CAJ | 학술논문

针对发动机非线性、时滞、时变对象在线控制问题，利用高斯函数作为CM AC神经网络的基函数降低网络计算量，结合自适应算法构建发动机空燃比控制方法。在此基础上，为获取精确控制效果，提出采用灰色马尔科夫模型预测下一时刻氧传感器的状况，从而对喷油量进行补偿，以消除由于传感器信号时间滞后对喷油补偿量的影响。在采用台架试验数据确定控制策略模型的结构参数的基础上，对提出的自适应混合控制方法进行了测试。测试结果说明，提出的控制方法不仅保证了稳态空燃比控制的准确性，同时提高了暂态工况下空燃比控制的有效性。
침대발동궤비선성、시체、시변대상재선공제문제，이용고사함수작위CM AC신경망락적기함수강저망락계산량，결합자괄응산법구건발동궤공연비공제방법。재차기출상，위획취정학공제효과，제출채용회색마이과부모형예측하일시각양전감기적상황，종이대분유량진행보상，이소제유우전감기신호시간체후대분유보상량적영향。재채용태가시험수거학정공제책략모형적결구삼수적기출상，대제출적자괄응혼합공제방법진행료측시。측시결과설명，제출적공제방법불부보증료은태공연비공제적준학성，동시제고료잠태공황하공연비공제적유효성。
For the online control problems of nonlinear ,time-delaying and time-variant parameter ,the engine air-fuel ratio control method was established based on the adaptive algorithm and the network calculation cutback caused by using the Gaussi-an function as the basis function of GMAC network .Then it was put forward that the following status of oxygen sensor was predicted with Gray-Markov model to decrease the sensor time-delaying influence on fuel compensation .The structural parame-ters of control model were determined with the bench test data and the proposed adaptive-hybrid control algorithm was further tested .The results show that the control method can ensure the control accuracy of air-fuel ratio under steady state and improve its control validity under transient state .