基于神经网络的气缸压力识别研究

对传统的内燃机气缸压力识别方法存在的问题 ,提出了应用人工神经网络方法进行气缸压力识别的新方法。以 BP网络构造气缸压力识别模型。通过对网络的训练 ,用实测的缸盖振动信号识别气缸压力。结果表明 ,利用神经网络进行内燃机气缸压力识别 ,识别结果的重复性好 ,精度较高     

基于EMD奇异值分解与马氏距离的气阀故障诊断

设置排气阀的不同间隙及用新气阀模拟轻微漏气,构造气阀机构的四种常见工作状态,然后针对非平稳的缸盖振动信号,介绍了一种可以处理非平稳信号的新方法,应用Hilbert-Huang 变换的核心内容--经验模态分解法对非平稳信号进行分解,以降低原始信号中的非平稳性.利用经验模态分解和奇异值分解得到缸盖振动信号的故障特征参数,然后用少量的样本数据训练得到四种常见工作状态的模式向量,最后利用马氏距离判别函数进行气阀机构故障状态的识别.实验结果表明通过小样本即可完成模型的训练,且训练一旦完成,对未知样本的分类速度和识别率都很高,便于实现气阀机构故障的在线实时监测与诊断.     

EWT-FastICA在内燃机振动信号识别中的应用

内燃机广泛应用于工程、动力等领域,然而内燃机因燃烧和机械运动引起的冲击与振动导致其减振降噪一直是研究的热点,而如何准确识别振源则是减振的前提.本文针对振源盲分离时观测信号不少于源信号数目要求不易满足的问题,利用经验小波变换(Empirical wavelet transform,EWT)结合快速独立成分分析(Fast independent component analysis,FastICA)实现对内燃机振源信号的识别.首先使用时域同步平均法对内燃机缸盖的振动信号进行预处理,然后进行经验小波变换,之后再利用皮尔逊相关系数选择有效经验模态分量作为快速独立成分分析(FastICA)的输入,最终分离结果表明:该方法可以有效地从内燃机缸盖振动信号中识别出燃烧信号和气阀机构开启时的气体冲击信号.     

基于变分模态分解和鲁棒性独立成分分析的内燃机缸盖振动信号分离

针对缸盖振动信号中燃烧信号和活塞敲击信号在时频域混叠严重难以分离的问题,用基于VMD和RobustICA的方法分离各独立源信号。通过试验测量内燃机单通道缸盖振动信号,首先对测得的信号进行消除趋势项及滑动平均等预处理,用VMD算法对预处理后的信号进行分解;然后用RobustICA算法提取独立成分,并用组合模态函数法对时域和频域相似性较高的分量成分进行组合;最后结合频谱分析、连续小波变换、相干函数法及倒拖试验对分离得到的结果进行识别验证。研究结果表明：在不同的试验工况下,该方法可以有效地从缸盖振动信号中分离出燃烧信号和活塞敲击信号。     

HHT与Elman神经网络在离心泵故障振动信号处理中的应用

提出了一种Hilbert-Huang变换(HHT)和Elman神经网络相结合的离心泵振动信号故障诊断新方法.首先,将离心泵振动信号时间序列数据经验模态分解(Empirical Mode Decomposition,简称EMD),然后经过Hilbert-Huang变换获得各模态(Intrinsic Mode Functions,简称IMF)的能量,并以"能量比"为元素,构造离心泵振动信号的特征向量,根据Elman神经网络模型能够逼近任意非线性函数的特点和具有反映系统动态特性的能力,利用Elman神经网络模型实现离心泵故障的诊断.实验研究结果表明该方法对离心泵振动信号故障具有很高的诊断率.     

基于深度学习理论的发动机气门机构故障识别

为了正确诊断和识别发动机气门机构故障,提出一种基于经验模态分解和堆栈式稀疏自编码器深度学习模型的发动机气门机构故障识别算法。以发动机缸盖振动信号为信号源,对振动信号做经验模态分解,提取各个本征模态分量的时域和频域特征构成故障特征向量集,作为故障识别的样本变量。通过稀疏自编码非监督学习网络对输入向量进行特征学习,并将单层网络堆栈成深度网络,最后采用少量有标签数据对整个深度学习模型进行微调训练,建立气门机构故障识别模型。试验结果表明,EMD-SSAE混合深度学习模型能够有效的识别气门机构的故障状态,并且比EMD-SVM和EMD-BPNN模型获得更高的识别准确率。     

基于角域同步平均技术的内燃机失火故障诊断

在利用缸盖振动信号诊断内燃机失火故障时,由于发动机工作背景噪声复杂,必须消除信号中的非周期分量和随机干扰,保留与发动机工作循环有关的周期分量。为了解决时域同步平均方法在转速波动时振动信号存在的不同步问题,提出了以旋转角度信号作为同步触发基准的角域同步平均技术,对内燃机缸盖振动加速度信号进行了处理,有效地削弱了随机噪声的干扰。分析了缸盖振动信号中不同激励源产生的响应分量与发动机失火故障的关系,利用缸盖振动加速度信号中的各个瞬态冲击响应更加易于识别,能有效地对内燃机失火故障进行诊断。     

基于Hilbert——数学形态变换的柴油机燃烧始点辨识

为了识别燃烧始点,本文对比了基于油压、缸压以及振动信号识别方法的优缺点,最后采用振动信号识别.在燃烧始点气缸压力急剧升高,缸盖振动响应也更为强烈.通过对信号进行一系列的Hilbert变换、形态滤波、差分运算、最后设置阈值即可提取燃烧始点.实验分析得出康明斯K38发动机燃烧始点约在上止点前20°CA,结果表明该方法具有稳定性、准确性和可行性.     

基于相关能量波动评估的学习样本筛选与深度置信神经网络的滚动轴承故障诊断研究

在数据驱动的滚动轴承状态智能辨识中,针对辨识模型构建过程中由于学习样本“质量差”造成其故障识别率低的问题,提出一种如何筛选学习样本的准则来提升基于深度置信神经网络滚动轴承智能辨识模型识别率的方法。首先,基于变分模态分解将具有时变调制特性的滚动轴承振动信号分解为有限个表征原信号不同成分的本征模函数分量;其次,根据其故障能量波动及其相关来量化并评估上述每个分量包含故障成分的比重,并据此对振动信号进行筛选重构来获取学习样本;最后,将上述学习样本集作为深度置信网络的输入来构建滚动轴承的故障辨识模型。实验结果表明,所提方法不仅筛选出滚动轴承振动信号中包含故障主成分的本征模函数分量并实现学习样本集构建,而且提高了基于振动数据的滚动轴承状态辨识模型的故障识别率。     

运用希尔伯特黄变换的桥梁颤振导数识别

提出了一种基于希尔伯特黄变换(Hilbert-Huang Transform,简称HHT)识别桥梁颤振导数的方法.根据HHT理论,首先利用经验模态分解(Emprical Mode Decomposition,简称EMD)将桥梁节段模型风洞试验中实测得到的自由振动加速度响应信号分解成一系列本征模函数(Intrinsic Mode Function,简称IMF)分量,对幅值最大的IMF分量作希尔伯特变换,得到其瞬时振幅和瞬时频率.通过信号处理,识别出结构的固有频率和模态阻尼比.其次,对自由振动时程进行时域拟合,获得结构的复振型.最后通过状态方程确定系统的8个颤振导数.数值仿真算例表明,该方法具有较好的抗噪性和可靠性,风洞试验也验证了该方法的可行性和有效性.     

分形维数在内燃机振动诊断中的应用

将分形理论引入内燃机的振动诊断中,根据内燃机的配气定时,着重研究了缸盖振动信号中对应燃烧段的数据,计算其关联维数,将关联维数用于刻划内燃机缸盖在气门不同状态时表现的非线性行为,从而进行故障诊断与分类。结果表明,当气门在不同状态时,缺盖振动信号中对应燃烧段数据的关联维数是不同的,可以将其作为判断气门漏气的一个诊断特征量。     

Cylinder pressure reconstruction based on complex radial basis function networks from vibration and speed signals

Methods to measure and monitor the cylinder pressure in internal combustion engines can contribute to reduced fuel consumption, noise and exhaust emissions. As direct measurements of the cylinder pressure are expensive and not suitable for measurements in vehicles on the road indirect methods which measure cylinder pressure have great potential value. In this paper, a non-linear model based on complex radial basis function (RBF) networks is proposed for the reconstruction of in-cylinder pressure pulse waveforms. Input to the network is the Fourier transforms of both engine structure vibration and crankshaft speed fluctuation. The primary reason for the use of Fourier transforms is that different frequency regions of the signals are used for the reconstruction process. This approach also makes it easier to reduce the amount of information that is used as input to the RBF network. The complex RBF network was applied to measurements from a 6-cylinder ethanol powered diesel engine over a wide range of running conditions. Prediction accuracy was validated by comparing a number of parameters between the measured and predicted cylinder pressure waveform such as maximum pressure, maximum rate of pressure rise and indicated mean effective pressure. The performance of the network was also evaluated for a number of untrained running conditions that differ both in speed and load from the trained ones. The results for the validation set were comparable to the trained conditions.     

Multivariate statistical analysis strategy for multiple misfire detection in internal combustion engines

This paper proposes a multivariate statistical analysis approach to processing the instantaneous engine speed signal for the purpose of locating multiple misfire events in internal combustion engines. The state of each cylinder is described with a characteristic vector extracted from the instantaneous engine speed signal following a three-step procedure. These characteristic vectors are considered as the values of various procedure parameters of an engine cycle. Therefore, determination of occurrence of misfire events and identification of misfiring cylinders can be accomplished by a principal component analysis (PCA) based pattern recognition methodology. The proposed algorithm can be implemented easily in practice because the threshold can be defined adaptively without the information of operating conditions. Besides, the effect of torsional vibration on the engine speed waveform is interpreted as the presence of super powerful cylinder, which is also isolated by the algorithm. The misfiring cylinder and the super powerful cylinder are often adjacent in the firing sequence, thus missing detections and false alarms can be avoided effectively by checking the relationship between the cylinders.     

内燃机气阀漏气故障的高阶谱分析

将高阶谱理论引入内燃机的振动诊断中,分析了不同状态时缸盖表面振动信号的三阶谱特性,并计算出三阶谱的峰值,用于刻划各状态时缸盖系统的非线性行为。结果表明:正常状态时缸盖表面振动信号的三阶谱接近为零,可以认为缸盖系统是线性系统;当气阀发生漏气故障时,缸盖表面振动信号的三阶谱就会出现较大的峰值,而且不同状态时所对应的峰值也存在着较大差别,说明不同气阀漏气状态时缸盖系统表现出不同程度的非线性。可以将三阶谱的峰值作为判断气阀是否漏气的一个诊断特征量,同时也为诊断内燃机气阀的早期漏气故障提供了依据。     

图像处理在内燃机故障诊断中的应用研究

在应用图像处理进行故障诊断的基础上 ,探讨了应用这一技术进行内燃机故障诊断与状态监测时应注意的一些问题。指出利用缸盖振动信号小波包分解后的时 -频分布图的灰度直方图进行故障诊断的效果并不好 ,并分析了原因 ;进行了信号加噪声与不加噪声的诊断效果对比 ,发现噪声对基于图像处理的气阀机构故障诊断的影响不大 ,验证了这一方法的工程实用性和可行性     

Torsional system parameter identification of internal combustion engines under normal operation

For internal combustion engines, lumped-mass models of the crankshaft system are frequently used for torque estimation in control and diagnostic applications, such as cylinder balancing and misfire detection. Due to inherent model uncertainties and changing system dynamics it may be necessary to adapt the model parameters from time to time in order to preserve the required model accuracy. In this paper a frequency-domain method for on-line identification of the parameters describing the torsional dynamics of internal combustion engines is presented. In the proposed method, the engine is excited by adjusting the cylinder-wise injected fuel amounts, and the measured responses in torsional vibration frequency components are used for parameter estimation. As the fuel-injection adjustments can be determined in such a way that the net indicated torque is unaffected, the identification can be performed on-line without disturbing normal engine operation. The procedure can be applied to estimate the torsional stiffness and damping parameters of the flexible coupling connecting the engine and the load. In addition, the gains which describe how the cylinder-wise fuel injections affect the amplitudes of relevant torsional vibratory frequency components are obtained. The parameter identification method is successfully evaluated in full-scale engine tests on a 6.6 MW six-cylinder medium-speed common-rail diesel engine.     

利用振动加速度识别HCCI发动机燃烧始点

燃烧始点是均质压燃(Homogenous charge compression ignition,HCCI)燃烧控制的基本反馈参数,基于振动加速度信号对HCCI发动机燃烧始点的识别进行研究.建立295HCCI发动机有限元分析模型,基于模型模拟分析了缸盖表面振动加速度信号与缸内压力二次导数之间的关系,指出缸内峰值压力出现...     

基于Hilbert—数学形态变换的柴油机燃烧始点辨识

为了识别燃烧始点，本文对比了基于油压、缸压以及振动信号识别方法的优缺点，最后采用振动信号识别。在燃烧始点气缸压力急剧升高，缸盖振动响应也更为强烈。通过对信号进行一系列的Hilbert变换、形态滤波、差分运算、最后设置阈值即可提取燃烧始点。实验分析得出康明斯K38发动机燃烧始点约在上止点前20℃A，结果表明该方法具有稳定性、准确性和可行性。     

A methodology for combustion detection in diesel engines through in-cylinder pressure derivative signal

In-cylinder pressure measurement has historically been used for off-line combustion diagnosis, but online application for real-time combustion control has become of great interest. This work considers low computing-cost methods for analysing the instant variation of the chamber pressure, directly obtained from the electric signal provided by a traditional piezoelectric sensor. Presented methods are based on the detection of sudden changes in the chamber pressure, which are amplified by the pressure derivative, and which are due to thermodynamic phenomena within the cylinder. Signal analysis tools both in time and in time-frequency domains are used for detecting the start of combustion, the end of combustion and the heat release peak. Results are compared with classical thermodynamic analysis and validated in several turbocharged diesel engines.