Search

This paper presents a new approach for reliability test sampling plan of Railway equipment. Based on the interval censored failure data, the mean rank order method is used for data pre-process and the least square method is used for fitting the parameters of Weibull distribution, where, the best estimations of distribution parameters are derived. The sampling inspection plan for interval censored data is established when the lifetime follows the Weibull distribution. In the case study, we applied the proposed method to design the reliability test sampling plan of an electric multiple units.

The failure data of the electromechanical product satisfies a certain law that a batch of failure data usually meet a variety of distributions at the same time. As to this phenomenon, an optimization method of evaluation model of reliability which is combined with the concept of entropy was put forward in this work. In the optimization process, several empirical models and multiple evaluation indexes were proposed for selecting a best model of reliability evaluation model, and entropy weight was defined to show the importance of evaluation indexes.

The high reliability and the short times carried of quantitative risk assessment of Filling system in launching site, the failure data is small sample data. In this paper, we will focus on these small sample data and take processing analysis. A method based on Bayesian method of failure data processing analysis of Filling system has been proposed, which firstly analyzes the characteristic of failure data of Filling system. Secondly, we classify the failure data into running failure data and demanding data according to the data types, and further analyze various uncertainty distributions under this two kinds of data type. What’s more, we use Bayesian method to solve the problems of various failure data. Lastly, we combine the real circumstance of Filling system and take the Filling system as example, and choose the filling pump, shut-off valve and pneumatic ball valve these three different failure types to use Bayesian method to analyze, and obtain various prior distributions and posterior distributions of different failure equipment, and give the simulation results.

As the Wireless Sensor Networks (WSNs) are widely implemented in various fields recent years, the quality of WSNs has been increasingly concerned. WSNs can usually be divided into sub-nets, which assumed to work or fail independently. Through the failure data of those sub-nets, the additive NHPP model for reliability evaluation is composed, and then the maximum likelihood estimation is applied to estimate the unknown parameters in the model. Finally, the simulation shows that the additive NHPP model is better than general NHPP model under certain circumstances.

The common faults of mine ventilator are researched in this paper, and rotor misalignment, unbalance, oil whirl, surge and other faults and fault characterization of the generation mechanism are analyzed. The faults diagnosis system is designed based on rough neural network. First, the characteristics of the type of fault for fan failure data collection, including vibration and temperature signals. Then, the pretreated sample data using rough set attribute reduction method to delete redundant attributes. Finally, the sample data is divided into training and testing samples, were used to train and test the neural network classifier. Experiments show that the system is reliable, diagnostic yield, improved ventilator system security, expanding the scope of application of rough sets.

The expert system that based on knowledge base of association rule and to discuss fault diagnosis expert system for the application of fault diagnosis in the Electrical Multiple Unit (EMU) was described. Through the analysis of failure data of EMU operation group to find out the potential relationship between the state data and fault data from a large number of operation and maintenance history data of EMU. This kind of relationship was stored in the knowledge base. Fault diagnosis expert system model is established, which based on the fault intelligent diagnosis expert of EMU. It can achieve dynamic fault diagnosis of the EMU in the process of running and ensure its safe operation.

There is often more than one failure at the same time when NC machine tools fails, but only one failure data is recorded which leads to that the reliability model of NC machine tools is not rational. In addition, the assumption of “repaired as good as new” for NC machine tools which consists of hydraulic, mechanic and other complicated subsystems is questionable and unreasonable. In this paper a method of reliability modeling using Blocksim and Weibull++ for NC machine tools is proposed. In the end, one example is provided and examined to show its potential applications and benefits. The result shows that the method of reliability modeling is more scientific and more in line with the practice.

As the tool magazine and automatic tool changer (ATC) are the important function units to reserve and change tools for the machining center, the tool-changing rate affects their reliability life which directly influences the performance level of the machining center. However, the reliability of the tool magazine and ATC has not yet got a reasonable evaluation index. Directing at the features of the tool magazine and ATC system, the mean tool change numbers between failures evaluation index is proposed, which is used to calculate the mean time to repair and estimate the collected failure data of the chain-type tool magazine and ATC. It can be seen that the tool magazine with continuous tool-changing adopts the mean tool change numbers between failures as the reliability evaluation index.

According to the main characteristics and functions of a high-speed punch, the system was divided into nine function subsystems systematically. Based on failure data, the failure positions, failure mode, failure causes and failure cause classification were detailed statistical analyzed. And the weakest links and potential shortcomings of the high-speed punch were pointed out. Some failure mechanisms of the high-speed punch, such as oil/gas leakage, crankshaft failure and so on, were analyzed especially. The improvement measures, which lay the foundation for the reliability growing of the high-speed punch, were drawn based on the failure and failure mechanisms analysis.

This paper discusses the principle and method of Double-Crossed Step-Down-Stress Accelerated Life Testing (DCSDS-ALT) for pneumatic cylinders. As to pneumatic cylinder, the step-down-stress testing failure physics can be described as cumulative degradation model. Temperature and frequency are normally chosen as the test stresses. The failure data obtained under DCSDS-ALT testing steps can be converted to those under constant stress testing. And the reliability specifications can be derived accordingly. To compare Double-stress ALTs with traditional constant stress testing shows Double-stress ALTs can meet the accuracy demand. The 5% of average lifetime estimation error and 1.85% of the characteristic lifetime error are very satisfying for pneumatic industrial lifetime prediction.

An automatic tool changer (ATC) is one of the most important subsystems of a computerized numerical control (CNC) lathe. The field failure analysis is very important to improve the reliability of the ATC. The structure and mechanism of an ATC is briefly introduced. Its brief reliability block diagram is set up. With this understanding, through the collection of the field failure data, a coding system to code failure data is devised and a failure analysis database of CNC lathes is accrued. The failure mode and cause are analyzed to indicate the weak subsystem of the ATC. Some suggestions on improving the reliability have been proposed.