Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “loading pattern optimization”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

112 records · Page 7
All sourcesNASA NTRSDOE OSTI
DOE OSTI2018

Prefabricated High-Strength Rebar Systems with High-Performance Concrete for Accelerated Construction of Nuclear Concrete Structures

This project aimed to reduce the field construction times and fabrication costs of reinforced concrete (RC) nuclear structures through the use of: 1) high-strength steel deformed reinforcing bar (rebar); 2) prefabricated high-strength rebar (HSR) assemblies with headed anchorages; and 3) high-strength concrete (HSC). The focus of the project was not to develop new innovations on materials, but rather to fill a major knowledge gap on the effectiveness, code conformity, and viability (i.e., practicality, commercial availability) of existing high-strength materials in nuclear structures, especially in shear walls, their connections/joints, and around penetrations. These advances promote the Nuclear Energy Enabling Technologies Advanced Methods for Manufacturing (NEET-1) goals of accelerating deployment schedules by at least 6 months and reducing component fabrication costs by 20% or more. Toward this vision, the following objectives were identified and completed: Develop limit-benefit and cost-benefit analysis frameworks: The structural and economic limits and benefits of HSR and HSC in nuclear walls were evaluated. Importantly, it was found that modest increases in concrete strength up to 10.0 ksi and rebar yield strength up to 100 ksi result in the most economic designs and improved structural performance (Barbachyn et al., 2017a and Barbachyn et al., 2017b). Also, an industry survey was distributed to determine the cost benefits associated with prefabrication. Importantly, on-site worker-hours are significantly reduced, allowing for significantly accelerated construction schedules (Devine et al. 2018b). Develop an optimization methodology for structural design: An optimization study was deemed not necessary since the trivial solution with maximized use of high-strength materials and prefabrication always resulted in the largest cost and time savings. Instead, the benefits (and limits) of high-strength materials and prefabrication on construction costs and on-site construction time were investigated through a parametric study. Cost-reductions through the use of HSR and HSC were found to be greater than 40%. Also, on-site construction time of nuclear shear walls can be reduced by up to 83% through the use of prefabricated rebar assemblies. Conduct experimental evaluations of nuclear shear walls and wall-foundation joints: A large scale experimental program was completed on shear-critical RC cantilever deep beams. From these experiments, it was demonstrated that increasing rebar yield strength had a greater effect on increasing lateral strength than increasing the concrete compressive strength. Further, using HSR together with HSC resulted in the highest lateral strength and displacement ductility (Devine et al. 2018a). Additionally, large-scale reversed-cyclic lateral load tests were conducted on RC shear walls. It was demonstrated that through the utilization of HSR and HSC, a wall with 55% reduced vertical and horizontal rebar area was able to achieve 91% of the peak lateral strength of a conventional wall using state-of-practice materials with the same geometry. Further, trim reinforcement with headed anchorages around the penetrations proved to be effective. Develop validated numerical simulation models: Numerical modeling was completed in two phases during this project using 3 different finite element software packages. The first phase used existing measured data from previous experimental studies on stocky RC walls to validate simplified predictive models. Models using two different programs were validated based on a database of 38 stocky walls (Barbachyn et al., 2017a). The second phase occurred after experimental testing using a third and more detailed finite element package. Prediction improvements for the initial stiffness, lateral deformations, damage propagation, and failure mechanisms were targeted. The resulting detailed models more accurately predicted the peak lateral strength, initial and secant stiffnesses, lateral deformations, and cracking patterns of the wall test specimens subjected to reversed-cyclic lateral loading as compared to the simplified pretest models. Develop validated design procedures, tools, and criteria: This project evaluated the accuracy of existing closed-form design equations from U.S. building design standards and previous research (Barbachyn et al. 2017a and Devine et al. 2018a). It was shown that the current predictive equations for shear strength capacity need to be revisited as they provide significant scatter and gross overestimates, indicating the potential for unconservative designs. Further, for both the deep beam and shear wall specimens, the numerical models and flexural capacity predictions provided the most accurate strength predictions. Develop cost-effective field construction procedures: To promote the use of prefabricated rebar assemblies, the following construction conditions where prefabrication would be used instead of in-place rebar tying were identified from an industry survey: 1) to save on construction schedule; 2) to improve safety and/or quality control; 3) for areas with heavy rebar congestion; and 4) for structures with significant repetition in rebar layout. All of these conditions directly apply and benefit nuclear construction where construction costs and schedules are extensive, safety and quality are of utmost importance, the reinforcement is extremely dense, and the reinforcement layout is repetitive in many regions of the shear walls. Additionally, an experimental study was conducted to investigate the changes in bar spacing (with respect to code-required rebar spacing and placement tolerances) when a horizontally assembled prefabricated two-dimensional (2D) rebar mat or three-dimensional (3D) rebar cage is lifted (or tripped) to a vertical position. It was found that the bars involved in both the lateral translation of the rebar assemblies (i.e., movement of the assembly in its horizontal position) and the tripping process were most susceptible to large spacing changes, and thus, their positions within the assembly would need to be checked after the placement of the assembly in place within the formwork. Importantly, the spacings of all of the other bars in the assembly were not affected significantly by the lateral translation and tripping process, thereby eliminating the need to recheck their positions within the assembly (Devine et al. 2018b).

36 MATERIALS SCIENCE↗
DOE OSTI2021

Storage Enabled Flexibility of Conventional Generation Assets (StorFlex)

The power systems have faced progressively more demanding operational requirements over the last two decades. Several factors contribute to these challenging operating conditions, including load growth, aging infrastructure, increasing penetrations of distributed energy resources (DERs), electrification of the economy, and policy initiatives such as decarbonization. The power system and its components must provide high operational flexibility to mitigate these challenges. For example, the proliferation of intermittent DERs such as wind and solar has increased the need for conventional generation assets like hydropower plants to respond to sudden load-generation imbalances. The higher flexibility requirements for hydropower plants cause more wear and tear, potentially shortening the useful lifespan of hydropower turbines. To reduce the need for hydropower plants to follow sudden changes in the dispatch signal, we investigate their combined operation with the energy storage systems (ESSs; “ESS-based hybridization”). Our analyses focuses on improving the lifespan of hydropower plants through ESS-based hybridization. Wear and tear on hydropower turbines (particularly Francis turbines) is modeled using a loss-of-life concept that is based on damage experienced by the turbine due to various cycles of operation. Then, we show that using ESSs to offset some of the high variation increases the remaining life of the hydropower plants. To demonstrate this, a few modeling tools were developed for this work: (1) a dynamic model for various components of the turbine and its governor; (2) a control strategy that assigns a slow-varying dispatch signal to a hydropower unit versus a fastmoving signal to ESS, such that the overall power request remains the same; and (3) models for the financial analysis to quantify the economic merits of such a framework. We used the models we developed to analyze the dispatch pattern of an actual hydropower plant with a power output of 50 MW and a head height of 152 m. This work showed that ESS-based hybridization could extend the life of the hydropower plant by 5% on average. This extension in life was then used to estimate the economic benefit in terms of cost deferrals associated with hydropower plant maintenance and replacement: on average, $3.6 million. Sensitivity analysis with respect to the size of ESS and cost of turbines was performed to show the variation in benefits over the range of turbine costs and ESS sizes. Crucially, stacking damage reduction and lifetime extension with other ESS value streams such as providing ancillary services could substantially increase the financial benefits of ESS-based hybridization. The higher costs associated with ESS of appropriate size would make more financial sense when multiple value streams are stacked and co-optimized to extract the maximum benefit. This dimension will be explored in future work.

13 HYDRO ENERGY↗
DOE OSTI2019

Data Locality Enhancement of Dynamic Simulations for Exascale Computing (Final Report)

The development of modern processors exhibits two trends that complicate the optimizations of modern software. The first is the increasing sensitivity of processors' throughput to irregularities in computation. With more processors produced through a massive integration of simple cores, future systems will increasingly favor regular data-level parallel computations, but deviate from the needs of applications with complex patterns. Some evidences are already shown on Graphic Processing Units (GPU): Irregular data accesses (e.g., indirect references A[D[i]]) and conditional branches are limiting many GPU applications' performance at a level an order of magnitude lower than the peak of GPU. The second hardware trend is the growing gap between memory bandwidth and the aggregate speed—that is, the sum of all cores' computing power—of a Chip Multiprocessor (CMP). Despite the capped growth of the peak CPU speed, the aggregate speed of a CMP keeps increasing as more cores get into a single chip. It is expected that by 2018, node concurrency in an exascale system will increase by hundreds of times, whereas, memory bandwidth will expand by only 10 to 20 times. Consequently, data movement and storage is expected to consume more than 70% of the total system power. Bridging this gap is difficult; the complexities of modern CMP memory hierarchy make it even harder: Data cache becomes shared among computing units, and the sharing is often non-uniform—whether two computing units share a cache depends on their proximity and the level of the cache. On the recent IBM Power7 architecture, for instance, four hardware contexts (or SMT threads) in a core share the entire memory hierarchy, all cores in one chip share an on-chip L3 cache, and cores across chips share L3 and main memory through off-chip connections. These two trends complicate the translation of computing power into performance, especially for a program with either intensive data accesses or complex patterns in data accesses or control flow paths. Unfortunately, both attributes present and will persist in a class of important applications. For instance, many scientific simulations deal with a large volume of data. And meanwhile, as most real-world processes are non-uniform and evolving (e.g., the evolution of a galaxy or the process of a drug injection), both the computations and data accesses of these programs tend to be irregular and dynamically changing. Currently, the lack of support to these applications on modern CMP severely limits their performance. On GPU, as our recent study shows and other studies echo, performance enhancement of a factor of integers is possible when memory accesses or control flows are streamlined for a set of GPU applications. On multicore CPU, our studies show that traditional locality enhancement, for being oblivious to the new features of multicore memory hierarchy, may even cause large slowdown to data-intensive dynamic applications. The severity of the issues is expected to worsen as the two hardware trends continue. Some recent studies try to match software with the trends, but in a limited scope or manner. For irregularities on GPU, most studies focus on irregularities analyzable through static analysis (e.g., data accesses in regular loops). Dynamic irregularities are harder to address because the needed analysis and transformations typically have to happen at run time. Some other research resorts to hardware extensions, an actual adoption of which is unclear for the entailed space cost and complexity. For data locality, recent years have seen some exploitations of the new memory hierarchy on multicore for performance, but most of them are on process or thread scheduling, rather than program transformations. Our recent study reveals that program-level transformations may magnify the scheduling benefits by a factor of seven, concluding that program-level transformation should play a central role for data locality enhancement on modern CMP. But research in this direction has been sparse, and most have focused on data layout or cache performance modeling, rather than program transformations to match with the new memory hierarchy features. Overall, it is still an open question how to bridge the gap between dynamic computations and the two prominent properties of modern processors. The goal of this project is to develop a set of techniques and software tools to enhance the matching between memory accesses in dynamic simulations and the prominent features of modern and future CMP, alleviating the memory performance issues for petascale and exascale computing. This report summarizes the discoveries and products produced throughout this project. It includes free launch, a new software approach to overcoming the shortcomings of both methods; coherence-free multiview, an approach that allows multiple views of a single data object to co-exist on GPU memory during a GPU kernel execution; algorithmic optimizations to data analytics problems, especially those that involve lots of distance calculations. Supporting dynamic parallelism is important for GPU to benefit a broad range of applications. There are currently two fundamental ways for programs to exploit dynamic parallelism on GPU: a software-based approach with software-managed worklists, and a hardware-based approach through dynamic subkernel launches. Neither is satisfactory. The former is complicated to program and is often subject to some load imbalance; the latter suffers large runtime overhead. In this work, we propose free launch, a new software approach to overcoming the shortcomings of both methods. It allows programmers to use subkernel launches to express dynamic parallelism. It employs a novel compiler-based code transformation named subkernel launch removal to replace the subkernel launches with the reuse of parent threads. Coupled with an adaptive task assignment mechanism, the transformation reassigns the tasks in the subkernels to the parent threads with a good load balance. The technique requires no hardware extensions, immediately deployable on existing GPUs. It keeps the programming convenience of the subkernel launch-based approach while avoiding its large runtime overhead. Meanwhile, its superior load balancing makes it outperform manual worklist-based techniques by 3X on average. The work was published at the 48th Annual IEEE/ACM International Symposium on Microarchitecture (Micro'2015). A Graphic Processing Unit (GPU) system is typically equipped with many types of memory (e.g., global, constant, texture, shared, cache). Data placement determines what data are placed on which type of memory, essential for GPU memory performance. Prior optimizations of data placement always require a single view of a data object on memory, which limits the optimization effectiveness. In this work, we propose coherence-free multiview, an approach that allows multiple views of a single data object to co-exist on GPU memory during a GPU kernel execution. We demonstrate that under certain conditions, the multiple views can remain incoherent while facilitating enhanced data placement. We present a theorem and some compiler support to ensure the soundness of the usage of coherence-free multiview. We further develop reference-discerning data placement, a new way to enhance data placements on GPU. It enables more flexible data placements by using coherence-free multiview to leverage the slack in coherence requirement of some GPU programs. Experiments on three types of GPU systems show that, with less than 200KB space cost, the new data placement technique can provide a 1.6X average (up to 4.27X) speedup. In addition, we have examined algorithmic optimizations to data analytics problems, especially those that involve lots of distance calculations. Computing distances among data points is an essential part of many important algorithms in data analytics, graph analysis, and other domains. In each of these domains, developers have spent significant manual effort optimizing algorithms, often through novel applications of the triangle equality, in order to minimize the number of distance computations in the algorithms. In this work, we observe that many algorithms across these domains can be generalized as an instance of a generic distance-related abstraction. Based on this abstraction, we derive seven principles for correctly applying the triangular inequality to optimize distance-related algorithms. Guided by the findings, we develop Triangular OPtimizer (TOP), the first software framework that is able to automatically produce optimized algorithms that either matches or outperforms manually designed algorithms for solving distance-related problems. TOP achieves up to 237x speedups and 2.5X on average. The work has been published at the 32nd International Conference on Machine Learning (ICML'15) and the 41st International Conference on Very Large Data Bases (VLDB'15).

42 ENGINEERING↗
DOE OSTI2018

Modular Controllable Transformers (MCT)

Large Power Transformers (LPTs) are at the heart of the electricity infrastructure and are critical pieces in today’s grid. Aging assets, long turn-around times, transportation delays and dependence on foreign manufacturing for these components have created conditions for points of failure across the system. Moreover, there is no available dynamic control on these passive sections of the grid. For all the automation embedded on the low voltage side of the low voltage side of the system there is absolutely no flexibility on the high voltage side. Changes in power flow patterns are brought about by altering dispatch patterns for generators which requires a coordinated effort and is often time consuming. Traditional FACTS based approaches such as UPFC or HVDC light involve massive modifications to the existing infrastructure. Recent instances of physical attacks on LPTs and the slow recovery process associated with fixing or replacing these damaged units has created an increased awareness for this issue. LPT designs are highly customized owing to specific field requirements. Long lead times coupled with transportation delays make procuring and installing LPTs a lengthy process. In an attempt to minimize the recovery time following an LPT failure, transformer manufacturers have developed mobile transformer units. These units typically consist of either a single-phase or three-phase transformer mounted on a truck and characterized by compact form factors. By reducing the assembly time and transport time these units offer a fast and temporary solution to LPT failures. Several efforts led by DOE, Edison Electric Company (EEI) and North American Electric Reliability Corporation (NERC) have led to the RecX program. RecX transformers consist of single phase transformers transported on specialized assemblies to enable fast restoration and minimal installation time. NERC and EEI have instituted transformer sharing programs like Spare Equipment Database (SED) as well as Spare Transformer Equipment Database (STEP) which encourage utilities to maintain and share a fleet of LPTs. Mobile transformers are typically limited to less than 100 MVA to enable transportation on a trailer. Also, the efficiency of mobile transformer units and RecX transformers is lower since they are temporary solutions. The ideal replacement solution would be one that could utilize a set of standardized transformers while incorporating flexibility to influence metrics like power flow, apparent impedance and voltage profiles. In view of these issues the team at Georgia Tech proposed a concept called Modular Controllable Transformers (MCTs). This project was commissioned as a part of DOE’s Transformer Resiliency and Advance Component (TRAC) program. The approach consists of splitting one large unit rated upwards of 100 MVA with smaller modular, standardized units. Each of these smaller units are augmented with a converter to add power flow, voltage and impedance control. Splitting the power level into multiple units has two important implications. By making smaller units, the transportation times can be reduced drastically. Moreover, even though these units have very specific impedance and other parameters, the augmented converters can make it emulate the unit being replaced. Thus, faster replacement times and added controls are achieved making this an extremely resilient approach. In the event of one modular units failure, the other two controllable units can operate in the most resilient fashion keeping the net system overloads to a minimum. Moreover, for the same probability of failure only a fraction of the system capacity is lost. The converter is augmented on to the tertiary winding of the transformer making it retrofittable in the field without massive modifications to the system. The material proposed in this report highlights design considerations to minimize recovery time as well as results displaying overall improvement in resiliency. The cost analysis of this approach shows significant savings compared to UPFC and HVDC lite based approaches. Extensive analysis was conducted on numerous aspects of this project over the course of this project. Simulation studies were conducted to prove the 5MVA converter design. Challenging BIL constraints and transient issues were addressed over the course of this project. Simulations on EMTP platform presented in consequent sections were used to finely tune the control for the same. A final design for a back-to-back 5MVA neutral-point-clamped (NPC) converter was completed in this project. Furthermore, switching components like high power IGBTs from IXYS Westcode Press-Pack series and ABB Stackpack series were identified for this application. These devices exhibit a blocking voltage of 6.5 kV and a nominal collector current of 900 A. Potential housing options as well as existing commercial option were identified for this converter design. A 56 MVA modular transformer unit was designed with Delta Star Inc. with consultation and inputs from all project members. The transformer cooling systems were chosen to be Oil Natural Air Forced (ONAF) to optimize ease of transport. A full Product Requirement Document (PRD) for the converter and the transformer was generated. The team also investigated key aspects of resiliency and how to quantify the same. A metric was developed to analyze resiliency and numerous simulation cases were presented. Increased (N-1) resiliency and outage resistance was showcased through simulation on the Texas Large scale power system. Dynamic control capabilities were also showcased on numerous system level simulations. An IEEE 30 bus system was used to show the ability of the MCT approach to reduce line congestion. Power flow control; a part of the MCTs abilities, showcased a 29% reduction in transformer loading on the same system. Full control over tie line flows was demonstrated on a modified IEEE 13 bus system. Dynamic LTC capabilities were shown through simulation and were compared to traditional LTCs. Furthermore, the ability of the MCTs to alter apparent transformer impedance was simulated to promote standardization of modular transformer units while retaining the ability to emulate a varied range of impedance characteristics. Thus granular control over power flow, voltages, impedances and was presented. Regulatory analysis conducted by Oak Ridge National Laboratory (ORNL) further bolstered the efficacy of the MCT approach and verified the added resiliency achieved due to this approach. As a conclusion, it was proved that the resilience improvement obtained from using the MCT approach is justified by just one failure in a 30-year period. This means that the benefits obtained from using the MCT approach far outweigh the cost of setting this approach up. As a part of the future efforts, the team is proposing building a (12.47-24) kV/400 kVA scaled down unit to control 5MVA of power flows. Emphasis would be laid on using commercial components to speed up the build of these units. Further the actual transport of the modular transformer unit would be a part of this operation and would provide insight into transportation issues. The approach shows much promise and adds key dynamic control on passive sub transmission and transmission sectors.

24 POWER TRANSMISSION AND DISTRIBUTION↗