorcycle Engine And Study On Methods Of Misfire Detection

With more and more strict exhaust emission standard in recent years, the motorcycle which can not be up to the emission standard will not be launched. Electronic controlled fuel injection(EFI) technology is the best choice for the current motorcycle environment protection technology because of its excellent fuel economy and low emission. So, motorcycle EFI system has a huge market in the future.Just because the above-mentioned reasons, motorcycle EFI system is studied carefully. According to the current status of electronic controlled motorcycle in China, a motorcycle EFI system is developed in this thesis after studying a electronic controlled motorcycle. The system is composed of electronic control system, fuel supply system, ignition system and intake system. Using the signals of some sensors as inputs, such as CPS(Crankshaft Position Sensor), MAT(Manifold Air Temperature sensor), CTS(CoolantTemperature Sensor), TPS(Throttle Position Sensor) and OS(Oxygen Sensor), the electronic control system can confirm the operating state of the engine, then it controls fuel pulse width and ignition advance angle which can be got from the MAPs stored in the memory of ECU. In order to improve the efficiency of TWC(Three-Way Catalytic Converter), a closed-loop control arithmetic is applied in this thesis, and it is proved during tests that it can keep air-fuel ratio around stoichiometric values. The design of electronic control system is the core of a motorcycle EFI system and an important part in this thesis, and it includes hardware design, software design and anti-interference design. In this thesis the hardware circuits are composed of power supply circuit, input signal processing circuit, output signal driving circuit, input and output interface circuit, communicating circuit. At the precondition of meeting reliability and precision, the system should be designed as simply as possible, then the cost of the system will be cut down. In order to up to higher efficiency and expansibility, the structure which includes main program and module progrom is taked, then the different functions can be distributed reasonably. The software system is made up of main progrom, analog signals acquisition module, starting module, accelerating module, decelerate module, steady state module and overspeed cut off module. The anti-interference design is very important because it affects the stability of the system.In this thesis a calibrating system developed using Visual C++ 6.0 is designed to acquire the state data of engine and optimize the controlled parameters and MAPs. After plenty of bench tests, it is proved that the system developed in this thesis can keep the motorcycle engine operating steadily at all states and having good emission characteristic.When misfire occurs, the emission characteristic of engine and the stability and security of vehicle will worsen. So on the base of the motorcycle EFI system developed in this thesis, two methods of misfire detection are studied.The method of misfire detection using speed signal of the engine can detect misfire as soon as it happens. Along with engine speed higher and higher, the inertia of the flywheel becomes biger and biger, so it results in the characteristic of the misfire included in engine speed signal weaker and weaker. Just because of this, this method can not detection misfire accurately. Using the signal of wide range oxygen sensor as misfire detection signal, the characteristic of the misfire and the delay of the sensor signal vary slightly with engine speed increasing; the characteristic of the misfire becomes clear and the delay of the sensor signal decrease with engine load increasing, and it has the misfire detection easier. Because the factors affecting the signal of wide range oxygen sensor increase when engine operates at transient state, the effect of misfire detection using this method is not as well as that at other states.In recent years, wavelet analysis is more and more often applied to analyse and process signals. In this thesis the signals of engine speed sensor and wide range oxygen sensor are analysed and the noise in them is reduced using wavelet analysi. The processed engine speed signal displays the characteristic of the misfire more clearer, and it can be used to detecte misfire even though the speed of the engine is high. The same effect is obtained by processing the signal of wide range oxygen sensor using wavelet analysis.