Diesel engines have a wide range of functions in the industrial and military fields. An urgent problem to be solved is how diagnose identify their faults effectively timely. In this paper, diesel engine acoustic fault diagnosis method based on variational modal decomposition mapping Mel frequency cepstral coefficients (MFCC) long-short-term memory network proposed. Variational mode (VMD) used remove noise from original signal differentiate into multiple modes. The sound pressure signals...

Abstract The traditional prediction of remaining useful life (RUL) for bearings cannot be calculated in parallel and requires manual feature extraction artificial label construction. Therefore, this article proposes a two-stage framework predicting the RUL bearings. In first stage, an unsupervised approach using temporal convolutional network (TCN) is employed to construct health indicator (HI). This helps reduce human interference reliance on expert knowledge. second based neural...

This paper addresses a selective maintenance optimization problem for fuzzy multi-state system composed of elements. Due to insufficient data and unpredictable external working conditions, both the performance capacity states transition intensities elements ca nnot be known precisely, but are represented by numbers. Additionally, durations break succeeding mission also treated as values. To maximize probability successfully completing mission, model is proposed identify an optimal subset...

Due to the relative insufficiencies of conventional time-domain waveform and spectrum analysis in fault diagnosis research, a diesel engine method based on Stacked Sparse Autoencoder Support Vector Machine is proposed this study. The consists two main steps. first step utilize (SSAE) reduce feature dimension multi-sensor vibration information; when compared with other reduction methods, approach can better capture nonlinear features, so as cope reduction. second diagnosing faults,...

The construction of health indicators (HI) for traditional deep learning requires human training labels and poor interpretability. This paper proposes an HI method based on Stacked Sparse Autoencoder (SSAE) combines SSAE with Long short-term memory (LSTM) network to predict the remaining useful life (RUL). Extracting features from a single domain may result in insufficient feature extraction cannot comprehensively reflect degradation status information mechanical equipment. In order solve...

Reliability allocation is an important step of system design. The design scheme for the directly under influence results. AGREE one major method reliability allocation. In this paper, deficiency was discussed, such as depiction importance and its applicability. Fault damage degree fault were introduced. Through improvement traditional method, new models both serial parallel connection systems designed. Because not only results will be more practicability, but also application range wider. At...

In both industrial and military fields, there is such a kind of complicated system termed as phased-mission system, which executes missions composed several different phases in sequence. The structure, failure behavior, working conditions may change from phase to phase. duration each involved random follows probability distribution, the suffer some events resulting simultaneous failures elements with probabilities. order guarantee completes successfully, selective maintenance model for...

This article addresses a selective maintenance optimisation problem for systems subject to random common cause failures. A system is likely suffer from several failures during given mission. Random events, which occur with specific probability distribution, may result in the simultaneous of multiple elements. Because time one most crucial resources, time-based imperfect model proposed quantify efficiency each candidate action. To meet demands next mission, optimally identify subset...

Selective maintenance is a process of identifying the sets components to be repaired and corresponding activities performed when given limited resources, such as time, budget, repairman, spares, etc. Examples selective abound in applications manufacturing systems, military power generation In this paper, simulation method proposed optimally select schemes, including selected tasks allocation with objective maximising system availability. Genetic algorithm (GA) adopted allocate repairman. An...

The gearbox is one of the key components many large mechanical transmission devices. Due to complex working environment, vibration signal stability gear box poor, fault feature extraction difficult, and diagnosis accuracy makes it difficult meet expected requirements. To solve this problem, paper proposes a method based on an optimized stacked denoising auto encoder (SDAE) kernel extreme learning machine (KELM). Firstly, particle swarm optimization algorithm in adaptive weight (SAPSO) was...

To improve the efficiency and scientificity of Reliability Centered Maintenance (RCM) analysis, in paper, Case-Based Reasoning (CBR) maintenance decision models are both successfully introduced into analysis process, a framework for case-based model-based RCM is presented, critical components which, such as case representation, reasoning mechanism etc, discussed detail. The idea based on fact that historical records similar items can be referenced used current new item, furthermore...

For the optimization of complex system RCM tasks, group maintenance policy is introduced, and mathematical models are established to minimize whole cost per unit time. The traditional periodic replacement functional check considering potential failure analyzed, respectively; integrating several types preventive maintenance, model obtained. By joint execution times loss shut-down reduced can be saved. Finally, effectiveness proposed illustrated through a numerical example.

The robust problem of age replacement models in maintenance management is discussed this paper. characters optimal solutions models, with cost and availability objects, are presented. A new algorithm based on Newton method established to calculate the interval rate. Comparing TTT (Total Time Test) plot method, much faster calculation better accuracy were achieved. Some values a parameter fixed investigate variation costs when changing other parameters. Weibull distribution used as typical...

Abstract A high-performance frequency synthesizer based on phase-locked loop (PLL) technology is proposed for high-power, solid-state microwave sources. The PLL circuit simulated, optimized, and analyzed by using ADIsimPLL software. methods to improve the phase noise performance of source are summarized. Using ADF4351, STM32, ADL5611, other devices complete hardware software control part design, low successfully established. can output signal in range 2.3 G-2.5 G, within - 97.30 dBc@1 KHz,...

Selective maintenance is a process of identifying the sets components to be repaired and corresponding activities performed when given limited resources, such as time, budget, repairman, spares, etc. Examples selective abound in applications manufacturing systems, military power generation In this paper, simulation method proposed optimally select schemes, including selected tasks allocation with objective maximising system availability. Genetic algorithm (GA) adopted allocate repairman. An...

This paper presents a methodology to perform cost-benefit analysis on the application of Prognostics and Health Management (PHM) technology for equipment. The purpose is estimate economic feasibility PHM technology. discusses calculation metrics technology, determines potential Return Investment (ROI) Payback Period approach relative unscheduled maintenance policy using Monte Carlo simulation. case study kind electronic equipment indicates proposed can be used as useful tool aiding decision-making

We proposed a new approach for target value sequencing problem using grey analytic network process(ANP). Target evaluation index system was constructed located at artillery battlefield, the weight decided based on process, evaluated league and general vector were theory. The case proves reliability of results validity this method. This method provides commander with reference to theoretic basis decision-making sequencing.

The maintenance and support system is analyzed on surface-to-air missile, the process model proposed repair process, prevention spare parts supply equipments usage can be used in supportability, it has general use reference to other equipments.

As a strategy of maintenance, RBM (Risk Based Maintenance) are playing more and important role in the maintenance large equipment corporations. After analyzing meaning characteristics RBM, authors despicted decision flow base on RBM. The risk assessment model weapon materiel was constructed index analyzed detail. Finally, made some useful suggestions implementing mode based

In order to identify rational maintenance level for the fault of Self-Propelled Gun, this paper proposes a severity degree criterion and adopts fuzzy Petri net establish corresponding information expression model allusion recoil mechanism Gun. analysis method graph theory is also adopted deal with qualitative evaluation database.