基于油液监测的石化机组磨损评价方法研究

发布时间：2018-05-29 11:23

本文选题：泵 + 油液监测技术　；参考：《太原理工大学》2017年硕士论文

【摘要】：泵作为石油化工行业最常见的物料输送和能量转换设备,在整个工艺流程中扮演着非常重要的角色,然而在实际应用中却经常因摩擦磨损而发生机械故障。润滑油作为泵常用的润滑剂有着减小摩擦磨损、降低温度、缓冲减震和形成密封的作用,同时会携带有大量设备磨损和润滑的相关信息,通过对这些摩擦信息的处理和研究便可得知泵目前的磨损状况和未来的变化趋势,本文在基于油液监测技术的基础上,提出了一种模糊层析综合评价方法,用于评判泵的磨损状况。基于油液监测技术的泵磨损状态模糊层析综合评价主要包括了以下四个内容:首先,介绍油液监测技术等相关方面知识,并针对设备现场工况制定了相应的监测方法、采样要求、采样周期和结果反馈形式等,以及各检测技术手段的基本原理和检测结果表征意义;其次,根据检测技术和手段的不同建立了评价指标体系,该体系包含目标层、准则层和指标层,并对指标层各指标参数进行了分类研究。不同指标对评价结果的贡献值不同,在评判过程中这种不同主要通过权重来体现。所以以准则层评价指标为单位,分别对其下属的指标层指标进行主观赋权和客观赋权,再用乘法集成法进行组合赋权。模糊综合评价的核心是隶属关系的建立,本文的隶属关系以隶属函数来体现,指标层各指标与目标层评价集各因素的隶属函数由模糊分布法得到。评价矩阵和权重集的合成需要通过模糊算子(模糊合成模型),本文所选用的模糊算子为。由因素集、评价集、权重集、隶属关系和模糊合成模型,就构建成了基于油液监测技术的泵磨损状态模糊层析综合评价模型。最后,以现场采集油样的检测数据为依据,对泵磨损状态进行评价,评价方向采用自下而上的逐层评价方向。最后,以评价结果中隶属于正常磨损状态的隶属度为预测参数,采用灰色预测模型GM(1,1)对设备磨损状态进行预测,将预测结果与实际评价结果相对比,计算预测精度,验证预测的正确性。
[Abstract]:As the most common material transportation and energy conversion equipment in petrochemical industry, pump plays a very important role in the whole technological process. However, in practical application, mechanical failure often occurs due to friction and wear. As a common lubricant of pump, lubricating oil can reduce friction and wear, reduce temperature, buffer vibration and form seal, and it will carry a lot of information about equipment wear and lubrication. Through the processing and research of these friction information, we can know the current wear condition and the future change trend of the pump. Based on the oil monitoring technology, a fuzzy chromatographic comprehensive evaluation method is put forward in this paper. Used to judge the wear and tear of the pump. The comprehensive evaluation of pump wear state fuzzy chromatography based on oil monitoring technology mainly includes the following four contents: firstly, the related knowledge of oil monitoring technology is introduced, and the corresponding monitoring method is developed for the field working condition of the equipment. Sampling requirements, sampling period and the form of result feedback, as well as the basic principles of the detection techniques and the significance of the detection results. Secondly, according to the different detection techniques and means, the evaluation index system is established, which includes the target layer. Criterion layer and index layer, and the index layer of the index parameters are classified. The contribution value of different indexes to the evaluation results is different, which is mainly reflected by the weight in the process of evaluation. So take the criterion level evaluation index as the unit, respectively carry on the subjective weight and the objective weight to its subordinate index layer index, then use the multiplication integration method to carry on the combination weight. The core of fuzzy comprehensive evaluation is the establishment of membership relationship. The membership relationship of this paper is reflected by membership function. The membership function of each index of index layer and evaluation set of target layer is obtained by fuzzy distribution method. The combination of evaluation matrix and weight set needs to pass through fuzzy operators (fuzzy composition model), the fuzzy operator selected in this paper is. Based on the factor set, evaluation set, weight set, membership relation and fuzzy synthesis model, a comprehensive evaluation model of pump wear state fuzzy chromatography is established based on oil monitoring technology. Finally, based on the oil samples collected on the spot, the wear status of the pump is evaluated, and the evaluation direction is the bottom-up evaluation direction. Finally, the wear state of the equipment is predicted by using the grey prediction model GM1Q1). The prediction results are compared with the actual evaluation results, and the accuracy of prediction is calculated by comparing the prediction results with the actual evaluation results, taking the membership degree of the evaluation results as the prediction parameters, and using the grey prediction model GM1 / 1) as the prediction parameter. Verify the correctness of the prediction.
【学位授予单位】：太原理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TQ051.21;TE65

【相似文献】