Random matrix theory for modeling uncertainties in computational mechanics

This paper deals with data uncertainties and model uncertainties issues in computational mechanics. If data uncertainties can be modeled by parametric probabilistic methods, for a given mean model, a nonparametric probabilistic approach can be used for modeling model uncertainties. The first part is devoted to random matrix theory for which we summarize previous published results and for which two new ensembles of random matrices useful for the nonparametric models are introduced. In a second part, the nonparametric probabilistic approach of random uncertainties is presented for linear dynamical systems and for nonlinear dynamical systems constituted of a linear part with additional localized nonlinearities. In a third part, a new method is proposed for estimating the parameters of the nonparametric approach from experiments. Finally, examples with experimental comparisons are given.     

Experimental identification of an uncertain computational dynamical model representing a family of structures

We are interested in constructing an uncertain computational model representing a family of structures and in identifying this model using a small number of experimental measurements of the first eigenfrequencies. The prior probability model of uncertainties is constructed using the generalized probabilistic approach of uncertainties which allows both system-parameters uncertainties and model uncertainties to be taken into account. The parameters of the prior probability model of uncertainties are separately identified for each type of uncertainties, yielding an optimal prior probability model. The optimal prior stochastic computational model allows a robust analysis for the family of structures to be carried out.     

Robust Pareto active suspension design for vehicle vibration model with probabilistic uncertain parameters

Using the robust design of a vehicle vibration model considering uncertainties can elaborately show the effects of those unsure values on the performance of such a model. In this paper, probabilistic metrics, instead of deterministic metrics, are used for a robust Pareto multi-objective optimum design of five-degree of freedom vehicle vibration model having parameters with probabilistic uncertainties. In order to achieve an optimum robust design against probabilistic uncertainties existing in reality, a multi-objective uniform-diversity genetic algorithm (MUGA) in conjunction with Monte Carlo simulation is used for Pareto optimum robust design of a vehicle vibration model with ten conflicting objective functions. The robustness of the design obtained using such a probabilistic approach is shown and compared with that of the design obtained using deterministic approach.     

Multi-objective genetic programming approach for robust modeling of complex manufacturing processes having probabilistic uncertainty in experimental data

In this paper, a multi-objective uniform-diversity genetic programming (MUGP) algorithm deployed for robust Pareto modeling and prediction of complex nonlinear processes using some input-output data table. The uncertainties included in measured data are considered to obtain more robust models. The considered benchmarks are an explosive cutting and forming processes, in which the nonlinear behavior between the input and output of processes are detected using MUGP. For both case studies, a multi-objective modeling and prediction procedure firstly performed using deterministic data. Secondly, the same identification procedure carried out using probabilistic uncertainty in the experimental input-output data. The objective functions considered are namely, training error, prediction error and number of tree nodes (complexity of models) in the deterministic approach. Accordingly, the mean and standard deviation of training error and prediction error are considered in robust Pareto modeling and prediction of such processes. In this way, Pareto front of such modeling and prediction is first obtained for both explosive cutting and forming processes with deterministic data. Such Pareto front is then obtained using experimental input-output-data having probabilistic uncertainty in input parameters through a Monte Carlo simulation (MCS) approach. In addition, it has been shown that for both cases, the trade-off models obtained from deterministic data have significant biases when tested on data with probabilistic uncertainty. Finally, the obtained results of such multi-objective robust model identification show promising results in terms of compensating uncertainty in the experimental input-output-data.     

A procedure of parameter inversion for a nonlinear constitutive model of soils with shield tunneling

The identification of the parameters of a nonlinear constitutive model of soil mass is based on an inverse analysis procedure, which consists of minimizing the objective function representing the difference between the experimental data and the calculated data of the mechanical model. A gradient-based optimization algorithm is developed for estimating model parameters of soils in earth pressure balance (EPB) shield tunneling. The parameter values of the nonlinear constitutive model are searched for by using the Levenberg–Marquardt approximation which can provide fast convergence. The parameter identification results illustrate that the proposed parameter inversion procedure has not only higher computing efficiency but also better identification accuracy. The results from the model are compared with simulated observations. The models are found to have good predictive ability and are expected to be very useful for estimating model parameters for soils in EPB shield tunneling.     

Flexural wave propagation in fluid-conveying carbon nanotubes with system uncertainties

Many studies have reported that the material properties of carbon nanotubes (CNTs) show a wide range and exhibit a great of uncertainties. The uncertainty, in turn, will affect the physical behaviors of CNTs. In this paper, an iterative algorithm-based interval analysis method is proposed to deal with the flexural wave propagation characteristics of fluid-conveying CNTs with system uncertainties. To make the conclusion more objective, the properties of the material and fluid are all considered as uncertain-but-bounded parameters, which can effectively describe the uncertainties where few data are available to perform the probabilistic analysis. The upper and lower bounds of the wave dispersion curves are predicted to clarify the influences of the uncertain material and fluid properties on the wave propagation behaviors of fluid-conveying CNTs. It is demonstrated that the widths of the wave frequency and phase velocity behave different at different wavelengths. Besides, the bounds predicted by the probabilistic model are given to verify the present model, and the present model is also validated by comparing with the Monte Carlo simulation. The present model provides some useful guides for using CNTs to convey fluid flows.     

An evolutionary Bayesian belief network methodology for optimum management of groundwater contamination

An integrated methodology, based on Bayesian belief network (BBN) and evolutionary multi-objective optimization (EMO), is proposed for combining available evidence to help water managers evaluate implications, including costs and benefits of alternative actions, and suggest best decision pathways under uncertainty. A Bayesian belief network is a probabilistic graphical model that represents a set of variables and their probabilistic relationships, which also captures historical information about these dependencies. In complex applications where the task of defining the network could be difficult, the proposed methodology can be used in validation of the network structure and the parameters of the probabilistic relationship. Furthermore, in decision problems where it is difficult to choose appropriate combinations of interventions, the states of key variables under the full range of management options cannot be analyzed using a Bayesian belief network alone as a decision support tool. The proposed optimization method is used to deal with complexity in learning about actions and probabilities and also to perform inference. The optimization algorithm generates the state variable values which are fed into the Bayesian belief network. It is possible then to calculate the probabilities for all nodes in the network (belief propagation). Once the probabilities of all the linked nodes have been updated, the objective function values are returned to the optimization tool and the process is repeated. The proposed integrated methodology can help in dealing with uncertainties in decision making pertaining to human behavior. It also eliminates the shortcoming of Bayesian belief networks in introducing boundary constraints on probability of state values of the variables. The effectiveness of the proposed methodology is examined in optimum management of groundwater contamination risks for a well field capture zone outside Copenhagen city.     

Robust identification method for nonlinear model structures and its application to high-performance aircraft

A common assumption is that the model structure is known for modelling high performance aircraft. In practice, this is not the case. Actually, structure identification plays the most important role in the processing of nonlinear system modelling. The integration of mode structure identification and parameter estimation is an efficient method to construct the model for high performance aircraft, which is nonlinear and also contains uncertainties. This article presents an efficient method for identifying nonlinear model structure and estimating parameters for high-performance aircraft model, which contains uncertainties. The parameters associated with nonlinear terms are considered one after the other if they should be included in the nonlinear model until a stopping criterion is met, which is based on Akaike's information criterion. A numerically efficient U-D factorisation is presented to avoid complex computation of high-order matrices. The proposed method is applied to flight test data of a high-performance aircraft. The results demonstrate that the proposed method could obtain the good aircraft model with a reasonably good fidelity based on the comparison with flight test data.     

Improved hybrid method as a robust tool for reliability-based design optimization

In the engineering problems, the randomness and the uncertainties of the distribution of the structural parameters are a crucial problem. In the case of reliability-based design optimization (RBDO), it is the objective to play a dominant role in the structural optimization problem introducing the reliability concept. The RBDO problem is often formulated as a minimization of the initial structural cost under constraints imposed on the values of elemental reliability indices corresponding to various limit states. The classical RBDO leads to high computing time and weak convergence, but a Hybrid Method (HM) has been proposed to overcome these two drawbacks. As the hybrid method successfully reduces the computing time, we can increase the number of variables by introducing the standard deviations as optimization variables to minimize the error values in the probabilistic model. The efficiency of the hybrid method has been demonstrated on static and dynamic cases with extension to the variability of the probabilistic model. In this paper, we propose a modification on the formulation of the hybrid method to improve the optimal solutions. The proposed method is called, Improved Hybrid Method (IHM). The main benefit of this method is to improve the structure performance by much more minimizing the objective function than the hybrid method. It is also shown to demonstrate the optimality conditions. The improved hybrid method is next applied to two numerical examples, with consideration of the standard deviations as optimization variables (for linear and nonlinear distributions). When integrating the improved hybrid method within the probabilistic model variability, we minimize the objective function more and more.     

Reliability-based structural optimization with probability and convex set hybrid models

For structural systems exhibiting both probabilistic and bounded uncertainties, it may be suitable to describe these uncertainties with probability and convex set models respectively in the design optimization problem. Based on the probabilistic and multi-ellipsoid convex set hybrid model, this paper presents a mathematical definition of reliability index for measuring the safety of structures in presence of parameter or load uncertainties. The optimization problem incorporating such reliability constraints is then mathematically formulated. By using the performance measure approach, the optimization problem is reformulated into a more tractable one. Moreover, the nested double-loop optimization problem is transformed into an approximate single-loop minimization problem by considering the optimality conditions and linearization of the limit-state function, which further facilitates efficient solution of the design problem. Numerical examples demonstrate the validity of the proposed formulation as well as the efficiency of the presented numerical techniques.     

Evaluating probabilistic models with uncertain model parameters

Probabilistic models are commonly used to evaluate quality attributes, such as reliability, availability, safety and performance of software-intensive systems. The accuracy of the evaluation results depends on a number of system properties which have to be estimated, such as environmental factors or system usage. Researchers have tackled this problem by including uncertainties in the probabilistic models and solving them analytically or with simulations. The input parameters are commonly assumed to be normally distributed. Accordingly, reporting the mean and variances of the resulting attributes is usually considered sufficient. However, many of the uncertain factors do not follow normal distributions, and analytical methods to derive objective uncertainties become impractical with increasing complexity of the probabilistic models. In this work, we introduce a simulation-based approach which uses Discrete Time Markov Chains and probabilistic model checking to accommodate a diverse set of parameter range distributions. The number of simulation runs automatically regulates to the desired significance level and reports the desired percentiles of the values which ultimately characterises a specific quality attribute of the system. We include a case study which illustrates the flexibility of this approach using the evaluation of several probabilistic properties.     

A nonparametric approach to design robust controllers for uncertain systems: Application to an air flow heating system

This paper presents an approach to design robust fixed structure controllers for uncertain systems using a finite set of measurements in the frequency domain. In traditional control system design, usually, based on measurements, a model of the plant, which is only an approximation of the physical system, is first built, and then control approaches are used to design a controller based on the identified model. Errors associated with the identification process as well as the inevitable uncertainties associated with plant parameter variations, external disturbances, measurement noise, etc. are expected to all contribute to the degradation of the performance of such a scheme. In this paper, we propose a nonparametric method that uses frequency-domain data to directly design a robust controller, for a class of uncertainties, without the need for model identification. The proposed technique, which is based on interval analysis, allows us to take into account the plant uncertainties during the controller synthesis itself. The technique relies on computing the controller parameters for which the set of all possible frequency responses of the closed-loop system are included in the envelope of a desired frequency response. Such an inclusion problem can be solved using interval techniques. The main advantages of the proposed approach are: (1) the control design does not require any mathematical model, (2) the controller is robust with respect to plant uncertainties, and (3) the controller structure can be chosen a priori, which allows us to select low-order controllers. To illustrate the proposed method and demonstrate its efficacy, an application to an air flow heating system is presented.     

Structural anti-optimization with interval design parameters

The anti-optimization problem of structures with uncertain design variables is studied by combing the conventional optimization and interval analysis. The uncertain design parameters, which usually exist in the object function and constraint conditions, are modeled as interval sets. The proposed method can endure the variation of structural performance resulting from the variation of uncertain design parameters. According to the variation range of them, the range or interval of the optimal objective function and the optimal solution can be determined. In this sense, the optimal solution is one domain rather than a point. Numerical examples are used to illustrate the feasibility and superiority of the non-probabilistic optimization method in comparison with the conventional and probabilistic optimization methods.     

Inverse propagation of uncertainties in finite element model updating through use of fuzzy arithmetic

A fuzzy finite element model updating (FFEMU) method is presented in this study for the damage detection problem. The uncertainty caused by the measurement noise in modal parameters is described by fuzzy numbers. Inverse analysis is formulated as a constrained optimization problem at each α-cut level. Membership functions of each updating parameter which correspond to reduction in bending stiffness of the finite elements is determined by minimizing an objective function using a hybrid version of genetic algorithms (GA) and particle swarm optimization method (PSO) which is very efficient in terms of accuracy and robustness. Practical evaluation of the approximate bounds of the interval modal parameters in FFEMU iterations is addressed. A probabilistic analysis is performed using Monte Carlo simulation (MCS) and the results are compared with presented FFEMU method. It is apparent from numerical simulations that the proposed method is well capable in finding the membership functions of the updating parameters within reasonable accuracy. It is also shown that the results obtained by FFEMU are in good agreement with the MCS results while FFEMU is not as computationally expensive as the MCS method. Nevertheless, the proposed FFEMU do not required derivatives of the objective function like existing methods except in the deterministic case.     

Fault detection for uncertain LPV systems using probabilistic set-membership parity relation

This paper considers fault detection of uncertain linear parameter varying systems that have polynomial dependence on parametric uncertainties. A conventional set-membership (SM) approach is able to ensure zero false alarm rate (FAR) by using conservative threshold sets, but usually results in a high missed detection rate (MDR) due to equally treating all uncertainty realizations without distinguishing between high and low probability of occurrence. To address this limitation, a probabilistic SM parity relation approach is proposed to exploit probabilistic information on the parametric uncertainties, which results in a reduced MDR by admitting an acceptable FAR. The parity relation is first polynomially parameterized with respect to uncertain parameters. Then, Gaussian mixtures are adopted to efficiently compute uncertainty propagation from stochastic uncertainties to the residual distribution. To achieve an acceptable FAR, a non-convex confidence set of residuals – represented by a union of ellipsoids – is determined for the consistency test. The effectiveness of the proposed approach is illustrated using a continuous stirred tank reactor example including performance comparisons with a deterministic zonotope-based method.     

Probabilistic dual heuristic programming-based adaptive critic

Adaptive critic (AC) methods have common roots as generalisations of dynamic programming for neural reinforcement learning approaches. Since they approximate the dynamic programming solutions, they are potentially suitable for learning in noisy, non-linear and non-stationary environments. In this study, a novel probabilistic dual heuristic programming (DHP)-based AC controller is proposed. Distinct to current approaches, the proposed probabilistic (DHP) AC method takes uncertainties of forward model and inverse controller into consideration. Therefore, it is suitable for deterministic and stochastic control problems characterised by functional uncertainty. Theoretical development of the proposed method is validated by analytically evaluating the correct value of the cost function which satisfies the Bellman equation in a linear quadratic control problem. The target value of the probabilistic critic network is then calculated and shown to be equal to the analytically derived correct value. Full derivation of the Riccati solution for this non-standard stochastic linear quadratic control problem is also provided. Moreover, the performance of the proposed probabilistic controller is demonstrated on linear and non-linear control examples.     

An extended probabilistic method for reliability analysis under mixed aleatory and epistemic uncertainties with flexible intervals

The reliability analysis approach based on combined probability and evidence theory is studied in this paper to address the reliability analysis problem involving both aleatory uncertainties and epistemic uncertainties with flexible intervals (the interval bounds are either fixed or variable as functions of other independent variables). In the standard mathematical formulation of reliability analysis under mixed uncertainties with combined probability and evidence theory, the key is to calculate the failure probability of the upper and lower limits of the system response function as the epistemic uncertainties vary in each focal element. Based on measure theory, in this paper it is proved that the aforementioned upper and lower limits of the system response function are measurable under certain circumstances (the system response function is continuous and the flexible interval bounds satisfy certain conditions), which accordingly can be treated as random variables. Thus the reliability analysis of the system response under mixed uncertainties can be directly treated as probability calculation problems and solved by existing well-developed and efficient probabilistic methods. In this paper the popular probabilistic reliability analysis method FORM (First Order Reliability Method) is taken as an example to illustrate how to extend it to solve the reliability analysis problem in the mixed uncertainty situation. The efficacy of the proposed method is demonstrated with two numerical examples and one practical satellite conceptual design problem.     

UMOTS: an uncertainty-aware multi-objective genetic algorithm-based static task scheduling for heterogeneous embedded systems

Increasing manufacturing process variations due to aggressive technology scaling in addition to heterogeneity in design components are expected to cause serious challenges for future embedded system design steps including task scheduling. Process variation effects along with increased complexity in embedded applications result in design uncertainties, which in turn, reduce the accuracy and efficiency of traditional design approaches with deterministic values for the design component parameters. In this paper, a multi-objective task scheduling framework is proposed for embedded systems considering uncertainties in both hardware and software component parameters. The tasks which are modeled as a task graph are scheduled on a specific hardware platform consisting of processors and communication parts. Uncertainty is considered in both software (task parameters) and hardware (processor and communication parameters) of the embedded system. UMOTS takes advantages of a Monte-Carlo-based approach within a multi-objective genetic algorithm to handle the uncertainties in model parameters. The proposed approach finds the Pareto frontier, which is robust against uncertainties, in the objective space formed by performance, energy consumption, and reliability. The efficiency of UMOTS is investigated in the experimental results using real-application task graphs. In terms of Scheduling Length Ratio (SLR) and speedup, UMOTS provides 27.8% and 28.6% performance improvements in comparison to HSHD, one state-of-the-art task scheduling algorithm. Additionally, UMOTS, which is based on a multi-objective genetic optimization algorithms, finds robust Pareto frontier with 1%, 5% and 10% uncertainty in design indicators with respect to design limitations.