Masdi Muhammad

Methodology for short-term performance prognostic of gas turbine using recurrent neural network

The issue of performance prognosis has been a topic of considerable interest in industrial condition monitoring applications. An innovative data driven prognostic methodology has been introduced in the current study by utilizing artificial recurrent neural network (RNN) approach which intends to improve the capability of equipment performance prediction within a specified short time bound even with limited available data. The ability of the approach is demonstrated using condition monitoring parameters collected from a 20 MW industrial gas turbine. An appropriate selection and fusion of measured variables has been employed to feed RNN with the most influential performance information. The analysis demonstrated that the developed prognostic approach has a great potential to provide an accurate short term forecast of equipment performance which can be invaluable for maintenance strategy and planning.

Time series method for machine performance prediction using condition monitoring data

Accurate machine performance prediction is crucial to an effective maintenance strategy for improved reliability and to reduce total maintenance cost. In this study, a time series neural network based approach is introduced to achieve more accurate and reliable performance prediction of machine using condition monitoring data source. The proposed time series model utilizes the various measured condition monitoring data at the current and previous inspection marks as the inputs, and the machine output performance as the targets for the model. To validate the model, it considers a two-shaft industrial gas turbine as a case study. The collected condition monitoring data are used to train and validate the proposed model. Results showed that the proposed time series method could predict the performance of the gas turbine power output with more accuracy and better results.

Integrating Reliability Analysis in Life Cycle Cost Estimation of Heat Exchanger and Pump

This paper presents a mathematical model to estimate the life cycle cost (LCC) of heat exchanger and pump. Maintenance cost, down time cost and acquisition costs are calculated. The main uncertainty in calculating these costs are prediction of number of failure and cumulative down time. Number of failure is determined using failure and repair time density function. According to the characteristic that the cumulative failure probability observed, a Weibull distribution model is used. The scale and shape parameters of the Weibull are extracted from the published data. The results of the study show that 71.3% loss in the reliability of heat exchanger and 34.2% reliability loss in pump could lead to 66.2 % increment of the total cost. The reliability of the system decreases because of number of failures will increase each year, and this failure leads to unavailability of the system.Therefore in order to achieve higher system effectiveness and reduce the total LCC, the reliability of the systems need to be increased through proper maintenance policies and strategies. The results of the study could assist the managers to make decision with high degree of accuracy.

Power and Thermal Efficiency Study of Offshore Gas Turbines under Various Ambient Temperatures

Gas turbines offer a reduced weight and compact solution for installation on offshore platforms and floating facilities. The purpose of this study is to examine the influence of various parameters on offshore gas turbines performance. Operating measurements of a 23MW gas turbine installed at an offshore oil and gas plant in east of Peninsular Malaysia was used for model verification and evaluation. The results showed that the gas turbine performance improvements involve the study of a wide range of different parameters including ambient temperature, compression ratio, fuel-air ratio and operating load. These achieved relations will help in appropriate assessment of offshore gas turbines thermal efficiency.

A Practical Method for Analyzing Offshore Gas Compressor System Maintenance Data

A systematic approach with proper statistical analysis techniques for analyzing maintenance data can give insight on how well the performance of the existing system. The objective of this study is to present a methodology for systematically analyzing the maintenance data of offshore gas compressor system to gain insight about the system performance and identify the critical factors influencing the performance. The study approach is based on problem and data‐lead rather than technique‐driven. The results of trend test propose that the system under studied can be modeled using a simple Homogeneous Poisson process (HPP) process where the failure rate is constant. Analyses of covariates are done using Kaplan Meier and Proportional hazards models. The results indicate that the preventive maintenance (PM) plus engine wash has a significance influence on the system failure distribution. This covariate is found to play a positive role in extending the inter‐arrival failure times thus improving the system performance.

Probabilistic Life Cycle Costing: A Case Study of Centrifugal Pumps

Life cycle costing (LCC) has been gaining attention in industries as a decision-making tool for the management of assets. LCC is generally recognized as a valuable tool in making an optimal decision considering the total cost of ownership rather than just the initial acquisition cost. The deterministic LCC model is commonly used in many plants; however, the deterministic model only inherently encompasses uncertainty factors, i.e., the economic issue alone, and cannot practically and effectively handle the ambiguous other uncertainties such as changes in interest rate, cost of production loss per hour due to unexpected failure, and labor cost per failure, to name a few. The life cycle cost of a repairable system is closely linked to and highly influenced by its maintenance cost which includes its reliability, maintainability, and maintenance support. Thus, to incorporate these uncertainties into LCC, one needs to consider the application of the reliability engineering principles to evaluate the probabilistic nature of the equipment failures and repairs.

Continuous Life Cycle Cost Model for Repairable System

Traditionally, the estimation of maintenance cost of a repairable system was evaluated using discrete approach based on estimated number of system failure, cost of repair as well as the interest rates. As maintenance cost represents a significant portion of overall life cycle cost (LCC), accurate estimation of maintenance cost would influence LCC analysis. However, in actuality the failure of the repairable system occurs in a continuous probabilistic manner thus the assumption of discrete occurrence is rather inaccurate. This paper presents an alternative continuous LCC model to better represent the actual operating phenomena of repairable system. The model was established based on the widely used Weibull distribution probability density function and continuous combined interest method. The result of the developed LCC model was then validated using Monte Carlo method. The result indicates that the continuous LCC model is able to accurately estimate LCC for any given time that can be useful in decision making based on life cycle cost.

Identification of Critical Factors to System Performance Using Availability Modeling and Simulation Analysis

Availability analysis presents a means to understand the impact of existing maintenance system and maintenance resources to the overall system operational availability. The practical method for conducting availability analysis of a plant system at operation phase is illustrated and discussed via a case study of an acid gas removal system of gas processing plant. This study demonstrates that the availability modeling and simulation is effective in assessing the existing and future system configurations and determining possible impacts and critical factors to systems availability. These findings can significantly assist management to make right actions in improving plant system performances.

Reliability assessment framework for repairable system

System reliability assessment serves as one of the decision tools in selecting the right maintenance strategy. However, selecting the right reliability model can be a formidable task given the vast number of available reliability prediction models This paper presents a framework of selecting the right model based on system failure data with special emphasis on generalized renewal process (GRP) for system that exhibits failure trending The results indicate a better fit for the data with GRP compared with life data analysis approach.

Reliability assessment of multi-state repairable system subject to minimal repairs and constant demand

Effective maintenance management is essential to reduce the adverse effect of equipment failure to operation. This can be accomplished by accurately predicting the equipment failure such that appropriate actions can be planned and taken in order to minimize the impact of equipment failure to operation. This paper presents a model to assess system reliability for a degraded multi-state system based on discrete time Markov process and continuous time Markov process. The selection of which model to use is based on the type of available data. The system degradation was quantified by discrete level of systems performance rate with system states ranging from perfect functioning state to complete failure. At any point in time, the system can experience random failures from any degraded state upon which general repair will be performed. This research also explored a method of estimating of transition probabilities as well as definition of states for the Markov process by utilizing system performance data and data clustering method. The results proved the applicability of both discrete time Markov chain and continuous time Markov process in assessing the reliability of multi-state systems using the systems performance data. The results are then utilized to perform equipment replacement analysis due to deterioration based on the expected demand.