SEARCH · Search NASA

Results for “Dynamic scheduling”

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.

SourceResearch connectionFrom yearThrough year

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

At least 127 records · Page 7

All sources NASA NTRS DOE OSTI

NASA NTRS1973

A vehicle scheduling algorithm using non-serial discrete dynamic programming with space shuttle applications

Description of the development and operation of a vehicle-scheduling algorithm which has applications to the NASA problem of assigning payloads to space delivery vehicles. The algorithm is based on a discrete, integer-valued, nonserial, dynamic-programming solution to the classical problem of developing resource utilization plans with limited resources. The algorithm places special emphasis on incorporating interpayload (precedence) relationships; maintaining optimal alternate schedule definitions (a unique feature of dynamic programming) in the event of contingencies (namely, resource inventory changes) without problem resolution; and, by using a special information storage technique, reducing the computational complexity of solving realistic problems.

Dupnick, E.↗

DOE OSTI2017

A novel approach to estimate emissions from large transportation networks: Hierarchical clustering-based link-driving-schedules for EPA-MOVES using dynamic time warping measures

We present that EPA-MOVES (Motor Vehicle Emission Simulator) is often integrated with traffic simulators to assess emission levels of large-scale urban networks with signalized intersections. High variations in speed profiles exist in the context of congested urban networks with signalized intersections. The traditional average-speed-based emission estimation technique with EPA-MOVES provides faster execution while underestimates the emissions in most cases because of ignoring the speed variation at congested networks with signalized intersections. In contrast, the atomic second-by-second speed profile (i.e., the trajectory of each vehicle)-based technique provides accurate emissions at the cost of excessive computational power and time. We addressed this issue by developing a novel method to determine the link-driving-schedules (LDSs) for the EPA-MOVES tool. Our research developed a hierarchical clustering technique with dynamic time warping similarity measures (HC-DTW) to find the LDS for EPA-MOVES that is capable of producing emission estimates better than the average-speed-based technique with execution time faster than the atomic speed profile approach. We applied the HC-DTW on a sample data from a signalized corridor and found that HC-DTW can significantly reduce computational time without compromising the accuracy. The developed technique in this research can substantially contribute to the EPA-MOVES-based emission estimation process for large-scale urban transportation network by reducing the computational time with reasonably accurate estimates. This method is highly appropriate for transportation networks with higher variation in speed such as signalized intersections. Lastly, experimental results show error difference ranging from 2% to 8% for most pollutants except PM 10 .

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

DOE OSTI2017

A dynamic control approach for energy-efficient production scheduling on a single machine under time-varying electricity pricing

This paper proposes a dynamic control algorithm to enable an energy-aware single machine scheduling under the time-varying electricity pricing policy, in which price rates remain fixed day-to-day over the season. The key issue is to assign a set of jobs to available time periods where different electricity prices are assigned, while considering requested due dates of jobs so as to minimize total penalty costs for earliness and tardiness of jobs and total energy consumption costs, simultaneously. As the first contribution of this study, we develop a new mixed integer nonlinear programming (MINLP) model that aims at determining job arrival times and resulting earliness and tardiness of jobs and energy consumption costs for machine idle and normal processing. Second, an efficient heuristic approach based on continuous-time variable control models and algorithm is developed. The proposed heuristic adaptively changes job arrival times and due dates, which finally determine production sequence over the time periods of different electricity prices, machine turn-off, and machine idle with minimum energy consumption costs and just-in-time (JIT) penalty. Energy and JIT performance of the proposed approach is examined using real energy and machining parameters of a HAAS machine and compared to those of the metaheuristic approach. For relatively large size data groups, the proposed approach incurs about higher energy consumption costs on average, which are offset by up to lower JIT costs, resulting in lower total costs on average compared to the metaheuristic approach. Further, the proposed time-scaled heuristic algorithm yields extremely short computational time, which enables production managers to flexibly select proper production strategies and to implement them for different production environments.

42 ENGINEERING↗

NASA NTRS1991

User interface issues in supporting human-computer integrated scheduling

The topics are presented in view graph form and include the following: characteristics of Operations Mission Planner (OMP) schedule domain; OMP architecture; definition of a schedule; user interface dimensions; functional distribution; types of users; interpreting user interaction; dynamic overlays; reactive scheduling; and transitioning the interface.

Cooper, Lynne P.↗

DOE OSTI2019

Sky Survey Scheduling Using Reinforcement Learning

Modern cosmic sky surveys (e.g., CMB S4, DES, LSST) collect a complex diversity of astronomical objects. Each of class objects presents different requirements for observation time and sensitivity. For determining the best sequence of exposures for mapping the sky systematically, conventional scheduling methods do not optimize the use of survey time and resources. Dynamic Sky survey scheduling is an np-hard problem that has been therefore treated primarily with heuristic methods. We present an alternative scheduling method based on reinforcement learning (RL) that aims to optimize the use of telescope resources for scheduling sky surveys.

79 ASTRONOMY AND ASTROPHYSICS↗

NASA NTRS2003

Interleaved Observation Execution and Rescheduling on Earth Observing Systems

Observation scheduling for Earth orbiting satellites solves the following problem: given a set of requests for images of the Earth, a set of instruments for acquiring those images distributed on a collecting of orbiting satellites, and a set of temporal and resource constraints, generate a set of assignments of instruments and viewing times to those requests that satisfy those constraints. Observation scheduling is often construed as a constrained optimization problem with the objective of maximizing the overall utility of the science data acquired. The utility of an image is typically based on the intrinsic importance of acquiring it (for example, its importance in meeting a mission or science campaign objective) as well as the expected value of the data given current viewing conditions (for example, if the image is occluded by clouds, its value is usually diminished). Currently, science observation scheduling for Earth Observing Systems is done on the ground, for periods covering a day or more. Schedules are uplinked to the satellites and are executed rigorously. An alternative to this scenario is to do some of the decision-making about what images are to be acquired on-board. The principal argument for this capability is that the desirability of making an observation can change dynamically, because of changes in meteorological conditions (e.g. cloud cover), unforeseen events such as fires, floods, or volcanic eruptions, or un-expected changes in satellite or ground station capability. Furthermore, since satellites can only communicate with the ground between 5% to 10% of the time, it may be infeasible to make the desired changes to the schedule on the ground, and uplink the revisions in time for the on-board system to execute them. Examples of scenarios that motivate an on-board capability for revising schedules include the following. First, if a desired visual scene is completely obscured by clouds, then there is little point in taking it. In this case, satellite resources, such as power and storage space can be better utilized taking another image that is higher quality. Second, if an unexpected but important event occurs (such as a fire, flood, or volcanic eruption), there may be good reason to take images of it, instead of expending satellite resources on some of the lower priority scheduled observations. Finally, if there is unexpected loss of capability, it may be impossible to carry out the schedule of planned observations. For example, if a ground station goes down temporarily, a satellite may not be able to free up enough storage space to continue with the remaining schedule of observations. This paper describes an approach for interleaving execution of observation schedules with dynamic schedule revision based on changes to the expected utility of the acquired images. We describe the problem in detail, formulate an algorithm for interleaving schedule revision and execution, and discuss refinements to the algorithm based on the need for search efficiency. We summarize with a brief discussion of the tests performed on the system.

Khatib, Lina↗

NASA NTRS2016

Utilizing the ISS Mission as a Testbed to Develop Cognitive Communications Systems

The ISS provides an excellent opportunity for pioneering artificial intelligence software to meet the challenges of real-time communications (comm) link management. This opportunity empowers the ISS Program to forge a testbed for developing cognitive communications systems for the benefit of the ISS mission, manned Low Earth Orbit (LEO) science programs and future planetary exploration programs. In November, 1998, the Flight Operations Directorate (FOD) started the ISS Antenna Manager (IAM) project to develop a single processor supporting multiple comm satellite tracking for two different antenna systems. Further, the processor was developed to be highly adaptable as it supported the ISS mission through all assembly stages. The ISS mission mandated communications specialists with complete knowledge of when the ISS was about to lose or gain comm link service. The current specialty mandated cognizance of large sun-tracking solar arrays and thermal management panels in addition to the highly-dynamic satellite service schedules and rise/set tables. This mission requirement makes the ISS the ideal communications management analogue for future LEO space station and long-duration planetary exploration missions. Future missions, with their precision-pointed, dynamic, laser-based comm links, require complete autonomy for managing high-data rate communications systems. Development of cognitive communications management systems that permit any crew member or payload science specialist, regardless of experience level, to control communications is one of the greater benefits the ISS can offer new space exploration programs. The IAM project met a new mission requirement never previously levied against US space-born communications systems management: process and display the orientation of large solar arrays and thermal control panels based on real-time joint angle telemetry. However, IAM leaves the actual communications availability assessment to human judgment, which introduces unwanted variability because each specialist has a different core of experience with comm link performance. Because the ISS utilizes two different frequency bands, dynamic structure can be occasionally translucent at one frequency while it can completely interdict service at the other frequency. The impact of articulating structure on the comm link can depend on its orientation at the time it impinges on the link. It can become easy for a human specialist to cross-associate experience at one frequency with experience at the other frequency. Additionally, the specialist's experience is incremental, occurring one nine-hour shift at a time. Only the IAM processor experiences the complete 24x7x365 communications link performance for both communications links but, it has no "learning capability." If the IAM processor could be endowed with a cognitive ability to remember past structure-induced comm link outages, based on its knowledge of the ISS position, attitude, communications gear, array joint angles and tracking accuracy, it could convey such experience to the human operator. It could also use its learned communications link behaviors to accurately convey the availability of future communications sessions. Further, the tool could remember how accurately or inaccurately it predicted availability and correct future predictions based on past performance. The IAM tool could learn frequency-specific impacts due to spacecraft structures and pass that information along as "experience." Such development would provide a single artificial intelligence processor that could provide two different experience bases. If it also "knew" the satellite service schedule, it could distinguish structure blockage from schedule or planet blockage and then quickly switch to another satellite. Alternatively, just as a human operator could judge, a cognizant comm system based on the IAM model could "know" that the blockage is not going to last very long and continue tracking a comm satellite, waiting for it to track away from structure. Ultimately, once this capability was fully developed and tested in the Mission Control Center, it could be transferred on-orbit to support development of operations concepts that include more advanced cognitive communications systems. Future applications of this capability are easily foreseen because even more dynamic satellite constellations with more nodes and greater capability are coming. Currently, the ISS fully employs a 300 million bit-per-second (Mbps) return link for harvesting payload science. In the coming eighteen months, it will step up to 600 Mbps. Already there is talk of a 1.2 billion bit-per-second (Gbps) upgrade for the ISS and laser comm links have already been tested from the ISS. Every data rate upgrade mandates more complicated and sensitive communications equipment which implies greater expertise invested in the human operator. Future on-orbit cognizant comm systems will be needed to meet greater performance demands aboard larger, far more complicated spacecraft. In the LEO environment, the old-style one-satellite-per-spacecraft operations concept will give way to a new concept of a single customer spacecraft simultaneously using multiple comm satellites. Much more highly-dynamic manned LEO missions with decades of crew members potentially increase the demand for communications link performance. A cognizant on-board communications system will meet advanced communications demands from future LEO missions and future planetary missions. The ISS has fledgling components of future exploration programs, both LEO and planetary. Further, the Flight Operations Directorate, through the IAM project, has already begun to develop a communications management system that attempts to solve advanced problems ideally represented by dynamic structure impacting scheduled satellite service. With an earnest project to integrate artificial intelligence into the IAM processor, the ISS Program could develop a cognizant communications system that could be adapted and transferred to future on-orbit avionics designs.

Jackson, Dan↗

NASA NTRS2016

Utilizing the ISS Mission as a Testbed to Develop Cognitive Communications Systems

The ISS provides an excellent opportunity for pioneering artificial intelligence software to meet the challenges of real-time communications (comm) link management. This opportunity empowers the ISS Program to forge a testbed for developing cognitive communications systems for the benefit of the ISS mission, manned Low Earth Orbit (LEO) science programs and future planetary exploration programs. In November, 1998, the Flight Operations Directorate (FOD) started the ISS Antenna Manager (IAM) project to develop a single processor supporting multiple comm satellite tracking for two different antenna systems. Further, the processor was developed to be highly adaptable as it supported the ISS mission through all assembly stages. The ISS mission mandated communications specialists with complete knowledge of when the ISS was about to lose or gain comm link service. The current specialty mandated cognizance of large sun-tracking solar arrays and thermal management panels in addition to the highly-dynamic satellite service schedules and rise/set tables. This mission requirement makes the ISS the ideal communications management analogue for future LEO space station and long-duration planetary exploration missions. Future missions, with their precision-pointed, dynamic, laser-based comm links, require complete autonomy for managing high-data rate communications systems. Development of cognitive communications management systems that permit any crew member or payload science specialist, regardless of experience level, to control communications is one of the greater benefits the ISS can offer new space exploration programs. The IAM project met a new mission requirement never previously levied against US space-born communications systems management: process and display the orientation of large solar arrays and thermal control panels based on real-time joint angle telemetry. However, IAM leaves the actual communications availability assessment to human judgement, which introduces unwanted variability because each specialist has a different core of experience with comm link performance. Because the ISS utilizes two different frequency bands, dynamic structure can be occasionally translucent at one frequency while it can completely interdict service at the other frequency. The impact of articulating structure on the comm link can depend on its orientation at the time it impinges on the link. It can become easy for a human specialist to cross-associate experience at one frequency with experience at the other frequency. Additionally, the specialist's experience is incremental, occurring one nine-hour shift at a time. Only the IAM processor experiences the complete 24x7x365 communications link performance for both communications links but, it has no "learning capability." If the IAM processor could be endowed with a cognitive ability to remember past structure-induced comm link outages, based on its knowledge of the ISS position, attitude, communications gear, array joint angles and tracking accuracy, it could convey such experience to the human operator. It could also use its learned communications link behaviors to accurately convey the availability of future communications sessions. Further, the tool could remember how accurately or inaccurately it predicted availability and correct future predictions based on past performance. The IAM tool could learn frequency-specific impacts due to spacecraft structures and pass that information along as "experience." Such development would provide a single artificial intelligence processor that could provide two different experience bases. If it also "knew" the satellite service schedule, it could distinguish structure blockage from schedule or planet blockage and then quickly switch to another satellite. Alternatively, just as a human operator could judge, a cognizant comm system based on the IAM model could "know" that the blockage is not going to last very long and continue tracking a comm satellite, waiting for it to track away from structure. Ultimately, once this capability was fully developed and tested in the Mission Control Center, it could be transferred on-orbit to support development of operations concepts that include more advanced cognitive communications systems. Future applications of this capability are easily foreseen because even more dynamic satellite constellations with more nodes and greater capability are coming. Currently, the ISS fully employs its high-data-rate return link for harvesting payload science. In the coming months, it will double that data rate and is forecast to fully utilize that capability. Already there is talk of an upgrade that quadruples the current data rate allocated to ISS payload science before the end of its mission and laser comm links have already been tested from the ISS. Every data rate upgrade mandates more complicated and sensitive communications equipment which implies greater expertise invested in the human operator. Future on-orbit cognizant comm systems will be needed to meet greater performance demands aboard larger, far more complicated spacecraft. In the LEO environment, the old-style one-satellite-per-spacecraft operations concept will give way to a new concept of a single customer spacecraft simultaneously using multiple comm satellites. Much more highly-dynamic manned LEO missions with decades of crew members potentially increase the demand for communications link performance. A cognizant on-board communications system will meet advanced communications demands from future LEO missions and future planetary missions. The ISS has fledgling components of future exploration programs, both LEO and planetary. Further, the Flight Operations Directorate, through the IAM project, has already begun to develop a communications management system that attempts to solve advanced problems ideally represented by dynamic structure impacting scheduled satellite service. With an earnest project to integrate artificial intelligence into the IAM processor, the ISS Program could develop a cognizant communications system that could be adapted and transferred to future on-orbit avionics designs.

Jackson, Dan↗

NASA NTRS1992

IOPS advisor: Research in progress on knowledge-intensive methods for irregular operations airline scheduling

Our research focuses on the problem of recovering from perturbations in large-scale schedules, specifically on the ability of a human-machine partnership to dynamically modify an airline schedule in response to unanticipated disruptions. This task is characterized by massive interdependencies and a large space of possible actions. Our approach is to apply the following: qualitative, knowledge-intensive techniques relying on a memory of stereotypical failures and appropriate recoveries; and quantitative techniques drawn from the Operations Research community's work on scheduling. Our main scientific challenge is to represent schedules, failures, and repairs so as to make both sets of techniques applicable to the same data. This paper outlines ongoing research in which we are cooperating with United Airlines to develop our understanding of the scientific issues underlying the practicalities of dynamic, real-time schedule repair.

Borse, John E.↗

DOE OSTI2017

Exploiting Vector and Multicore Parallelsim for Recursive, Data- and Task-Parallel Programs

Modern hardware contains parallel execution resources that are well-suited for data-parallelism-vector units-and task parallelism-multicores. However, most work on parallel scheduling focuses on one type of hardware or the other. In this work, we present a scheduling framework that allows for a unified treatment of task- and data-parallelism. Our key insight is an abstraction, task blocks, that uniformly handles data-parallel iterations and task-parallel tasks, allowing them to be scheduled on vector units or executed independently as multicores. Our framework allows us to define schedulers that can dynamically select between executing task- blocks on vector units or multicores. We show that these schedulers are asymptotically optimal, and deliver the maximum amount of parallelism available in computation trees. To evaluate our schedulers, we develop program transformations that can convert mixed data- and task-parallel pro- grams into task block-based programs. Using a prototype instantiation of our scheduling framework, we show that, on an 8-core system, we can simultaneously exploit vector and multicore parallelism to achieve 14×-108× speedup over sequential baselines.

Ren, Bin↗

NASA NTRS2006

Conflict-Aware Scheduling Algorithm

conflict-aware scheduling algorithm is being developed to help automate the allocation of NASA s Deep Space Network (DSN) antennas and equipment that are used to communicate with interplanetary scientific spacecraft. The current approach for scheduling DSN ground resources seeks to provide an equitable distribution of tracking services among the multiple scientific missions and is very labor intensive. Due to the large (and increasing) number of mission requests for DSN services, combined with technical and geometric constraints, the DSN is highly oversubscribed. To help automate the process, and reduce the DSN and spaceflight project labor effort required for initiating, maintaining, and negotiating schedules, a new scheduling algorithm is being developed. The scheduling algorithm generates a "conflict-aware" schedule, where all requests are scheduled based on a dynamic priority scheme. The conflict-aware scheduling algorithm allocates all requests for DSN tracking services while identifying and maintaining the conflicts to facilitate collaboration and negotiation between spaceflight missions. These contrast with traditional "conflict-free" scheduling algorithms that assign tracks that are not in conflict and mark the remainder as unscheduled. In the case where full schedule automation is desired (based on mission/event priorities, fairness, allocation rules, geometric constraints, and ground system capabilities/ constraints), a conflict-free schedule can easily be created from the conflict-aware schedule by removing lower priority items that are in conflict.

Wang, Yeou-Fang↗

NASA NTRSDate not supplied

Emulated Spacecraft Communication Testbed for Evaluating Cognitive Networking Technology

The ability to emulate the full space protocol stack is an essential aspect required to evaluate and mature cognitive communication capabilities. The interaction between the physical layer and network layers is key to developing network optimizations for a dynamic and complex environment. We present a laboratory testbed for the evaluation of cognitive radio and networking techniques applied to space communications. The testbed is a high fidelity, flight-like hardware testbed consisting of software-defined radios, channel emulators, modems, and orbital analysis and scheduling software. The testbed uses RF links with signal quality, propagation delay, and Doppler effects driven by orbital mechanics simulations of emulated spacecraft. Our framework enables control of link bidirectionality, data rates, and interference sources. In addition to hardware radio nodes, the testbed can incorporate virtualized emulated nodes for larger and more challenging network scenarios. Our approach to a cognitive communication system uses delay tolerant networking (DTN) to mitigate the challenges of the space environment. While many DTN networks use only preplanned schedules, our system uses User-Initiated Service (UIS) to dynamically schedule service providers. Software-defined radio allows the system to adapt to a variety of service providers. Integration of DTN, UIS, and software-defined radio technologies provides a framework for the implementation of a cognitive communication system. This paper describes the testbed capabilities, network emulation approach, component integration, and initial end-to-end testing results.

cognitive radio↗

NASA NTRS1981

Real time simulation of computer-assisted sequencing of terminal area operations

A simulation was developed to investigate the utilization of computer assisted decision making for the task of sequencing and scheduling aircraft in a high density terminal area. The simulation incorporates a decision methodology termed Constrained Position Shifting. This methodology accounts for aircraft velocity profiles, routes, and weight classes in dynamically sequencing and scheduling arriving aircraft. A sample demonstration of Constrained Position Shifting is presented where six aircraft types (including both light and heavy aircraft) are sequenced to land at Denver's Stapleton International Airport. A graphical display is utilized and Constrained Position Shifting with a maximum shift of four positions (rearward or forward) is compared to first come, first serve with respect to arrival at the runway. The implementation of computer assisted sequencing and scheduling methodologies is investigated. A time based control concept will be required and design considerations for such a system are discussed.

Dear, R. G.↗

NASA NTRSDate not supplied

Advances in Modeling Solar System Internet Structures and their Data Flows

With an ever-increasing presence in space, there is also an increasing burden on existing communications infrastructure. We are heading towards an inflection point where the traditional approach of scheduled, single-path communications for space will no longer be viable. One answer is Delay Tolerant Networking (DTN), which takes the once disparate system of point-to-point links and unifies them in a networked architecture, thereby making communications more scalable. However, much work remains for discovering and harnessing the underlying theory of DTN. For example, in the terrestrial setting the interplay between routing domains is well-understood, however this is not the case in DTNs. In this paper, we build up the fundamental foundations of DTN, with an emphasis on modeling time varying networks and data flows across them, with examples of cross-domain routing in a DTN. A lofty goal of DTN is to enable the so-called Solar System Internet (SSI), which implies a standardized and robust suite of protocols. These protocols include routing across disconnected networks using store, carry, and forward mechanisms, which is necessary due to the disconnections, delays, and mobility intrinsic to space networks. Due to these factors, each of which generalize traditional networking, there is a deep and rich theory of DTNs. Here we build off of past successes to broaden this theory while striving to keep actionable results a goal for future implementations and operations. The approach includes modeling the unicast, broadcast, and multicast communications using the language of hypergraphs, which capture the geometric properties of such networked communications algebraically. Also inherent to these networks is their time-varying nature, particularly given mobility, and hence we also cultivate modeling techniques that respect this time dependence. This leads us to develop models using tools from category theory and algebraic geometry, which provide a language well-suited to describing synchronization and optimization over such networks. We also introduce and study a novel generalization of curvature applicable to time-evolving networks, which provides quantitative controls on diffusion processes on the network. Because an interplanetary network would feature links with propagation delays the preclude discovery (feedback) mechanisms, they will always feature a scheduled component. However, it is beneficial to support discovery where possible. While DTNs do not yet have strong definitions for their analogues of autonomous systems or network areas, we show how to join dynamic and schedule-based routing domains, using the language of sheaves, which marks progress towards such definitions. We conclude with a discussion of the progress made, as well as suggestions for future work.

Delay Tolerant Networking↗

DOE OSTI2021

Identification and Online Updating of Dynamic Models for Demand Response of an Industrial Air Separation Unit

Demand-response operation of air separation units requires frequent changes in production rate(s), and scheduling calculations must explicitly consider process dynamics to ensure feasibility of the solutions. To this end, scale-bridging models (SBMs) approximate the scheduling-relevant dynamics of a process and its controller in a low-order representation. In contrast to previous works that have employed nonlinear SBMs, this paper proposes linear SBMs, developed using time-series analysis, to facilitate online scheduling computations. Using a year-long industrial dataset, we nd that compact linear SBMs are suitable approximations over typical scheduling horizons, but that their accuracies are unpredictable over time. We introduce a strategy for online updating of the SBMs, based on Kalman ltering schemes for online parameter estimation. The approach greatly improves the accuracy of SBM predictions and will enable the use of linear SBM-based demand-response scheduling in the future.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

NASA NTRS1994

Solution and reasoning reuse in space planning and scheduling applications

In the space domain, as in other domains, the CSP (Constraint Satisfaction Problems) techniques are increasingly used to represent and solve planning and scheduling problems. But these techniques have been developed to solve CSP's which are composed of fixed sets of variables and constraints, whereas many planning and scheduling problems are dynamic. It is therefore important to develop methods which allow a new solution to be rapidly found, as close as possible to the previous one, when some variables or constraints are added or removed. After presenting some existing approaches, this paper proposes a simple and efficient method, which has been developed on the basis of the dynamic backtracking algorithm. This method allows previous solution and reasoning to be reused in the framework of a CSP which is close to the previous one. Some experimental results on general random CSPs and on operation scheduling problems for remote sensing satellites are given.

Verfaillie, Gerard↗

DOE OSTI2020

Emerging Technologies for Improved Plug Load Management Systems: Learning Behavior Algorithms and Automatic and Dynamic Load Detection

Plug loads are responsible for a significant portion of the energy consumed in commercial buildings, yet their distributed and ever-changing nature makes them one of the most challenging building end uses to manage. Plug load management systems exist today that utilize smart plugs to meter and control devices at the outlet level, however, their uptake has been relatively slow in part due to the significant labor required for installation and maintenance. Learning behavior algorithms and automatic and dynamic load detection have been identified as two technology areas that could accelerate the adoption of plug load management systems by reducing these labor demands and providing additional energy efficiency and non-energy benefits. Learning behavior algorithms learn occupant behavior and adjust plug load management systems accordingly, allowing for the automatic creation of optimized control schedules. Automatic and dynamic load detection allows a plug load management system to identify devices as they are plugged in to a building and keeps the system up to date as devices are moved throughout a building. In this paper, we present our findings with respect to the current state of these two technologies based on a review of existing research and patents, as well as a series of interviews with companies working in the plug load space. We have found that, as of now, no commercialized solutions exist for these plug load technologies and that more work is needed to bring them to market. In addition, we summarize our findings related to the technology challenges, market barriers, drivers, and opportunities for these technologies moving forward.

30 DIRECT ENERGY CONVERSION↗

DOE OSTI2018

SuperNeurons: Dynamic GPU Memory Management for Training Deep Neural Networks

Going deeper and wider in neural architectures improves their accuracy, while the limited GPU DRAM places an undesired restriction on the network design domain. Deep Learning (DL) practitioners either need to change to less de- sired network architectures, or nontrivially dissect a network across multiGPUs. These distract DL practitioners from concentrating on their original machine learning tasks. We present SuperNeurons: a dynamic GPU memory scheduling runtime to enable the network training far beyond the GPU DRAM capacity. SuperNeurons features 3 memory optimizations, Liveness Analysis, Unified Tensor Pool, and Cost-Aware Recomputation; together they effectively reduce the network-wide peak memory usage down to the maximal memory usage among layers. We also address the performance issues in these memory-saving techniques. Given the limited GPU DRAM, SuperNeurons not only provisions the necessary memory for the training, but also dynamically allocates the memory for convolution workspaces to achieve the high performance. Evaluations against Caffe, Torch, MXNet and TensorFlow have demonstrated that SuperNeurons trains at least 3.2432 deeper network than current ones with the leading performance. Particularly, SuperNeurons can train ResNet2500 that has 104 basic network layers on a 12GB K40c.

Wang, Linnan↗