Search NASA⌕ Search

SEARCH · Search NASA

Results for “Heuristic 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.

At least 55 records · Page 3

Adaptive language model training for molecular design

Abstract The vast size of chemical space necessitates computational approaches to automate and accelerate the design of molecular sequences to guide experimental efforts for drug discovery. Genetic algorithms provide a useful framework to incrementally generate molecules by applying mutations to known chemical structures. Recently, masked language models have been applied to automate the mutation process by leveraging large compound libraries to learn commonly occurring chemical sequences (i.e., using tokenization) and predict rearrangements (i.e., using mask prediction). Here, we consider how language models can be adapted to improve molecule generation for different optimization tasks. We use two different generation strategies for comparison, fixed and adaptive. The fixed strategy uses a pre-trained model to generate mutations; the adaptive strategy trains the language model on each new generation of molecules selected for target properties during optimization. Our results show that the adaptive strategy allows the language model to more closely fit the distribution of molecules in the population. Therefore, for enhanced fitness optimization, we suggest the use of the fixed strategy during an initial phase followed by the use of the adaptive strategy. We demonstrate the impact of adaptive training by searching for molecules that optimize both heuristic metrics, drug-likeness and synthesizability, as well as predicted protein binding affinity from a surrogate model. Our results show that the adaptive strategy provides a significant improvement in fitness optimization compared to the fixed pre-trained model, empowering the application of language models to molecular design tasks.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Teaching Game Theory to Kids and Limits of Prediction

I have once been asked to read a lecture to a group of 6th graders on Game Theory. After agreeing to it, I realized that explaining the game theory basics to 6th graders my be difficult, given that terms such as Nash equilibrium, minimax, maximin, optimization may not resonate in a 6th grade classroom. Instead I've introduced game theory using the rock-paper-scissors (RPS) game. Turns out kids are excellent gametheoreticians. In RPS, they understood both the benefits of randomizing their own strategy and of predicting their opponent's moves. They offered a number of heuristics both for the prediction and opening move. These heuristics included optimizing against past opponent moves, such as not playing rock if the opponent just played scissors, and playing a specific opening hand, such as "paper". Visualizing the effects of such strategic choices on-the-fly would be interesting and educational. This brief essay attempts demonstrating and visualizing the value of a few different strategic options in RPS. Specifically, we would like to illustrate the following: 1) what is the value of being unpredictable?; and 2) what is the value of being able to predict your opponent? In regard to prediction of human players, the question 2) has been reflected in Jon McLoone's entry in Wolfram Blog from January 20, 2014[1]. McLoone created a predictive algorithm for playing against human opponents, that learns to beat human opponents reliably after approximately 30 - 40 games. I use McLoone's implementation to represent a predictive and random strategies. The rest of this documents 1) investigates performance of this predictive strategy against a random strategy (which is optimal in RPS) and in 2) attempts to turn this predictive power against the predictive strategy by allowing the opponent the full knowledge of the predictor's strategy (but not the choices made using the strategy). This exposes a weakness in predictions made without taking risks into account by illustrating that predictive strategy may make the predictor predictable as well.

97 MATHEMATICS AND COMPUTING↗

Transient analysis and energy optimization of solar heating and cooling systems in various configurations

In this paper, a transient simulation model of solar-assisted heating and cooling systems (SHC) is presented. A detailed case study is also discussed, in which three different configurations are considered. In all cases, the SHC system is based on the coupling of evacuated solar collectors with a single-stage LiBr-H{sub 2}O absorption chiller, and a gas-fired boiler is also included for auxiliary heating, only during the winter season. In the first configuration, the cooling capacity of the absorption chiller and the solar collector area are designed on the basis of the maximum cooling load, and an electric chiller is used as the auxiliary cooling system. The second layout is similar to the first one, but, in this case, the absorption chiller and the solar collector area are sized in order to balance only a fraction of the maximum cooling load. Finally, in the third configuration, there is no electric chiller, and the auxiliary gas-fired boiler is also used in summer to feed the absorption chiller, in case of scarce solar irradiation. The simulation model was developed using the TRNSYS software, and included the analysis of the dynamic behaviour of the building in which the SHC systems were supposed to be installed. The building was simulated using a single-lumped capacitance model. An economic model was also developed, in order to assess the operating and capital costs of the systems under analysis. Furthermore, a mixed heuristic-deterministic optimization algorithm was implemented, in order to determine the set of the synthesis/design variables that maximize the energy efficiency of each configuration under analysis. The results of the case study were analyzed on monthly and weekly basis, paying special attention to the energy and monetary flows of the standard and optimized configurations. The results are encouraging as for the potential of energy saving. On the contrary, the SHC systems appear still far from the economic profitability: however, this is notoriously true for the great majority of renewable energy systems. (author)

14 SOLAR ENERGY↗

Inverse reinforcement learning control for building energy management

Reinforcement learning (RL) based control is widely considered a promising approach in building automation and control as it has demonstrated the potential to deal with complex objectives in adjacent domains like robotics, autonomous vehicles, gaming applications, and advertisement recommendations. When applied to any environment, model-free RL learns to improve its control performance over time without requiring a control model, by receiving and then analyzing feedback from the building environment after each control action. Operational objectives are becoming increasingly complex through the simultaneous consideration of thermal comfort, carbon emissions, grid services, and indoor air quality. In this context, conventional rule-based control approaches are proving sub-optimal, mostly heuristic, and inadequate. The model-free and self learning nature of RL appears promising and attractive as it may address the scalability issues associated with advanced control approaches. However, it suffers from long training times and unstable control behavior during the early stages of its learning process, which makes it unsuitable to be applied directly to buildings. This paper addresses these issues using an inverse reinforcement learning approach (IRL), a technique utilized to learn the objective of a controller agent which is considered an expert in its respective domain. Here, we consider a rule-based control as the expert demonstrator. IRL is different from a direct imitation (i.e., direct mapping of states to actions) of control actions as it tries to find the underlying intent of an expert's policy, providing the controller with a better-generalized policy for unseen states or environments with slightly different dynamics. This approach propels the RL controller's policy to levels similar to or better than that of a rule-based policy before it starts learning by interacting with the building. This makes RL for building energy management applications more practical as it prevents the erratic and exploratory behavior in the initial training period, simultaneously speeding up the learning process when compared to applying an untrained RL agent directly to a building environment.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

End-to-end protocol for high-quality quantum approximate optimization algorithm parameters with few shots

The quantum approximate optimization algorithm (QAOA) is a quantum heuristic for combinatorial optimization that has been demonstrated to scale better than state-of-the-art classical solvers for some problems. For a given problem instance, QAOA performance depends crucially on the choice of the parameters. While average-case optimal parameters are available in many cases, meaningful performance gains can be obtained by fine-tuning these parameters for a given instance. This task is especially challenging, however, when the number of circuit executions (shots) is limited. In this work, we develop an end-to-end protocol that combines multiple parameter settings and fine-tuning techniques. We use large-scale numerical experiments to optimize the protocol for the shot-limited setting and observe that optimizers with the simplest internal model (linear) perform best. We implement the optimized pipeline on a trapped-ion processor using up to 32 qubits and 5 QAOA layers, and we demonstrate that the pipeline is robust to small amounts of hardware noise. To the best of our knowledge, these are the largest demonstrations of QAOA parameter fine-tuning on a trapped-ion processor in terms of two-qubit gate count.

quantum algorithms & computation↗

Data-Driven Voltage Regulation of Distribution Grid Using Nonlinear Autoregressive Model with Exogenous Inputs (NARX)

This article proposes data-driven control via a nonlinear autoregressive model with exogenous inputs (NARX) for real-time voltage regulation of a modified feeder using reactive power sources. Traditional voltage control strategies rely on rule-based heuristics or optimization techniques, which often require detailed system models and extensive computational resources. The NARX-based controller learns system dynamics from historical data and predicts optimal reactive power dispatch in real-time for voltage correction. The proposed approach is evaluated on a power system feeder model under varying load and network conditions. Simulation results demonstrate that the NARX-based controller achieves improved voltage regulation, offering higher adaptability to system fluctuations. This study highlights the potential of data-driven control for enhancing the reliability of power distribution networks.

Donge, Vrushabh [ORNL] (ORCID:0000000306062803)↗

Impacts of Dispatch Strategies and Forecast Errors on the Economics of Behind-the-Meter PV-Battery Systems

To assess the economic value of batteries in hybrid PV-battery systems, one must create a dispatch profile for the battery. Many analyses of battery value assume perfect forecasts of PV generation and load, determining an upper limit on the value of the battery. Prior work that accounts for forecast uncer- tainty often does so in the context of a single dispatch algorithm, which does not provide a baseline for comparison. Furthermore, when multiple dispatch algorithms are assessed with uncertainty, the benefits considered are for diesel generation in a microgrid, not retail rate savings. This work addresses the gaps in the literature by comparing the performance of both heuristic and optimal dispatch algorithms for retail rate savings under forecast uncertainty, and provides comparisons of the robustness of these algorithms and their associated estimates of economic value. We find that using a perfect forecast can overestimate the value of hybrid PV-battery systems between 1% and 8% compared to the reality of using a day-ahead forecast, depending on the dispatch algorithm used. Thus, accounting for forecast uncertainty in system design and analysis will significantly improve the accuracy of modeled system values.

batteries↗

Energy Consumption and Cost Reduction of Future Light-Duty Vehicles through Advanced Vehicle Technologies: A Modeling Simulation Study Through 2050

The U.S. Department of Energy’s (DOE’s) Vehicle Technologies Office (VTO) and Hydrogen and Fuel Cell Technologies Office (HFTO) aim to develop sustainable, affordable and efficient technologies for transportation of goods and people. Translating investments in advanced transportation component technologies and powertrains to estimate vehicle-level fuel savings potential is critical for understanding DOE’s impact. In this work, we simulated technologies funded by VTO and HFTO for light duty vehicles. The simulations were performed across: Multiple powertrain configurations (i.e., conventional, power-split, extended-range electric vehicle, battery electric drive, and fuel-cell vehicles), Vehicle classes (i.e., compact car, midsize car, small sport utility vehicle [SUV], midsize SUV, and pickup trucks); and Fuels (i.e., gasoline, diesel, hydrogen, and battery electricity). These various technologies are assessed for six different timeframes: laboratory years 2015, 2020, 2025, 2030, and 2045. A delay of 5 years is assumed between laboratory year and model year (year technology is introduced into production). Finally, uncertainties are included for both technology performance and cost aspects by considering two cases: Low case, aligned with DOE technology manager estimates of expected original equipment manufacturer (OEM) improvements based on regulations, business as usual; and High case, aligned with aggressive technology advancements based on R&D targets developed through support by VTO & HFTO. These scenarios are not intended as predictions of future performances. The energy and cost impact of different technologies were estimated using Autonomie (www.autonomie.net), Argonne vehicle system simulation tool. Autonomie is a state-of-the-art vehicle system simulation tool used to assess the energy consumption, performance and cost of multiple advanced vehicle technologies across classes (from light to heavy duty), powertrains (from conventional to HEVs, FCEVs, PHEVs and BEVs), components and control strategies. Autonomie is packaged with a complete set of vehicle models for a wide range of vehicle classes, powertrain configurations and component technologies, including vehicle level and component level controls. These controls were developed and calibrated using dynamometer test data. Autonomie has been used to support a wide range of studies including analyzing various component technologies, sizing powertrains components for different vehicle requirements, comparing the benefits of powertrain configurations, optimizing both heuristic and route based vehicle energy control and predicting transportation energy use when paired with a traffic modeling tool such as POLARIS. This report documents the assumptions and estimates the vehicle-level energy consumption benefits and associated technology costs for the various types of light duty vehicles. All details of vehicle assumptions and simulation results are available in the spreadsheets accompanying this report.

33 ADVANCED PROPULSION SYSTEMS↗

A Comprehensive Simulation Study to Evaluate Future Vehicle Energy and Cost Reduction Potential

Under the umbrella of EERE’s Office of Sustainable Transportation, the U.S. Department of Energy’s (DOE) Vehicle Technologies Office (VTO) and Hydrogen and Fuel Cell Technologies Office (HFTO) seek to develop sustainable, affordable, and efficient technologies for transportation of goods and people. Translating investments in advanced transportation component technologies and powertrains to estimate the potential for vehicle-level fuel savings is critical to understanding DOE’s impact and success in this mission For this study, Argonne National Laboratory (Argonne) simulated technologies funded by VTO and HFTO for light-duty vehicles across the following: Powertrain configurations (conventional, power-split hybrid electric vehicle, extended-range electric vehicle, battery electric drive, and fuel-cell vehicles); Vehicle classes (compact car, mid-size car, small sport utility vehicle [SUV], mid-size SUV, and pickup truck); Fuels (gasoline, diesel, natural gas, hydrogen, and battery electricity). We assessed each technology for five different timeframes: laboratory years 2015 (reference), 2020, 2025, 2030, and 2045. We assumed a delay of 5 years between laboratory year and model year (i.e., the year the technology is introduced into production). Finally, we included uncertainties for both technology performance and cost by considering two cases (note that these cases are not intended as predictions of future performance): Low case , aligned with DOE technology manager estimates of expected original equipment manufacturer (OEM) improvements based on business as usual regulatory and market environments; High case , aligned with aggressive technology advancements based on research and development (R&D) targets developed through support by VTO and HFTO. We estimated the energy and cost impact of different technologies using Autonomie (Argonne undated), a state-of-the-art vehicle system simulation tool developed by Argonne and used to assess the energy consumption, performance, and cost of multiple advanced vehicle technologies. The tool comprises a complete set of vehicle models to assess impacts across a wide range of classes (from light- to heavy-duty), powertrain configurations (from conventional to hybrid electric vehicles [HEVs], fuel cell electric vehicles [FCEVs], plug-in hybrid electric vehicles [PHEVs], and battery electric vehicles [BEVs]), components, and control strategies, including vehicle-level and component-level controls developed and calibrated using dynamometer test data. Autonomie has been used to support a wide range of studies: analyzing various component technologies, sizing powertrain components to meet different vehicle requirements, comparing the benefits of powertrain configurations, optimizing both heuristic and route-based vehicle energy control, and predicting transportation energy use when paired with a traffic modeling tool such as POLARIS. This report documents the assumptions made and the vehicle-level energy consumption benefits and associated technology costs estimated for various types of light-duty vehicles. Details regarding vehicle assumptions and simulation results are available in the spreadsheets accompanying this report.

08 HYDROGEN↗

Impacts of Dispatch Strategies and Forecast Errors on the Economics of Behind-the-Meter PV-Battery Systems

To assess the economic value of batteries in hybrid PV-battery systems, one must create a dispatch profile for the battery. Many analyses of battery value assume perfect forecasts of PV generation and load, determining an upper limit on the value of the battery. Prior work that accounts for forecast uncertainty often does so in the context of a single dispatch algorithm, which does not provide a baseline for comparison. Furthermore, when multiple dispatch algorithms are assessed with uncertainty, the benefits considered are for diesel generation in a microgrid, not retail rate savings. This work addresses the gaps in the literature by comparing the performance of both heuristic and optimal dispatch algorithms for retail rate savings under forecast uncertainty, and provides comparisons of the robustness of these algorithms and their associated estimates of economic value. We find that using a perfect forecast can overestimate the value of hybrid PV-battery systems between 1% and 8% compared to the reality of using a day-ahead forecast, depending on the dispatch algorithm used. Thus, accounting for forecast uncertainty in system design and analysis will significantly improve the accuracy of modeled system values.

batteries↗

Impacts of Dispatch Strategies and Forecast Errors on the Economics of Behind-the-Meter PV-Battery Systems: Preprint

To assess the economic value of batteries in hybrid PV-battery systems, one must create a dispatch profile for the battery. Many analyses of battery value assume perfect forecasts of PV generation and load, determining an upper limit on the value of the battery. Prior work that accounts for forecast uncertainty often does so in the context of a single dispatch algorithm, which does not provide a baseline for comparison. Furthermore, when multiple dispatch algorithms are assessed with uncertainty, the benefits considered are for diesel generation in a microgrid, not retail rate savings. This work addresses the gaps in the literature by comparing the performance of both heuristic and optimal dispatch algorithms for retail rate savings under forecast uncertainty, and provides comparisons of the robustness of these algorithms and their associated estimates of economic value. We find that using a perfect forecast can overestimate the value of hybrid PV-battery systems between 1% and 8% compared to the reality of using a day-ahead forecast, depending on the dispatch algorithm used. Thus, accounting for forecast uncertainty in system design and analysis will significantly improve the accuracy of modeled system values.

batteries↗

OCTOKV: An Agile Network-Based Key-Value Storage System with Robust Load Orchestration

In this paper, we propose OctoKV, an innovative network-based key-value storage system. OctoKV addresses the repetitive address translation overhead associated with traditional key-value stores running on file systems on the client side. To mitigate this overhead, we implemented the key-value store on the server side using NVMe-oF and a user-level NVMe driver. In particular, we employed fine-grained resource monitoring and load balancing based on heuristics to optimize I/O performance. OctoKV is deployed on a Linux cluster with Intel SPDK. The extensive evaluation shows that OctoKV achieves lower I/O response times in comparison to traditional approaches where key-value stores run on the client side. Also, the proposed load balancing strategies efficiently enhance I/O response times by equally distributing the workload from overloaded cores to other cores.

Khan, Awais↗

Coordinated distribution network control of tap changer transformers, capacitors and PV inverters

A power distribution system operates most efficiently with voltage deviations along a feeder kept to a minimum and must ensure all voltages remain within specified limits. Recently with the increased integration of photovoltaics, the variable power output has led to increased voltage fluctuations and violation of operating limits. This study proposes an optimization model based on a recently developed heuristic search method, grey wolf optimization, to coordinate the various distribution controllers. Several different case studies on IEEE 33 and 69 bus test systems modified by including tap changing transformers, capacitors and photovoltaic solar panels are performed. Simulation results are compared to two other heuristic-based optimization methods: harmony search and differential evolution. Finally, the simulation results show the effectiveness of the method and indicate the usage of reactive power outputs of PVs facilitates better voltage magnitude profile.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Fast marching methods for the continuous traveling salesman problem

We consider a problem in which we are given a domain, a cost function which depends on position at each point in the domain, and a subset of points ('cities') in the domain. The goal is to determine the cheapest closed path that visits each city in the domain once. This can be thought of as a version of the Traveling Salesman Problem, in which an underlying known metric determines the cost of moving through each point of the domain, but in which the actual shortest path between cities is unknown at the outset. We describe algorithms for both a heuristic and an optimal solution to this problem. The order of the heuristic algorithm is at worst case M * N logN, where M is the number of cities, and N the size of the computational mesh used to approximate the solutions to the shortest paths problems. The average runtime of the heuristic algorithm is linear in the number of cities and O(N log N) in the size N of the mesh.

97 MATHEMATICS AND COMPUTING↗

K-Spin Hamiltonian for Quantum-Resolvable Markov Decision Processes

The Markov decision process is the mathematical formalization underlying the modern field of reinforcement learning when transition and reward functions are unknown. We derive a pseudo-Boolean cost function that is equivalent to a K-spin Hamiltonian representation of the discrete, finite, discounted Markov decision process with infinite horizon. This K-spin Hamiltonian furnishes a starting point from which to solve for an optimal policy using heuristic quantum algorithms such as adiabatic quantum annealing and the quantum approximate optimization algorithm on near-term quantum hardware. In arguing that the variational minimization of our Hamiltonian is approximately equivalent to the Bellman optimality condition for a prevalent class of environments we establish an interesting analogy with classical field theory. Along with proof-of-concept calculations to corroborate our formulation by simulated and quantum annealing against classical Q-Learning, we analyze the scaling of physical resources required to solve our Hamiltonian on quantum hardware.

Hamiltonian↗

Lagrangian relaxation based heuristics for a chance-constrained optimization model of a hybrid solar-battery storage system

Here, we develop a stochastic optimization model for scheduling a hybrid solar-battery storage system. Solar power in excess of the promise can be used to charge the battery, while power short of the promise is met by discharging the battery. We ensure reliable operations by using a joint chance constraint. Models with a few hundred scenarios are relatively tractable; for larger models, we demonstrate how a Lagrangian relaxation scheme provides improved results. To further accelerate the Lagrangian scheme, we embed the progressive hedging algorithm within the subgradient iterations of the Lagrangian relaxation. Lastly, we investigate several enhancements of the progressive hedging algorithm, and find bundling of scenarios results in the best bounds.

14 SOLAR ENERGY↗