Engineering Papers⌕ Search

Engineering topics

Kamath, C.

Publications and source records attributed to Kamath, C..

Determination of Process Parameters for High-Density, Ti-6Al-4V Parts Using Additive Manufacturing

In our earlier work, we described an approach for determining the process parameters that re- sult in high-density parts manufactured using the additive-manufacturing process of selective laser melting (SLM). Our approach, which combines simple simulations and experiments, was demon- strated using 316L stainless steel. We have also used the approach successfully for several other materials. This short note summarizes the results of our work in determining process parameters for Ti-6Al-4V using a Concept Laser M2 system.

36 MATERIALS SCIENCE↗

Regression with Small Data Sets: A Case Study using Code Surrogates in Additive Manufacturing

There has been an increasing interest in recent years in the mining of massive data sets whose sizes are measured in terabytes. While it is easy to collect such large data sets in some application domains, there are others where collecting even a single data point can be very expensive, so the resulting data sets have only tens or hundreds of samples. For example, when complex computer simulations are used to understand a scientific phenomenon, we want to run the simulation for many different values of the input parameters and analyze the resulting output. The data set relating the simulation inputs and outputs is typically quite small, especially when each run of the simulation is expensive. However, regression techniques can still be used on such data sets to build an inexpensive \surrogate" that could provide an approximate output for a given set of inputs. A good surrogate can be very useful in sensitivity analysis, uncertainty analysis, and in designing experiments. In this paper, we compare different regression techniques to determine how well they predict melt-pool characteristics in the problem domain of additive manufacturing. Our analysis indicates that some of the commonly used regression methods do perform quite well even on small data sets.

97 MATHEMATICS AND COMPUTING↗

Laser powder bed fusion additive manufacturing of metals; physics, computational, and materials challenges

The production of metal parts via laser powder bed fusion additive manufacturing is growing exponentially. However, the transition of this technology from production of prototypes to production of critical parts is hindered by a lack of confidence in the quality of the part. Confidence can be established via a fundamental understanding of the physics of the process. It is generally accepted that this understanding will be increasingly achieved through modeling and simulation. However, there are significant physics, computational, and materials challenges stemming from the broad range of length and time scales and temperature ranges associated with the process. In this study, we review the current state of the art and describe the challenges that need to be met to achieve the desired fundamental understanding of the physics of the process.

36 MATERIALS SCIENCE↗

Overview of modelling and simulation of metal powder bed fusion process at Lawrence Livermore National Laboratory

The metal laser powder bed fusion additive manufacturing process uses high power lasers to build parts layer upon layer by melting fine metal powders. Qualification of parts produced using this technology is broadly recognised as a significant challenge. Physics based process models have been identified as being foundational to qualification of additively manufactured metal parts. In the present article, a multiscale modelling strategy is described that will serve as the foundation upon which process control and part qualification can be built. This includes a model at the scale of the powder that simulates single track/single multilayer builds and provides powder bed and melt pool thermal data. A second model computationally builds a complete part and predicts manufactured properties (residual stress, dimensional accuracy) in three dimensions. Finally, modelling is tied to experiment through data mining.

36 MATERIALS SCIENCE↗

MINDES - Data Mining for Inverse Design (Final Report)

This is the final summary report on the work done as part of the ARRA-funded, MINDES: Data Mining for Inverse Design SciDAC-e project. The goal of this project was to apply data mining techniques to data generated by the Center for Inverse Design, an Energy Frontier Research Center (EFRC) of the Office of Science, US Department of Energy. This center is pursuing a new approach to material science; rather than using the conventional direct approach (“Given the structure, find the electronic properties”), they are using a “materials by inverse design” approach (“Given the desired property, find the structure”). The specific target properties of interest include general semiconductor optical and electrical properties.

97 MATHEMATICS AND COMPUTING↗

Towards Detecting Motifs in Time Series Data of Wind Energy

The sporadic behavior of wind currently hampers electrical companies from fully utilizing wind energy as a viable resource. Observations of reported wind generation data suggest that there may be diurnal patterns in wind generation. Predicting the magnitude and graphical shape of wind generation at different times during the day would help control room operators to better direct the assignment of power resources throughout the day. We will discuss the use of different methods of pattern detection and analysis to analyze time series data of wind generation from the wind farms in the mid-Columbia River Basin which provide power to the Bonneville Power Administration.

17 WIND ENERGY↗

WindSENSE Project Summary: FY2009-2011

Renewable resources, such as wind and solar, are providing an increasingly larger percentage of our energy needs. To successfully integrate these intermittent resources into the power grid while maintaining its reliability, we need to better understand the characteristics and predictability of the variability associated with these power generation resources. WindSENSE, a three year project at Lawrence Livermore National Laboratory, considered the problem of scheduling wind energy on the grid from the viewpoint of the control room operator. Our interviews with operators at Bonneville Power Administration (BPA), Southern California Edison (SCE), and California Independent System Operator (CaISO), indicated several challenges to integrating wind power generation into the grid. As the percentage of installed wind power has increased, the variable nature of the generation has become a problem. For example, in the Bonneville Power Administration (BPA) balancing area, the installed wind capacity has increased from 700 MW in 2006-2007 to over 1300 MW in 2008 and more than 2600 MW in 2009. To determine the amount of energy to schedule for the hours ahead, operators typically use 0-6 hour ahead forecasts, along with the actual generation in the previous hours and days. These forecasts are obtained from numerical weather prediction (NWP) simulations or based on recent trends in wind speed in the vicinity of the wind farms. However, as the wind speed can be difficult to predict, especially in a region with complex terrain, the forecasts can be inaccurate. Complicating matters are ramp events, where the generation suddenly increases or decreases by a large amount in a short time (Figure 1, right panel). These events are challenging to predict, and given their short duration, make it difficult to keep the load and the generation balanced. Our conversations with BPA, SCE, and CaISO indicated that control room operators would like (1) more accurate wind power generation forecasts for use in scheduling and (2) additional information that can be exploited when the forecasts do not match the actual generation. To achieve this, WindSENSE had two areas of focus: (1) analysis of historical data for better insights, and (2) observation targeting for improved forecasts. The goal was to provide control room operators with an awareness of wind conditions and energy forecasts so they can make well-informed scheduling decisions, especially in the case of extreme events such as ramps.

17 WIND ENERGY↗

Implementation of two projection methods on a shared memory multiprocessor - DEC VAX 6240

The relative performance of two iterative schemes, based on projection techniques, is compared on a shared memory multiprocessor - VAX 6240. The CG accelerated Block-SSOR method and the CG accelerated Symmetric-Kaczmarz method are considered for the solution of large sparse nonsymmetric systems of linear equations. It is shown that the regular structure of many matrices can be exploited by the CG-accelerated Block-SSOR method to provide good speedup in a multiprocessing environment. However, the CG accelerated Symmetric-Kaczmarz method, while being a viable alternative on a scalar machine, is unable to benefit from multiprocessing.

Kamath, C.↗