基于多维时间序列模型的内燃机磨损状态预测研究

现有的油液光谱数据预测方法仅考虑单一数据内部前后间的联系,忽视不同种类数据间的相互影响。多维时间序列模型能够将多种元素光谱数据融合起来同时进行建模,利用所建模型对光谱数据进行预测,提高预报精度。通过内燃机台架实验获得多种元素的光谱数据,选择典型的磨损元素Fe和Al、污染元素Si以及添加剂元素Mg作为分析元素,通过分析找出相关性较大的元素,利用多维时间序列模型对其进行预报,从而对内燃机的磨损状态进行准确判断。结果表明,将多维时间序列模型引入油液光谱数据预报能对内燃机的磨损状态进行准确预测。     

基于能量耗损的摩擦学系统状态识别方法研究

通过对摩擦学系统的能量流监测,结合状态识别方法的分析,提出了基于磨损与振动能量损耗监测方法,建立了能量损耗的机械设备状态监测框架模型。分析指出,磨损与振动都是能量耗损行为,且二者的相关性采用齿轮疲劳试验证实。研究表明,基于齿轮磨损的光谱分析的元素相对质量分数的累积相对标度和振动信号时域的均方根值的累积相对标度具有很好的相关性,这表明齿轮磨损和振动具有很好的相关性,因此摩擦学系统的状态识别与故障诊断采用能量损耗监测的方法是可行的。     

基于灰色理论的设备磨损状态辨识参数监测法的研究

灰色理论应用于滑油光谱分析的方法已经成为设备状态监测中油液监测的一种有效的建模手段。对取样时间间隔较长的机械设备磨损进行监测，以便得到设备磨损状态的预测信息。本文提出了基于灰色理论的GM(2，1)模型，对设备磨损状态进行预测和诊断分析，并与以往常用的GM(1，1)模型进行了比较，分析讨论这两种模型应用于设备磨损光谱元素浓度预测的精度，通过实例验证了GM(2，1)模型具有简单、准确和实用的特点。并在此基础上，又提出了应用灰色理论对设备磨损状态进行参数辨识，进而运用参数监测来对设备进行状态监测的方法，可以准确地发现系统的变化，为应用光谱分析法判定系统磨损的状态提供了一种定量的和定性的分析手段。     

基于决策树的摩擦学系统状态辨识的知识获取方法研究

摩擦学状态辨识实质上是分类问题，针对以往机器摩擦学状态判别主要依靠人工经验来完成所存在的缺陷，用知识发现的思想来解决摩擦学状态辨识的知识获取问题。采用加权ID3（iterative dichotomizer 3）算法来度量摩擦学状态监测实例表中各条件属性对状态辨识的重要性，建立了基于摩擦学状态监测实例库和决策树的知识获取方法模型。将模型应用于磨损试验的摩擦学状态辨识的知识挖掘分析，利用获得的知识对测试集进行状态识别，取得了良好的摩擦学状态辨识结果，从而为从监测实例中挖掘摩擦学状态辨识知识提供了方法与手段支持。     

基于铁谱技术的内燃机的应用研究

介绍了内燃机的磨损及典型磨粒特点,利用铁谱技术作为一种监测手段,通过磨粒分析来判断内燃机的磨损状态.     

内燃机缸套活塞环系统的摩擦学设计

本文在系统思想的指导下,采用理论分析与润滑计算相结合,对内燃机摩擦副缸套一活塞环系统在正常工作时的摩擦,润滑、磨损等摩擦学行为进行了探讨,并根据研究结果。以桶面环为对象,以Ｓ１９５单缸紫机机的参数为例,编掉了通用内燃机摩这设计软件,以实现对内燃机缸套一活塞环系统的摩擦学设计和使用效果的预测。     

基于长短时记忆卷积神经网络的刀具磨损在线监测模型

为了提高机械加工过程中刀具磨损在线监测的准确性,提出了一种基于长短时记忆卷积神经网络（LSTM-CNN）的刀具磨损在线监测模型。在该监测模型中,通过振动、力、声发射传感器对刀具切削过程中的振动、力和声发射信号进行采集,采集的数据其本质为时间序列数据。考虑采集数据的序列和多维度特性,采用LSTM-CNN网络对采集的数据进行序列和多维度特征提取,利用线性回归实现特征到刀具磨损值的映射。通过实验验证了该模型的有效性和可行性,模型的精度较其他几种方法有了较大的提高。     

基于支持向量机的摩擦学系统状态判别方法研究

传统的摩擦学系统状态的判别采用逐步判别分析法,该方法以油液监测历史数据为依据,通过从不同的油液监测方法中获取信息,构造出状态分类判别数学模型来进行状态判别,具有建模样本量大,建模时间长,对建模人员要求较高的缺点。通过引入支持向量机分类方法,缩短了对摩擦学系统进行状态判别的时间,提高了分类效率,最后通过实例说明了该方法的有效性、实用性和良好的推广应用前景。     

内燃机缸套-活塞环摩擦学研究回顾与展望

内燃机缸套-活塞环摩擦副是一个典型的摩擦学系统,其中含有多种类型的摩擦和磨损,润滑、摩擦、磨损的相互作用十分显著。其摩擦学性能对提高内燃机的可靠性和耐久性,保证内燃机经济、可靠地工作具有决定性的作用。其摩擦学问题的研究一直是人们关注的热点之一。     

基于监测实例的摩擦学系统状态辨识的知识获取

对于摩擦学系统监测样本相对比较少或者投入运行时间不长的机器,研究如何实现其摩擦学系统的智能状态辨识具有重要意义。建立基于摩擦学状态监测实例和决策树的知识获取方法模型,并应用于船舶柴油机摩擦学状态辨识的知识获取,利用获得的知识对测试集进行状态识别,取得良好的摩擦学状态辨识结果,从而为从监测实例中挖掘摩擦学状态辨识知识提供方法与手段支持。     

航空发动机油样光谱分析的PSO-LSSVM组合预测方法

油样光谱分析是航空发动机磨损状态监测与故障诊断的重要技术,基于光谱数据的航空发动机状态预测有利于发现航空发动机的早期磨损故障。根据光谱数据特征,选取AR模型、BP神经网络模型以及GM(1,1)预测模型作为基础模型,建立了基于最小二乘支持向量机的组合预测模型,同时,用粒子群算法对LSSVM的正则化参数以及核函数参数进行了优化。最后利用两组实际的航空发动机光谱分析数据对模型进行了验证,与基础模型的对比结果充分表明,提出的带粒子群优化的最小二乘支持向量机(the Least Squares Support Vector Machines with Particle SwarmOptimization-PSO-LSSVM)的非线性变权重组合预测模型具有更好的预测精度。     

Analytical approach to wear rate determination for internal combustion engine condition monitoring based on oil analysis

Wear has important, negative effects on the functioning of engine parts. Additionally, this situation is very difficult to evaluate accurately in oil analysis for engine condition monitoring. Original Equipment Manufacturers (OEM), lubricant suppliers and oil analysis laboratories provide specific guidelines for wear metal concentrations. These limits provide good general guidelines for interpreting oil analysis data, but do not take into account common factors that influence the concentration of wear debris and contaminants in an oil sample. These factors involve oil consumption, fresh oil additions, etc., and particular features such as engine age, type of service, environmental conditions, etc.In this paper, an analytical approach to enable a more accurate wear determination from engine oil samples is developed. The above factors are taken into account and an improved maintenance program for internal combustion engines based on oil analysis is developed.     

Application of grey modeling method to fitting and forecasting wear trend of marine diesel engines

Oil monitoring is an important and useful method for predicting wear failure, and has been used in diesel engines successfully. The diesel engine is the key power equipment in ships and it is a complicated tribological system with uncertainty and indetermination. Grey system theory is suitable for systems in which some information is clear and some is not, so it is feasible to study the wear process of diesel engines with this theory. The unequal interval revised grey model (UIRGM) (1,1) is presented in this paper, which is applicable to original series with unequal intervals and sharp variation. The model that is built is applied to fit and predict element concentration as determined by oil spectrometric analysis. It is proved that UIRGM (1,1) determines the exact turning point, and the fitting and prediction results are acceptable.     

基于铁谱分析的颗粒分类识别方法与应用

铁谱颗粒分析是机器磨损状态监测与维修决策制定最有效的油液分析方法。通过近年来开展工业企业机器油液监测积累的大量铁谱磨粒图像,进行基于不同的颗粒特征的分类识别探究,并基于不同颗粒形成机制与原因提出切合工业现场的润滑管理维保策略。应用实践表明,铁谱分析方法在机器磨损状态监测、润滑磨损诊断机制判别以及企业润滑管理提升活动中仍发挥着积极作用。     

润滑油光谱数据关联度分析在发动机状态监测中的应用研究

针对发动机润滑油液光谱分析数据处理方法中的不足,运用灰色趋势关联度分析方法,以稳定磨损期典型元素间的关联度为参照,对发动机系统状态进行了监测。分析表明,随着发动机运行摩托小时数的增加,元素间关联度会下降。通过分析典型故障取得了满意效果,说明该方法可以对发动机系统进行状态监测。     

An oil monitoring method of wear evaluation for engine hot tests

Abnormal wear of a piston ring-cylinder liner pair may happen after 9 min hot tests of internal combustion engines, while the engine performance parameters were within predetermined threshold ranges. Few differences were observed among oil samples from the engines with or without abnormal wear in the spectrometric and Kittiwake Analex PQ analysis. Therefore, a manual confirmation by disassembling the oil pan was often required. In this work, an oil monitoring method for wear evaluation of the engines was proposed. The oil samples were rapidly analyzed on site by on-line visual ferrograph (OLVF). For the abnormal engines, it was found that the index of particle coverage area (IPCA), characterizing the wear debris concentration, was low. Moreover, large debris was rarely observed on OLVF ferrograms, which was consistent with the results obtained from analytical ferrography, and the reason was analyzed and discussed. In addition, an on-site abnormal wear evaluation procedure for the 9 min hot tests was proposed based on a trained Naive Bayes Classifier. As observed from the results of 27 engines, 4 abnormal engines were found. Among one of them, longitudinal scratches were found on the cylinder wall, which were evaluated as abnormal wear by the classifier. This method can cut down the quantity of disassembly inspection and is more efficient.     

OIL MONITORING DIAGNOSTIC CRITERIONS BASED ON MAXIMUM ENTROPY PRINCIPLE

A method of applying maximum entropy probability density estimation approach to constituting diagnostic criterions of oil monitoring data is presented. The method promotes the precision of diagnostic criterions for evaluating the wear state of mechanical facilities, and judging abnormal data. According to the critical boundary points defined, a new measure on monitoring wear state and identifying probable wear faults can be got. The method can be applied to spectrometric analysis and direct reading ferrographic analysis. On the basis of the analysis and discussion of two examples of 8NVD48A-2U diesel engines, the practicality is proved to be an effective method in oil monitoring.     

Condition monitoring techniques for an internal combustion engine

Various condition monitoring techniques were applied during a laboratory engine test in order to understand the wear processes occurring and to determine a suitable method which could be applicable to the detection and diagnosis of abnormal engine conditions in practice. Fuel consumption rates were measured in conjunction with chemical analysis of the used oil by Fourier Transform Infrared Spectroscopy, and the various contaminant inspection methods. For the contaminant inspection of the wear debris in the lubricating oil, the quantitative and qualitative techniques of the Rotary Particle Depositor (RPD), Particle Quantifier (PQ), Spectrometric Oil Analysis, and Image Analyser Systems were applied and the results compared. The effectiveness of each to be able to respond to wear failure was also compared. Evaluation showed that, of the methods examined, the combination of RPD and PQ was the most suitable technique for the detection of abnormal wear occurring in the engine combustion zone. It is, therefore, suggested that this combination can be directly applied to internal combustion engines as an effective condition monitoring method.     

The development of Ferrography as a laboratory wear measurement method for the study of engine operating conditions on diesel engine wear

Ferrographic oil analysis techniques were used in a laboratory study of diesel engine wear. Data were developed supporting the concept of using the Severity Index I_s to rank the effect of engine operating conditions on wear. Results analyzing the Severity Index as a function of time and as a function of engine operating variables are presented. The Severity Index is also linearly correlated to spectrometric data (iron and lead concentrations in the used oil samples). Engine wear tends to increase with increase of either oil or coolant temperature. However, brake specific fuel consumption tends to decrease as oil and coolant temperatures increase, indicating a need for accurate temperature control for both mediums to minimize fuel consumption and wear. The heated Ferrogram analysis (HFA) technique was used to determine changes in the wear rates of specific engine parts with variation of the oil and coolant temperatures.     

A model and the methodology for determining wear particle generation rate and filter efficiency in a diesel engine using ferrography

A mathematical model was developed that describes the wear particle concentration as a function of time in a diesel engine. This model contains engine and lubrication system parameters that determine the concentration of wear particles in the engine sump. These variables are the oil system volume, oil flow rate, particle generation rate, filtering efficiency and the initial particle concentration. The model was employed to study the wear particle concentrations in the sump and the mass of particles in the filter for the Cummins VT-903 diesel engine. In addition, the model was used to develop a testing methodology for determining wear particle generation rates and filter efficiencies from used oil analysis. This testing methodology uses ferrography together with computer programs to yield accurate statistical information on the data as curve fitted to the model. The test set-up incorporated a remote-controlled sampling system that enabled the accurate and periodic taking of oil samples over an engine test approximately 5 h in duration.

Results of this research indicate that equilibrium wear particle concentrations increase with an increase in engine speed and load. The wear particle generation rate and filter efficiency as determined by the test methodology were found to decrease with an increase in engine speed and load. After oil and filter changes, the wear particle generation rate and filter efficiency continually increased with cumulative engine time up to approximately 11 h. The test methods used to obtain the results above were found to be repeatable to within ±15% and could conceivably be employed to determine wear parameters on other diesel engines as well as the effects that other engine variables such as lubricants, oil temperature, coolant temperature and engine components have on the wear parameters.