Time filter

Source Type

Moscow, Russia

Kirillov A.,SmartSys Prognosis Center | Kirillov S.,SmartSys Prognosis Center | Pecht M.,University of Maryland University College
Lecture Notes in Mechanical Engineering | Year: 2015

The target problems of PHM computing cluster for different types of maintenance: condition-based maintenance (CBM); predictive maintenance (PdM); self-maintenance and self-recovery, are discussed in this paper. All types of maintenance are determined by the current state of engineering, degree of wear, the duration and conditions of operation. Therefore, various types of maintenance define various financial and time costs. Recent trends in the development of different maintenance strategies are aimed at the creation of self-maintenance and self recovery engineering systems. However, to support such systems are required new models and methods of determining the technical state of engineering and its prognosis. That is, each of the stages of engineering object maintenance must be supported by appropriate methods of diagnosis of the condition of engineering objects, methods of accurate prognosis and assessment of time intervals of prognosis reliability. Consequently, the problem of supporting various types of maintenance and the development of appropriate formalisms, methods and algorithms to analyse the condition of object and prognosis for each type of maintenance should be included in the PHM problems. PHM problems should represent the universal concept and system of algorithms and rules, capable also to an estimation of efficiency of chosen strategy maintenance, minimization of cost and reduce operating costs. These necessary methods should diagnose condition at all operation phases and estimate time of achievement of borders of each operation phase. Thus, PHM problems and determining the time hierarchy predictors of engineering conditions become the base for the development of maintenance and self-maintenance, but not only that. The paper is a review of actual problems for PHM in the context of the practice of application of computing clusters, algorithms and scenarios for the organization of the global system of development of self-maintenance systems based on of computing clusters are specified. This review is based on testing the PHM hierarchical models and algorithms for the pilot version of the PHM cluster at the analysis of failures of internal combustion engines and mechanisms of high complexity. © Springer International Publishing Switzerland 2015.

Kirillova A.,SmartSys Prognosis Center | Kirillova S.,SmartSys Prognosis Center | Pechtb M.,University of Maryland University College
Chemical Engineering Transactions | Year: 2013

The work describes the computing cluster for early fault prognosis and estimates of RUL for technical objects: wind turbines, internal combustion engines, gas turbine, etc. Basic functions of the cluster, consisting of the algorithms are described: -Preventive smart monitoring to detect signs of failure; -Definition of computing models of the time evolution of features and accurate RUL estimate; -Analysis of telemetry data in order to detecting and further monitoring of the hidden signs which give early prognosis (preventive prognosis) of potential risks of failure in the absence of statistical characteristics; -Learning of recognizing automata (HMM, neural networks) for real-time autonomous systems of smart preventive monitoring. The computational algorithms are based on models that combine the methods of the theory of dynamical systems, the theory of stochastic processes, statistical physics and field theory. The organization of algorithms in the hierarchical structure based on the principles of degeneration of failure signs helped identify more early signs of failure. This, in turn, allows to calculate effective Predictive maintenance strategies. The first experimental results of the cluster operation are shown. In particular, it is show that the hierarchical approach to the RUL prognosis has the structure of one-dimensional graph. The experiment confirmed the need to integrate remote computing cluster with peripheral recognizing automata in a united network for effective organization of automata operation and their retraining. Copyright © 2013, AIDIC Servizi S.r.l.

Khodos A.,SmartSys Prognosis Center | Kirillov A.,SmartSys Prognosis Center | Kirillov S.,SmartSys Prognosis Center
Lecture Notes in Mechanical Engineering | Year: 2015

Actual trends in the construction of systems of various types of maintenance; rather their choice in a particular situation is usually determined by the technical state of engineering at the time. Accurate information about its operating condition and prognosis of all detected trends to the side of failures or increasing risks or failures are necessary for the correct determining the optimal maintenance strategy and in general strategy of operational service of engineering object. Thus the life cycle of engineering relatively the types of operational service and maintenance methods is divided into several time intervals determined by the conditions of the object and prognosis. Their maintenance sets of measures are determined at each time interval from operating measures of self-maintenance to end of life management: reuse-remanufacture-recycle. Choice of various types of maintenance is more effective at accurate determining the state of engineering at the failure progression timeline. Account of optimization problem of operating costs throughout the life cycle of engineering in many respects is determined by exact definition of engineering positions on the failure progression timeline and the rate of evolution to failure state of engineering at particular time interval. Thus, different types of monitoring for engineering are required for the construction of systems of local RUL assessments. The considered problems of the state and RUL assessments are solved by integration of remote computing PHM clusters and on-board diagnostic systems of engineering object. RUL evaluation methods in the monitoring process at all time stages of the engineering operating are demonstrated in this chapter. Maintenance strategy for all operating phases and some methods of self-maintenance available for use by varying the control parameters of an engineering object are discussed. In particular, the management solutions in order to minimize maintenance costs for certain time intervals of engineering object operating are given. © Springer International Publishing Switzerland 2015.

Khodos A.,SmartSys Prognosis Center | Kirilllov A.,SmartSys Prognosis Center | Kirillov S.,SmartSys Prognosis Center
Chemical Engineering Transactions | Year: 2013

The model and algorithm for RUL estimation of the rotating machinery is constructed. RUL estimates are presented analytically. Random walk model of finite segments of the wavelet coefficients of observed signal is considered as the original model. We consider all possible cases in the problems of multidimensional random walk vectors free walk, walk with limitations of walk in a multiply connected space. In this case, the problem of RUL estimate is reduced to the assessment of the achievement by the wandering vector the preassigned critical values or areas of critical values. For this the probability of transition from the initial value to some preassigned values is determined using the representation of this probability by the Feynman path integral. During the monitoring of the operation of mechanical block is determined the type of random walk of observable process. After that there is an automatic choice of model for RUL estimates. In some cases it is possible to calculate the Feynman path integral and result analytical formula for RUL estimates. In other cases, after select the type random walk for RUL estimates the solutions of evolution equations for the transition probabilities are studied numerically. It is shown that for the RUL estimates is needed mode of continuous or periodic monitoring, because the state of the mechanical system can change and therefore the conditions of applicability of the random walk model are changed. In this case, the correction of the calculated estimates is needed. Therefore, the paper discusses the issue about prognosis of changing the conditions of applicability of random walk models. The described algorithm is implemented in PHM computing cluster. For full analysis of life time estimates the cloud computing service is used. The experimental data are shown. Copyright © 2013, AIDIC Servizi S.r.l.

Iakimkin V.,SmartSys Prognosis Center | Kirillov A.,SmartSys Prognosis Center | Kirillov S.,SmartSys Prognosis Center
Chemical Engineering Transactions | Year: 2013

This paper describes methods and algorithms for preprocessing of the observed data to fault prognostics. The main focus is on identifying the hidden signs of fault (nonamplitude failure predictor) and the method, defined their evolution. The signal from the vibration sensor, installed on the engine case, is taken as the initial observed signal. Further analysis is based on the representation of the signal in the form of coefficients of its wavelet decomposition. Each coefficient of wavelet decomposition is represented as a discrete sequence of data. Further pre-processing signal is a representation of the signal in the form of finite segments of wavelet coefficients. Thus, the transition is made to the high-dimensional vector processes with discrete time. The last stage of preprocessing is the decomposition of vector processes on "amplitude" and nonamplitude "phase" components (K-decompositions). If the physics of process is known, methods of a nonlinear stochastic filtration are used for allocation of the hidden signs. In the opposite case, particularly interesting from a practical point of view, Kdecompositions of the wavelet coefficients segments are analysed. Pseudo-dynamics of phase vector component is analysed to detect of non-amplitude failure predictors. Algorithm and software in automatic mode analyses the evolution of the phase component on the basis of continuous or periodic monitoring. Prognosis of evolution of the phase variable and life time estimate is based on the definition of evolution equations, or by monitoring the entropy characteristics of the pseudo-phase component of Kdecomposition. Pseudo-phase component is physically interpreted as characteristics of process regularities. The described algorithm is implemented in PHM computing cluster. For full analysis of life time estimates the cloud computing service is used. Copyright © 2013, AIDIC Servizi S.r.l.

Discover hidden collaborations