A Comprehensive Investigation of Ensembles of Gaussian-Based and Gradient-Based Transformed Reliability Analyses: When and How to Use Them

In the past few years, researchers have begun to investigate the existence of arbitrary uncertainties in the design optimization problems. Most traditional reliability-based design optimization (RBDO) methods transform the design space to the standard normal space for reliability analysis but may not work well when the random variables are arbitrarily distributed. It is because that the transformation to the standard normal space cannot be determined or the distribution type is unknown. The methods of Ensemble of Gaussian-based Reliability Analyses (EoGRA) and Ensemble of Gradient-based Transformed Reliability Analyses (EGTRA) have been developed to estimate the joint probability density function using the ensemble of kernel functions. EoGRA performs a series of Gaussian-based kernel reliability analyses and merged them together to compute the reliability of the design point. EGTRA transforms the design space to the single-variate design space toward the constraint gradient, where the kernel reliability analyses become much less costly. In this paper, a series of comprehensive investigations were performed to study the similarities and differences between EoGRA and EGTRA. The results showed that EGTRA performs accurate and effective reliability analyses for both linear and nonlinear problems. When the constraints are highly nonlinear, EGTRA may have little problem but still can be effective in terms of starting from deterministic optimal points. On the other hands, the sensitivity analyses of EoGRA may be ineffective when the random distribution is completely inside the feasible space or infeasible space. However, EoGRA can find acceptable design points when starting from deterministic optimal points. Moreover, EoGRA is capable of delivering estimated failure probability of each constraint during the optimization processes, which may be convenient for some applications.

Design of Complex Engineering Systems Using Multiagent Coordination

In complex engineering systems, complexity may arise by design, or as a by-product of the system’s operation. In either case, the root cause of complexity is the same: the unpredictable manner in which interactions among components modify system behavior. Traditionally, two different approaches are used to handle such complexity: (i) a centralized design approach where the impacts of all potential system states and behaviors resulting from design decisions must be accurately modeled; and (ii) an approach based on externally legislating design decisions, which avoid such difficulties, but at the cost of expensive external mechanisms to determine trade-offs among competing design decisions. Our approach is a hybrid of the two approaches, providing a method in which decisions can be reconciled without the need for either detailed interaction models or external mechanisms. A key insight of this approach is that complex system design, undertaken with respect to a variety of design objectives, is fundamentally similar to the multiagent coordination problem, where component decisions and their interactions lead to global behavior. The design of a race car is used as the case study. The results of this paper demonstrate that a team of autonomous agents using a cooperative coevolutionary algorithm can effectively design a Formula racing vehicle.

An Extended Pattern Search Method for Offshore Floating Wind Layout and Turbine Geometry Optimization

An Extended Pattern Search (EPS) method is developed to optimize the layout and turbine geometry for offshore floating wind power systems. The EPS combines a deterministic pattern search with stochastic extensions. Three advanced models are incorporated: (1) a cost model considering investment and lifetime costs of a floating offshore wind farm comprised of WindFloat platforms; (2) a wake propagation and interaction model able to determine the reduced wind speeds downstream of rotating blades; and (3) a power model to determine power produced at each rotor, and includes a semi-continuous, discrete turbine geometry selection to optimize the rotor radius and hub height of individual turbines. The objective function maximizes profit by minimizing cost, minimizing wake interactions, and maximizing power production. A multidirectional, multiple wind speed case is modeled which is representative of real wind site conditions. Layouts are optimized within a square solution space for optimal positioning and turbine geometry for farms containing a varying number of turbines. Resulting layouts are presented; optimized layouts are biased towards dominant wind directions. Preliminary results will inform developers of best practices to include in the design and installation of offshore floating wind farms, and of the resulting cost and power production of wind farms that are computationally optimized for realistic wind conditions.

A Method for Classifying Products Designed for the Developing World

Researchers in the area of design for the developing world have synthesized knowledge from location-specific product case studies in the form of design guidance, which includes pitfalls, principles, and methods. Much of the design guidance relates to specific product classes and regions, while recent work is directed towards generalized principles. The aim of this paper is to fill gaps in product class-specific design guidance by creating larger groups of similar products, which share design characteristics. In this paper, we present a method for classifying products into such groups utilizing cluster analysis. We present a five-step method, which includes optional synthesis of design principles. The potential value of the method is demonstrated in a case study. The result included two distinct product groups, titled Products for Relief and Products for Development, and corresponding design principles for each group.

Architecting Fail-Safe Supply Networks

A fail-safe network is one that mitigates the impact of different uncertainty sources and provides the most profitable level of service. This is achieved by having 1) a structurally fail-safe topology against rare but high magnitude stochastic events called disruptions and 2) an operationally fail-safe flow dynamic against frequent but low magnitude stochastic events called variations. A structurally fail-safe network should be robust and resilient against disruptions. Robustness and resilience respectively determine how well and how quickly disruptions are handled by the SN. Flow planning must be reliable in an operationally fail-safe supply network against variations to provide the most profitable service level to customers. We formulate the problem of designing/redesigning fail-safe supply networks as a compromise Decision Support Problem. We analyze the correlations among robustness, resilience, and profit for supply networks and propose a method for supply network managers to use when they need to find a compromise among robustness, resilience, and profit.

Cyber-Empathic Design: A Data Driven Framework for Product Design

One of the critical tasks in product design is to map information from the consumer space to the design space. Currently, this process is largely dependent on the designer to identify and map how psychological and consumer level factors relate to engineered product attributes. In this way current methodologies lack provision to test a designer’s cognitive reasoning and could therefore introduce bias while mapping from consumer to design space. Also, current dominant frameworks do not include user-product interaction data in design decision making and neither do they assist designers in understanding why a consumer has a particular perception about a product. This paper proposes a new framework — Cyber-Empathic Design — where user-product interaction data is acquired via embedded sensors in the products. To understand the motivations behind consumer perceptions, a network of latent constructs is used which forms a causal model framework. Structural Equation Modeling is used as the parameter estimation and hypothesis testing technique making the framework falsifiable in nature. To demonstrate the framework and demonstrate its effectiveness a case study of sensor integrated shoes is presented in this work, where two models are compared — one survey based and using the Cyber-Empathic framework model. It is shown that the Cyber-Empathic framework results in improved fit. The case study also demonstrates the technique to test a designers’ cognitive hypothesis.

Deriving Infill Design of Fused Deposition Modelled Parts From Predicted Stress Profiles

The use of Fused Deposition Modelling (FDM) is increasing rapidly in both the commercial and industrial sectors as a means of rapidly prototyping geometrically complex parts. Particular affordances of FDM include the reduction of waste material during manufacture, the use of multiple materials within a single manufacturing process and the ability to manipulate the internal geometry of a part. The latter of which has seen the generation of many 2-dimensional repeating pattern structures such as square, rectilinear and hexagonal, as well as an emerging field of 3-dimensional structures. Although these patterns have provided stiffness and rigidity whilst reducing the production time of FDM prototypes, many do not consider the actual loading conditions of the part in-situ, where it is argued that further significant gains in the performance could be achieved. This includes further reduction in process time and increased part functionality. Thus, this paper presents initial work into the generation of an infill that is derived from the predicted stress profile for the part. This has been achieved through the post-processing of Finite Element (FE) models to identify the stress profile. Interpolation across these profiles leads to a set of aligned Bézier splines that enable the transmission of force and are also able to be manufactured using FDM. These splines are embedded within the typical slicing procedure of a part ahead of being manufactured on a FDM machine. Initial results from parts designed to support three-point bending loads show a 79% increase in the stiffness of the part alongside a consistent and repeatable mode of failure when compared to the commonly used honeycomb infill design.

Multi-Objective Optimal Design of Functionally Gradient Materials

The objective of this research is to optimally design both the topology and material distribution of functionally gradient material objects while considering more than one objective. Many techniques exist for both topology optimization and optimal placement of functionally gradient material within a single object, but combining the two is challenging. In addition, gradient materials allow customization of individual regions of a single part in order to achieve conflicting objectives or constraints. This paper shows a technique for concurrent topology and material gradient optimization within a single part while considering two conflicting objectives. The algorithm is applied to a standard topology optimization problem. The resulting gradient material designs have regions with distinct functionality and the material in these regions is chosen based on the regions function. In addition, a comparison of the gradient material design and a corresponding homogenous material design shows a significant improvement in the objective value for the gradient material design.

Topology Optimization Under Independent Multi-Load With Uncertainty

In this paper, topology optimization under multiple independent loadings with uncertainty is presented. In engineering practice, load uncertainty can be found in many applications. From the literature, researchers have focused mainly on problems containing only a single uncertain external load. However, such idealistic problems may not be very useful in engineering practice. Problems involving multi-loadings with uncertainty are more commonly found in engineering applications. This paper presents a method to solve a system which contains multiple independent loadings with load uncertainty. First, a two-level optimization problem is formulated. The upper level problem is a typical topology optimization problem to minimize the mean compliance in the design using the worst case conditions. The lower level optimization problem is to solve for the worst loadings corresponding to the critical structure response. At the lower level formulation, an unknown-but-bounded model is used to define uncertain loadings. There are two challenges in finding the worst loading case: non-convexity and multi-loadings. The non-convexity problem is addressed by reformulating the problem as an inhomogeneous eigenvalue problem by applying the KKT optimality conditions and the multi-uncertain loadings problem is solved by an iterative method. After the worst loadings are generated, the upper level problem can be solved by a general topology optimization method. The effectiveness of the proposed method is demonstrated by numerical examples.

A Cluster Analysis Study of Opportune Adoption of Electric Drive Vehicles for Better Greenhouse Gas Reduction

Adoption of electric drive vehicles (EDVs) presents an opportunity for reduction of greenhouse gas (GHG) emissions. From an individual vehicle standpoint however, the GHG reduction can vary significantly depending on the type of driving that the vehicle is used for. This is primarily due to conventional vehicles (CVs) having poor energy efficiency in stop-and-go city-like driving compared to their performance in steady highway-like driving. This study attempts to examine the magnitude of the differential in GHG reduction benefit for real driving behaviors obtained from California Household Travel Survey (CHTS-2013). Recorded vehicles speed traces are analyzed via a fuel economy simulator then a hybrid support vector clustering (SVC) technique is applied to form groups of vehicle samples with similar driving behaviors. Unlike many clustering techniques, SVC does not impose a pre-dictated number of clusters, but has a number of parameters that must be tuned in order to obtain meaningful results. Tuning of the parameters is performed via a multi-objective evolutionary algorithm (SPEA2) after formulating the cluster tuning as a two-objective problem that seeks to maximize: i) differential benefit in GHG reduction, and ii) fraction of the population that groups of vehicles represent. Results show that replacing a CV with its equivalent hybrid (HEV) can reduce GHG emissions per mile of driving by 2 to 2.5 times more for a group of vehicles (best opportune for an EDV) compared to the less opportune group.