论文部分内容阅读
为解决柴油机故障诊断问题,采用自适应神经模糊推理系统(ANFIS)建立其故障诊断模型。利用减法聚类方法确定模型初始结构,并采用由梯度下降算法和最小二乘算法所组成的混合学习算法优化模型参数。经文中试验数据检验,所建模型故障识别值与实际值之间的最大误差为10.16%,最小误差为0.115%,平均误差为2.26%,识别精度达到了97.74%。仿真结果表明,与BP网络模型相比,该模型收敛速度快,拟合能力强且诊断识别精度高,能够有效识别柴油机故障。
To solve the problem of diesel engine fault diagnosis, an adaptive neuro-fuzzy inference system (ANFIS) was used to establish its fault diagnosis model. Subtractive clustering is used to determine the initial structure of the model, and a hybrid learning algorithm composed of a gradient descent algorithm and a least square algorithm is used to optimize the model parameters. According to the test data in the text, the maximum error between the identified value and the actual value of the model is 10.16%, the minimum error is 0.115%, the average error is 2.26% and the recognition accuracy is 97.74%. The simulation results show that compared with the BP network model, the proposed model has the advantages of fast convergence, good fitting ability and high diagnostic accuracy, which can effectively identify diesel engine faults.