Search results

A probabilistic‐based design for turbine disk at high temperature can quantify risk and thus identify areas of possible overdesign (conservatism). Moreover, the need for cost‐effective designs has resulted in the development of probabilistic design to quantify the effects of these uncertainties so as to improve the reliability of the component. This paper aims to address these issues.

The flow for probabilistic design established through investigating traditional design methods of the turbine disk at high temperature is divided into the processes of crack initiation and crack growth to find important design inputs at each course, where the probabilistic design criterion has been built based on the deterministic criteria and successful experiences.

The probabilistic‐based design procedure has been demonstrated by taking the reliability design of crack initiation process for turbine disk as the example. The reliability analysis for the disk life after optimization analysis was completed by considering random parameters reflecting the uncertainties. The results showed there was a margin in design for disk life referred to as the probabilistic criterion. This measure was taken by redesigning the structure to reduce the disk's weight within the range of reliability.

The present study provides a method to design aero‐engine components based on probabilistic design for further research.

Moreover, the present study provides a way to design structures based on probabilistic design.

It is proved that probabilistic‐based design could produce a lower weight turbine disk by integrating well‐proved deterministic design methods and tools with probabilistic design techniques while maintaining low failure probability.

The purpose of this paper is to report a study in which the feasibility of conducting probabilistic finite element analysis (FEA) for crane hook design has been explored.

This paper presents the results of probabilistic analysis, where in the input random variables are varied and corresponding variations in the output parameters were observed. In this study, material properties and load have been considered as random input variables and the maximum stress, maximum deflection variations were considered as output random variables.

The probability of occurrence of output variation and the sensitivity of output variables on the input variables are the important results generated from this analysis. By performing this probabilistic analysis, the random selection of factor of safety could be avoided.

The implementation study has been carried out for a single product.

The usage of the approach will indicate the importance of probabilistic analysis in product design and development process. This process will enable the organization to compete in the global market.

A case study has been reported to indicate the feasibility of performing probabilistic FEA for crane hook design. Hence, the contributions are original.

This paper aims to reasonably quantify the radial deformation of turbine blade from a probabilistic design perspective. A probabilistic design for turbine blade radial deformation considering non-linear dynamic influences can quantify risk and thus control blade tip clearance to further develop the high performance and high reliability of aeroengine. Moreover, the need for a cost-effective design has resulted in the development of probabilistic design method with high computational efficiency and accuracy to quantify the effects of these uncertainties.

An extremum response surface method-based support vector machine (SVM-ERSM) was proposed based on SVM of regression to improve the computational efficiency and precision of blade radial deformation dynamic probabilistic design regarding non-linear material properties and dynamically thermal and mechanical loads.

Through the example calculation and comparison of methods, the results show that the blade radial deformation reaches at the maximum at t = 180 s; the probabilistic distribution and inverse probabilistic features of output parameters and the major factors (rotor speed and gas temperature) are gained; besides, the SVM-ERSM holds high computational efficiency and precision in the non-linear dynamic probabilistic design of aeroengine typical components.

The present efforts provide a method to design turbine besides other aeroengine components considering dynamic and non-linear factors base on probabilistic design for further research.

Moreover, the present study provides a way to design dynamic (motion) structures from a probabilistic perspective.

It is proved that the dynamic probabilistic design-based SVM-ERSM could produce a more reasonable blade radial deformation while maintaining low failure probability, as well as offer a useful reference for blade-tip clearance control and a promising insight to the optimal design of aeroengine typical components.

This paper aims to describe current trends in probabilistic structural fire engineering and provides a comprehensive summary of the state-of-the-art of performance-based structural fire engineering (PSFE).

PSFE has been introduced to overcome the limitations of current conventional design approaches used for the design of fire-exposed structures, which investigate assumed worst-case fire scenarios and include multiple thermal and structural analyses. PSFE permits buildings to be designed in relation to a level of life safety or economic loss that may occur in future fire events with the help of a probabilistic approach.

This paper brings together existing research on various sources of uncertainty in probabilistic structural fire engineering, such as elements affecting post-flashover fire development, material properties, fire models, fire severity, analysis methods and structural reliability.

Prediction of economic loss would depend on the extent of damage, which is further dependent on the structural response. The representative prediction of structural behaviour would depend on the precise quantification of the fire hazard. The incorporation of major uncertainty sources in probabilistic structural fire engineering is explained, and the detailed description of a pioneering analysis method called incremental fire analysis is presented.

The purpose of this paper is to estimate probabilistic and regional importance sensitivities of fatigue life, with respect to the laser peening (LP) parameters applied to a Titanium turbine disk.

The sensitivities were calculated from Monte Carlo (MC) analysis of 21,000 simulations and probabilistic sensitivity methods.

The probabilistic sensitivity results indicate that the peak pressure and the mid‐span are the most important variables. The regional importance sensitivity results indicate that probability of failure is the most sensitive to the left tail of peak pressure and middle region of mid‐span and the fatigue life mean is the most sensitive to the left tails of the peak pressure and the mid‐span.

The sensitivity results of this research indicate that more time and energy should be focused on managing peak pressure and mid‐span, as compared to the remaining variables, to design and improve the laser peening process.

The paper presents four sensitivity analysis approaches which were formulated and employed to estimate fatigue life sensitivities with respect to the LP variables.

Due to many different types of aleatory and epistemic uncertainty in soil properties, safety factor, which is assessed by deterministic analysis, is not reliable. The purpose of this paper is to determine the difference between critical slip surface in deterministic analysis and critical reliability slip surface in probabilistic analysis.

Deterministic analysis is formulated by the limit equilibrium methods, including Fellenius method, Bishop method, and Janbu’s simplified method. Then, the factor of safety is calculated for different slip surfaces. The stability of the soil is defined as the critical slip surface with the lowest factor of safety in each method. For probabilistic analysis, the value of reliability index, factor of safety, and probability of failure regarding given potential slip surface are considered as the stability index and obtained by the Monte Carlo simulation method.

To compare deterministic and probabilistic analysis as well as the influence of each of the aforementioned methods and stability index, a soil slope with three uncertainty parameters is analyzed and the results indicate that the critical slip surface is significantly different from critical reliability slip; however, the results from the above-mentioned methods are very close.

There are many other methods that could be studied; however, the most usual ones were employed. Furthermore, this study just consider the most important factors as the uncertainty parameters; nevertheless, it can be extended to more geotechnical parameters.

Although there are many studies in this field, the authors conduct a succinct but very noteworthy research to show the difference between the results of mentioned methods as well as deterministic and probabilistic approaches and their influence on slip surface.

This paper seeks to develop an approach to problem localization and an algorithm to address the issue of determining the dependencies among system metrics for automated system management in ubiquitous computing systems.

This paper proposes an approach to problem localization for learning the knowledge of dynamic environment using probabilistic dependency analysis to automatically determine problems. This approach is based on Bayesian learning to describe a system as a hierarchical dependency network, determining root causes of problems via inductive and deductive inferences on the network. An algorithm of preprocessing is performed to create ordering parameters that have close relationships with problems.

The findings show that using ordering parameters as input of network learning, it reduces learning time and maintains accuracy in diverse domains especially in the case of including large number of parameters, hence improving efficiency and accuracy of problem localization.

An evaluation of the work is presented through performance measurements. Various comparisons and evaluations prove that the proposed approach is effective on problem localization and it can achieve significant cost savings.

This study contributes to research into the application of probabilistic dependency analysis in localizing the root cause of problems and predicting potential problems at run time after probabilities propagation throughout a network, particularly in relation to fault management in self‐managing systems.

Kriging model has been widely adopted to reduce the high computational costs of simulations in Reliability-based design optimization (RBDO). To construct the Kriging model accurately and efficiently in the region of significance, a local sampling method with variable radius (LSVR) is proposed. The paper aims to discuss these issues.

In LSVR, the sequential sampling points are mainly selected within the local region around the current design point. The size of the local region is adaptively defined according to the target reliability and the nonlinearity of the probabilistic constraint. Every probabilistic constraint has its own local region instead of all constraints sharing one local region. In the local sampling region, the points located on the constraint boundary and the points with high uncertainty are considered simultaneously.

The computational capability of the proposed method is demonstrated using two mathematical problems, a reducer design and a box girder design of a super heavy machine tool. The comparison results show that the proposed method is very efficient and accurate.

The main contribution of this paper lies in: a new local sampling region computational criterion is proposed for Kriging. The originality of this paper is using expected feasible function (EFF) criterion and the shortest distance to the existing sample points instead of the other types of sequential sampling criterion to deal with the low efficiency problem.

In a conventional finite element analysis, material properties, dimensions and applied loads are usually defined as deterministic quantities. This simplifying assumption however, is not true in practical applications. Using statistics in engineering problems enables us to consider the effects of the input variables dispersion on the output parameters in an analysis. This provides a powerful tool for better decision making for more reliable design. In this paper, a probabilistic based design is presented which evaluates the sensitivity of a mechanical model to random input variables. To illustrate the effectiveness of this method, a simple bracket is analyzed for stress‐strain behavior using commercially available finite element software. Young’s modulus, applied pressure and dimensions are considered as random variables with Gaussian distribution and their effects on maximum stress and displacement is evaluated. The finite element results are compared with reliability based theoretical results which show very good agreement. This demonstrates the capability of commercially available software to handle probabilistic approach design.

The purpose of this paper is to present a fault tree (FT)-based approach for quantitative risk analysis in the construction industry that can take into account both objective and subjective uncertainties.

In this research, the identified basic events (BEs) are first categorized based on the availability of historical data into probabilistic and possibilistic. The probabilistic and possibilistic events are represented by probability distributions and fuzzy numbers, respectively. Hybrid uncertainty analysis is then performed through a combination of Monte Carlo simulation and fuzzy set theory. The probability of occurrence of the top event is finally calculated using the proposed FT-based hybrid uncertainty analysis method.

The efficiency of the proposed method is demonstrated by implementing in a real steel structure project. A quantitative risk assessment is performed for weld cracks, taking into account of both types of uncertainties. An importance analysis is finally performed to evaluate the contribution of each BE to the probability of occurrence of weld cracks and adopt appropriate response strategies.

In this research, the impact of objective (aleatory) dependence between the occurrences of different BEs and subjective (epistemic) dependence between estimates of the epistemically uncertain probabilities of some BEs are not considered. Moreover, there exist limitations to the application of fuzzy set rules, which were used for aggregating experts’ opinions and ranking purposes of the BEs in the FT model. These limitations can be investigated through further research.

It is believed that the proposed hybrid uncertainty analysis method presents a robust and powerful tool for quantitative risk analysis, as both types of uncertainties are taken into account appropriately.