Multiobjective Optimization Under Uncertainty in Advanced Abrasive Machining Processes via a Fuzzy-Evolutionary Approach

This paper considers multiobjective optimization under uncertainty (MOOUC) for the selection of optimal cutting conditions in advanced abrasive machining processes. Processes considered are water-jet machining, abrasive water-jet machining and ultra-sonic machining. Decisions regarding the cutting conditions can involve optimization for multiple competing goals; such as surface finish, machining time and power consumption. In practice, there is also an issue of variations in the ability to attain the performance goals. This can be due to limitations in machine accuracy or variations in material properties of the workpiece and/or abrasive particles. The approach adopted in this work relies on a Strength Pareto Evolutionary Algorithm (SPEA2) framework, with specially tailored dominance operators to account for probabilistic aspects in the considered multiobjective problem. Deterministic benchmark problems in the literature for the considered machining processes are extended to include performance uncertainty, and then used in testing the performance of the proposed approach. Results of the study show that accounting for process variations through a simple penalty term may be detrimental for the multiobjective optimization. On the other hand, a proposed Fuzzy-tournament dominance operator appears to produce favorable results.

Diagnostics and Prognostics of Lithium-Ion Batteries

This paper investigates recent research on battery diagnostics and prognostics especially for Lithium-ion (Li-ion) batteries. Battery diagnostics focuses on battery models and diagnosis algorithms for battery state of charge (SOC) and state of health (SOH) estimation. Battery prognostics elaborates data-driven prognosis algorithms for predicting the remaining useful life (RUL) of battery SOC and SOH. Readers will learn not only basics but also very recent research developments on battery diagnostics and prognostics.

Non-Negative Matrix Factorization Based Uncertainty Quantification Method for Complex Networked Systems

The behavior of large networked systems with underlying complex nonlinear dynamic are hard to predict. With increasing number of states, the problem becomes even harder. Quantifying uncertainty in such systems by conventional methods requires high computational time and the accuracy obtained in estimating the state variables can also be low. This paper presents a novel computational Uncertainty Quantifying (UQ) method for complex networked systems. Our approach is to represent the complex systems as networks (graphs) whose nodes represent the dynamical units, and whose links stand for the interactions between them. First, we apply Non-negative Matrix Factorization (NMF) based decomposition method to partition the domain of the dynamical system into clusters, such that the inter-cluster interaction is minimized and the intra-cluster interaction is maximized. The decomposition method takes into account the dynamics of individual nodes to perform system decomposition. Initial validation results on two well-known dynamical systems have been performed. The validation results show that uncertainty propagation error quantified by RMS errors obtained through our algorithms are competitive or often better, compared to existing methods.

Impact of Technology Infusion on System Complexity and Modularity

Complex systems often have long life cycles with requirements that are likely to change over time. Therefore, it is important to be able to adapt the system accordingly over time. This is often accomplished by infusing new technologies into the host system in order to update or improve overall system performance. However, technology infusion often results in a disruption in the host system. This can take the form of a system redesign or a change in the inherent attributes of the system. In this study, we analyzed the impact of technology infusion on system attributes, specifically the complexity and modularity. Two different systems that were infused with new technologies were analyzed for changes in complexity and modularity.

Predicting the Benefits of Topology Optimization

Topology optimization is a systematic method of generating designs that maximize specific objectives. While it offers significant benefits over traditional shape optimization, topology optimization can be computationally demanding and laborious. Even a simple 3D compliance optimization can take several hours. Further, the optimized topology must typically be manually interpreted and translated into a CAD-friendly and manufacturing friendly design. This poses a predicament: given an initial design, should one optimize its topology? In this paper, we propose a simple metric for predicting the benefits of topology optimization. The metric is derived by exploiting the concept of topological sensitivity, and is computed via a finite element swapping method. The efficacy of the metric is illustrated through numerical examples.

High Penetration Residential Solar Photovoltaics and the Effects of Dust Storms on System Net Load

Residential solar photovoltaic (PV) systems are becoming increasingly common around the world. Much of this growth is attributed to a decreasing cost of solar PV modules, reduction in the cost of installation and other “soft costs,” along with net-metering, financial incentives, and the growing societal interest in low-carbon energy. Yet this steep rise in distributed, uncontrolled solar PV capacity is being met with growing concern in maintaining electric grid stability when solar PV reaches higher penetration levels. Rapid reductions in solar PV output create an immediate and direct rise in the net system load. Demand response and storage technologies can offset these fluctuations in the net system load, but their potential has yet to be realized through wide-scale commercial dissemination. In the interim these fluctuations will continue to cause technical and economic challenges to the utility and the end-user. Late-afternoon peak demands are of particular concern as solar PV drops off and household demand rises as residents return home. Transient environmental factors such as clouding, rain, and dust storms pose additional uncertainties and challenges. This study analyzes such complex cases by simulating residential loads, rooftop solar PV output, and dust storm effects on solar PV output to examine transients in the net system load. The Phoenix, Arizona metropolitan area is used as a case study that experiences dust storms several times per year. A dust storm is simulated progressing over the Phoenix metro in various directions and intensities. Various solar PV penetration rates are also simulated to allow insight into resulting net loads as PV penetration grows in future years.

A Novel Application of Gamification for Collecting High-Level Design Information

This paper presents a novel application of gamification for collecting high-level design descriptions of objects. High-level design descriptions entail not only superficial characteristics of an object, but also function, behavior, and requirement information of the object. Such information is difficult to obtain with traditional data mining techniques. For acquisition of high-level design information, we investigated a multiplayer game, “Who is the Pretender?” in an offline context. Through a user study, we demonstrate that the game offers a more fun, enjoyable, and engaging experience for providing descriptions of objects than simply asking people to list them. We also show that the game elicits more high-level, problem-oriented requirement descriptions and less low-level, solution-oriented structure descriptions due to the unique game mechanics that encourage players to describe objects at an abstract level. Finally, we present how crowdsourcing can be used to generate game content that facilitates the gameplay. Our work contributes towards acquiring high-level design knowledge that is essential for developing knowledge-based CAD systems.

Wave Energy Converter Array Optimization: A Review of Current Work and Preliminary Results of a Genetic Algorithm Approach Introducing Cost Factors

Currently, ocean wave energy is a novel means of electricity generation that is projected to potentially serve as a primary energy source in coastal areas. However, for wave energy converters (WECs) to be applicable on a scale that allows for grid implementation, these devices will need to be placed in close relative proximity to each other. From what’s been learned in the wind industry of the U.S., the placement of these devices will require optimization considering both cost and power. However, current research regarding optimized WEC layouts only considers the power produced. This work explores the development of a genetic algorithm (GA) that will create optimized WEC layouts where the objective function considers both the economics involved in the array’s development as well as the power generated. The WEC optimization algorithm enables the user to either constrain the number of WECs to be included in the array, or allow the algorithm to define this number. To calculate the objective function, potential arrays are evaluated using cost information from Sandia National Labs Reference Model Project, and power development is calculated such that WEC interaction affects are considered. Results are presented for multiple test scenarios and are compared to previous literature, and the implications of a priori system optimization for offshore renewables are discussed.

Modeling Consumer Decisions on Returning End-of-Use Products Considering Design Features and Consumer Interactions: An Agent Based Simulation Approach

As electronic waste (e-waste) becomes one of the fastest growing environmental concerns, remanufacturing is considered as a promising solution. However, the profitability of take back systems is hampered by several factors including the lack of information on the quantity and timing of to-be-returned used products to a remanufacturing facility. Product design features, consumers’ awareness of recycling opportunities, socio-demographic information, peer pressure, and the tendency of customer to keep used items in storage are among contributing factors in increasing uncertainties in the waste stream. Predicting customer choice decisions on returning back used products, including both the time in which the customer will stop using the product and the end-of-use decisions (e.g. storage, resell, through away, and return to the waste stream) could help manufacturers have a better estimation of the return trend. The objective of this paper is to develop an Agent Based Simulation (ABS) model integrated with Discrete Choice Analysis (DCA) technique to predict consumer decisions on the End-of-Use (EOU) products. The proposed simulation tool aims at investigating the impact of design features, interaction among individual consumers and socio-demographic characteristics of end users on the number of returns. A numerical example of cellphone take-back system has been provided to show the application of the model.

Autonomous Microgrid Design Using Classifier-Guided Sampling

Identifying high-performance, system-level microgrid designs is a significant challenge due to the overwhelming array of possible configurations. Uncertainty relating to loads, utility outages, renewable generation, and fossil generator reliability further complicates this design problem. In this paper, the performance of a candidate microgrid design is assessed by running a discrete event simulation that includes extended, unplanned utility outages during which microgrid performance statistics are computed. Uncertainty is addressed by simulating long operating times and computing average performance over many stochastic outage scenarios. Classifier-guided sampling, a Bayesian classifier-based optimization algorithm for computationally expensive design problems, is used to search and identify configurations that result in reduced average load not served while not exceeding a predetermined microgrid construction cost. The city of Hoboken, NJ, which sustained a severe outage following Hurricane Sandy in October, 2012, is used as an example of a location in which a well-designed microgrid could be of great benefit during an extended, unplanned utility outage. The optimization results illuminate design trends and provide insights into the traits of high-performance configurations.