SEARCH · Engineering Papers
Results for “communication constraints”
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.
FIRM: federated image reconstruction using multimodal tomographic data
Here, we propose a federated algorithm for reconstructing images using multimodal tomographic data sourced from dispersed locations, addressing the challenges of traditional unimodal approaches that are prone to noise and reduced image quality, as well as the limitations of centralized multimodal approaches that require extensive data transfer, leading to significant communication overhead, storage demands, and potential data privacy concerns. Our approach formulates a joint inverse optimization problem incorporating multimodality constraints and solves it in a federated framework through local gradient computations complemented by lightweight central operations, thereby ensuring data decentralization. Leveraging the connection between our federated algorithm and the quadratic penalty method, we introduce an adaptive step-size rule with guaranteed sublinear convergence. Numerical results demonstrate superior computational efficiency and improved image reconstruction quality compared to existing approaches.
Enabling Real-Time Communication in Multi-Agent Systems: A Graph Neural Network Based Approach
Global connectivity enables effective coordination in Multi-Agent Systems (MAS). Solving these connection problems under hardware constraints is an NP-hard non-Euclidean Degree Constrained Minimum Spanning Tree (DCMST) problem. Prior MAS controllers coordinate team movement for task completion and collision avoidance; some considering Line-of-Sight (LOS) maintenance but prioritizing flexibility over guarantees. Evolutionary Algorithms (EA) have been shown to find good solutions for DCMST, but their performance degrades with larger populations required to support a large MAS. We present a method based on edge graph attention networks, trained offline to reduce online computation times. Empirical comparisons with greedy polynomial-time solvers and EA show that our method leverages latent graph information to consistently find constraint-satisfying solutions in less time.
Distributed Coordination of Demand-side Flexible Resources in Microgrid with All-Time Feasibility
The prevalence of distributed renewable generators motivates microgrid operators to exploit demand-side flexible resources (DFRs). Due to their dispersed nature, distributed DFR coordination has been a vibrant research area, while there are several issues awaiting to be addressed. On one hand, DFR power is internally coupled through power flow, while DFR usually cannot access grid information. On the other hand, in time-restricted scenarios, solution feasibility cannot be guaranteed by conventional dual-based algorithms. To fill these gaps, we propose a distributed DFR coordination framework with all-time feasibility. The proposed framework accounts for the distinct access of microgrid entities to grid information. A distributed and all-time feasible algorithm is proposed for optimal DFR coordination, which allows DFRs to make local decisions without violating constraints throughout iterations. The effectiveness of the proposed algorithm is demonstrated through case studies. The impact of peer-to-peer communication links on algorithm convergence is also investigated, which emphasizes the balance between communication investment and algorithm performance.
Demonstration of a Novel Technology to Manage Electricity Demand in Grid-Independent Military Microgrids
This research was conducted by the National Renewable Energy Laboratory (NREL) in collaboration with the S&C Electric Inc. through funding provided by the ESTCP. The project demonstrates use of cybersecure Automated Demand Response (ADR) technology to effectively manage microgrid loads during grid-independent, also known as "islanded," operation. When military microgrids become isolated from the main electrical grid, they are required to balance electricity supply and demand locally. Given that local generation may be constrained, the prevailing strategy involves shedding all but the most critical loads by tripping smart circuit breakers, which then necessitate manual resetting. This approach is generally implemented at the building level, which means that the buildings with mission-critical activities are exempt from load management and remain fully powered, whereas those deemed non-critical can experience a complete loss of service. In this research we developed a method that allows building automation systems to selectively control their assets in response to load shedding request from a microgrid controller, avoiding total loss of service in contrast to the conventional control approach. A commercial OpenADR client server by GridFabric is used for communication between the microgrid controller and the building management system (BMS). The microgrid controller monitors both generation capacity and various assets within the microgrid and issues a demand reduction request when necessary. This request is communicated to the OpenADR server via Modbus. Upon receiving the request, the OpenADR server forwards it to the BMS utilizing the OpenADR protocol. The BMS is pre-configured with various levels of load reduction strategies based on the controllable assets available, allowing for a nuanced approach to demand reduction. Both lab and field tests were performed that considered load shedding needed to achieve closed transition into island mode and to accommodate changing loads and power source availability while islanded. A commercial microgrid controller was used for these tests with normal programming within the expected constraints of the system capabilities. That is, the solution did not require any specialized modification to the code base of the controller. Given the latency of the round-trip communication path between the microgrid controller and the various devices involved with the load shed processes, there are certain scenarios for which the demonstrated solution are appropriate and some which are not. The methods described in this report can be used for load shedding/restoration during transitions between islanded and grid-tied modes of operation, as well as accommodating normal variations in load and the need to remove a power source from operation for maintenance. These methods should not be used for scenarios that require load shedding within a second or two such as sudden and unanticipated significant load increases or loss of power sources through equipment faults.
Toward a persistent event-streaming system for high-performance computing applications
High-performance computing (HPC) applications have traditionally relied on parallel file systems and file transfer services to manage data movement and storage. Alternative approaches have been proposed that use direct communications between application components, trading persistence and fault tolerance for speed. Event-driven architectures, as popularized in enterprise contexts, present a compelling middle ground, avoiding the performance cost and API constraints of parallel file systems while retaining persistence and offering impedance matching between application components. However, adapting streaming frameworks to HPC workloads requires addressing challenges unique to HPC systems. This paper investigates the potential for a streaming framework designed for HPC infrastructures and use cases. We introduce Mofka, a persistent event-streaming framework designed specifically for HPC environments. Mofka combines the capabilities of a traditional streaming service with optimizations tailored to the HPC context, such as support for massively multicore nodes, efficient scaling for large producer-consumer workflows, RDMA-enabled high-performance network communications, specialized network fabrics with multiple links per node, and efficient handling of large scientific data payloads. Built using the Mochi suite of HPC data service components, Mofka provides a lightweight, modular, and high-performance solution for persistent streaming in HPC systems. We present the architecture of Mofka and evaluate its performance against Kafka and Redpanda using benchmarks on diverse platforms, including Argonne's Polaris and Oak Ridge's Frontier supercomputers, showing up to 8× improvement in throughput in some scenarios. We then demonstrate its utility in several real-world applications: a tomographic reconstruction pipeline, a workflow for the discovery of metal-organic frameworks for carbon capture, and the instrumentation of Dask workflows for provenance tracking and performance analysis.
Model-Free Control of Grid-Interactive Efficient Buildings Under Communication Time Delays
Grid-interactive efficient buildings (GEBs) have recently been used to enhance the reliability and stability of the electric grid through demand response (DR) programs. However, most existing DR control strategies require accurate modeling of the various building thermostatically controlled loads (TCLs) and are computationally expensive. To address these challenges, a model-free control (MFC)-based strategy has recently been introduced for coordinating and controlling GEBs. MFC is a data-enabled control strategy that is computationally efficient and does not require the analytical models of the various building equipment. In this paper, we numerically investigate the impact of communication time delays on the performance of MFC in maintaining the TCLs' temperatures within the desired comfort levels while meeting the assigned power allocation constraint.
A Review of Software for Designing and Operating Quantum Networks
Quantum networks development is crucial to realizing a production-grade network that can support distributed sensing, secure communication, and utility-scale quantum computation. However, the transition from laboratory demonstration to deployable networks requires software implementations of architectures and protocols tailored to the unique constraints of quantum systems. This paper reviews the current state of software implementations for quantum networks, organized around a three-plane abstraction of infrastructure, logical, and control/service planes. We cover software for both designing quantum network protocols (e.g., SeQUeNCe, QuISP, and NetSquid) and operating testbeds, with a focus on essential control/service plane functions such as entanglement, topology, and resource management, in a proposed taxonomy. Our review highlights a persistent gap between theoretical architecture and protocol proposals and their realization in simulators or testbeds, particularly in dynamic topology and network management. We conclude by outlining open challenges and proposing a roadmap for developing scalable software architectures to enable hybrid, large-scale quantum networks.
Deploying MARTe2 and MDSplus for Scalable Real-Time Control Systems: A Proof-of-Concept for the SMART Tokamak
The SMART tokamak is a spherical tokamak with flexible shaping that will need a modular and high-performance real-time control system. This work presents the implementation and validation of the MARTe2 framework at the Princeton Plasma Physics Laboratory (PPPL) as a testbed for supporting SMART (University of Seville, Spain). The system integrates MARTe2, MDSplus for data archiving, and user datagram protocol (UDP)-based communication with both synthetic and physical devices. A deterministic control loop was developed using custom control algorithms, showing the system’s ability to operate under strict real-time constraints. The resulting framework is scalable, maintainable, and adaptable to experimental tokamaks like SMART.
Integration of 5G and Time Sensitive Networks in Fossil Energy Generation Systems: A Case Study
Precise timing and data transmission within stringent time constraints are critical for numerous applications, such as robotics, virtual and augmented reality, industrial automation, energy and medical business, and various other sectors. Time-sensitive networks (TSN) and fifth-generation wireless communications (5G) are crucial for industrial communications, enabling convergent communication for various services using a common network core. Applications that are time-sensitive and require deterministic communications with low latency fall into this category, such as generation plant systems. This paper presents a simulated model of 5G-TSN for a fossil power plant based on the wired network parameters implemented in situ. Metrics analysis, comparison and future work are presented. © 2024 IEEE.
Stripline Block Sensor System
The Long Baseline Neutrino Facility requires a secondary sensor system to aid in the installation and replacement of the stripline blocks. These blocks are responsible for delivering large amounts of current to the horn and must be installed with high precision to avoid possible physical damage. There are many constraints due to the modules' surroundings, such as small clearances and a highly radioactive environment. Existing systems have already been implemented within the process; however, these systems offer limited information, leaving much risk during operations. This calls for a sufficiently accurate sensor system, which has a small space requirement and provides valuable information throughout the replacement period. A variety of methods and technologies were explored to provide an applicable approach to the issue. The proposed solution consisted of the use of retroreflective and target, however, it cannot provide positional feedback throughout the entire process. A proposed alternative utilizes a three-dimensional indoor positioning system to provide the operator with real-time positioning and rotational information. Sensors placed on the module and stripline would communicate with fixed receivers surrounding the target hall to provide precise positional feedback. However, this solution may still contain line-of-sight issues as well as installation complications due to the radioactive environment. Future work would consist of finding vendors capable of providing technology that fits our constraints, proof of concept tests for both solutions, implementation planning, and trials within the target hall itself. Due to its ability to provide continuous positional and rotational feedback, the indoor positioning system is the most suitable solution for accurately installing the stripline block.
DP-TwoLevel: two-stage gradient subspace learning for differentially private federated learning
Federated learning (FL) enables collaborative model training across distributed data sources without sharing raw data, but faces fundamental challenges in communication efficiency and privacy. Differentially private (DP) training mitigates information leakage but introduces noise that degrades model performance, especially in high-dimensional settings. We propose DP-TwoLevel, a hierarchical gradient projection method that improves utility under fixed DP constraints by exploiting low-dimensional structure in model updates. Our approach learns a two-level PCA-based representation of gradients and applies DP noise in a reduced-dimensional subspace, thereby lowering the effective noise magnitude while preserving dominant signal components. We evaluate the method across three datasets (MNIST, Fashion-MNIST, CIFAR-10) and three privacy regimes (ϵ∈0.5, 1.0, 2.0). Across nine experimental settings, DP-TwoLevel consistently outperforms DP-FedAvg, achieving an average accuracy improvement of 9.44%, with larger gains observed in lower ϵ(higher-noise) regimes (up to +22.31%). We further analyze scalability across models ranging from 100K to 1.49M parameters and identify a variance-based success criterion: performance remains strong when the projection preserves more than 75% of gradient variance, degrades in a marginal regime (65–75%), and fails below this threshold. Our results demonstrate that structure-aware dimensionality reduction can significantly improve the privacy–utility tradeoff in FL without modifying formal privacy guarantees. We also provide empirical evidence of scaling limitations for global projections and motivate per-layer extensions for larger models.
PowerMappeR: Power-Optimized Mapping of SNNs onto ReRAM Crossbars coupled via Packet-Switched NoCs
Many recent efforts in developing hardware-accelerated spiking neural networks (SNNs) are characterized by deep co-design between algorithms, architectures, and devices. Architectural advances overcome device constraints by coupling together many small resistive-RAM (ReRAM) crossbars via a network-on-chip (NoC) for neuromorphic component operation. Concurrently, improved SNN training methods increase accuracy and structural sparsity in networks despite growing problem sizes. Finally, compilers leverage these attributes to minimize area and inter-crossbar communication while mapping large SNNs to sophisticated architectures. However, for compiler-driven co-design to realize increasingly complex and profitable optimizations, a compile-time view of power consumption is critical. We present PowerMappeR to express and optimize over mapping-, architecture-, and device-specific power consumption information. By modeling the dynamic power of well-established components, we develop an integer linear programming (ILP)-based, encoding-agnostic, parametric power estimation model. Using this model, we demonstrate practical improvements in area and inter-crossbar communication by 0%–9.5% and 1.4%–5.1%, respectively. We also limit hotspot formation during optimization, achieving comparable or better results in targeted metrics with up to 96.4%–97.1% restriction of hotspot magnitude. Finally, we introduce profile-guided formulations to reduce worst-case and expected-case hotspot magnitude by 40.7%–69.5% and 40.6%–56.3%, respectively. Optimizing worst-case hotspot magnitude incidentally improves expected-case magnitude by 10.85%–33.45%. Reciprocally, optimizing expected-case magnitude incidentally improves worst-case magnitude by 4.33%–39.87%. Validation against hardware simulators confirms that PowerMappeR can decrease dynamic power consumption by 12.6%–27.3%.
Privacy-Preserving Control of Partitioned Energy Resources
Distributed energy resources are an increasingly important part of the electric grid. We examine the problem of partitioning a distributed energy resource among many users while providing privacy to them. In this model, clients can send requests to a server, the server can verify that the requests are valid and aggregate them, but it cannot see the actual values in the requests. Without privacy, each user is forced to reveal their daily schedule or energy use. Energy resources add a novel challenge that prior systems do not address: they require verifying limits on private power (a rate over time) and energy (a sum) values. Furthermore, the cryptographic mechanisms must run on embedded energy control systems. We describe Weft, a novel cryptographic system that verifies both power (rate) and energy (integral) constraints on private client values and aggregates them. The key insight behind the approach is to rely on additively homomorphic secret shares, which allows servers to compute sums from rates. We present 3 cryptographic proof systems with different system trade-off for embedded systems: bit-splitting proofs minimize memory use, sorting proofs minimize computation, and commitment proofs minimize network communication. Using bit-splitting proofs, it takes an IoT client using a CortexM microcontroller 4 minutes of compute time to privately control its share of an energy resource for a day at 20s granularity.
Network Slicing for Federated Learning in Operational Technology Environment
Industrial Control Systems (ICS) and Supervisory Control and Data Acquisition (SCADA) environments are essential to modern infrastructure, facing challenges in ensuring low-latency, high-throughput communication while mitigating cyber threats. This paper presents a framework integrating Federated Learning (FL) and network slicing with Quality of Service (QoS) to enable real-time monitoring without disrupting OT operations. Leveraging digital twin technology and Network Function Virtualization (NFV), the architecture supports predictive analytics and Industry 4.0 requirements. FL facilitates decentralized model training, preserving data privacy and scalability, though it introduces potential throughput constraints. Network slicing addresses this by creating dedicated virtualized segments optimized for performance and security. Advanced fault tolerance at the container and instance levels enhances system reliability. The proposed architecture ensures high throughput, low latency, and secure orchestration for real-time anomaly detection in OT networks. Performance evaluations validate its efficiency in throughput, deployment, and learning accuracy, providing a robust foundation for future ICS automation and data-driven decision-making.
A framework and tool for designing cost-effective, resilient, and circular net-zero supply chains under uncertainty with an application to multilayer plastic films
While 55% of Fortune 500 companies have committed to achieving net-zero emissions and/or zero-waste operations by 2035, only 2% are currently on track, revealing a critical gap between ambition and action. Designing supply chains that reduce both emissions and waste is a complex non-intuitive, multi-objective challenge, compounded by the high costs of new technologies and the need for resilient, profitable solutions. This paper aims to address this challenge by presenting a generic framework and multi-objective optimization formulation for designing cost-effective, circular, and resilient supply chains under uncertainty, implemented through a user-friendly decision-support tool with intuitive data visualization capabilities, enabling communication of results to both technical and non-technical stakeholders. We demonstrate the application of this framework in the context of multilayer plastic films (barrier films), which are widely used in food packaging and composite materials. The model quantifies trade-offs across three objectives: minimizing global warming potential, maximizing circularity, and minimizing cost. A key contribution of this work is the explicit modeling of technological resilience, the ability of supply chains to maintain function under disruption. In the cost-minimization case, the resilience constraint makes the design approximately three times more expensive in the short-term metric, but shifts the system from relying on a single recovery pathway to a portfolio of four recovery pathways, improving the robustness of the optimization solution under uncertainty. Lastly, we introduce TranZero, a decision-support tool that integrates material flow analysis, hotspot identification, and optimization-based scenario planning to support net-zero and circularity decisions.
Scalable Circuit Cutting and Scheduling in a Resource-constrained and Distributed Quantum System
Despite quantum computing's rapid development, current systems remain limited in practical applications due to their limited qubit count and quality. Various technologies, such as superconducting, trapped ions, and neutral atom quantum computing technologies are progressing towards a fault tolerant era, however they all face a diverse set of challenges in scalability and control. Recent efforts have focused on multi-node quantum systems that connect multiple smaller quantum devices to execute larger circuits. Future demonstrations hope to use quantum channels to couple systems, however current demonstrations can leverage classical communication with circuit cutting techniques. This involves cutting large circuits into smaller subcircuits and reconstructing them post-execution. However, existing cutting methods are hindered by lengthy search times as the number of qubits and gates increases. Additionally, they often fail to effectively utilize the resources of various worker configurations in a multi-node system. To address these challenges, we introduce FitCut, a novel approach that transforms quantum circuits into weighted graphs and utilizes a community-based, bottom-up approach to cut circuits according to resource constraints, e.g., qubit counts, on each worker. FitCut also includes a scheduling algorithm that optimizes resource utilization across workers. Implemented with Qiskit and evaluated extensively, FitCut significantly outperforms the Qiskit Circuit Knitting Toolbox, reducing time costs by factors ranging from 3 to 2000 and improving resource utilization rates by up to 3.88 times on the worker side, achieving a system-wide improvement of 2.86 times.
Performance and Achievable Rates of the Gottesman-Kitaev-Preskill Code for Pure-Loss and Amplification Channels
Quantum error-correction codes protect information from realistic noisy channels and lie at the heart of quantum computation and communication tasks. Understanding the optimal performance and other information-theoretic properties, such as the achievable rates, of a given code is crucial, as these factors determine the fundamental limits imposed by the encoding in conjunction with the noise channel. Here, we use the transpose channel to analytically obtain the near-optimal performance of any Gottesman-Kitaev-Preskill (GKP) code under pure loss and pure amplification. We present rigorous connections between GKP code’s near-optimal performance and its dual lattice geometry and average input energy. With no energy constraint, we show that when |𝜏/(1−𝜏)| is an integer, specific families of GKP codes simultaneously achieve the loss and amplification capacity. 𝜏 is the transmissivity (gain) for loss (amplification). Our results establish GKP code as the first structured bosonic code family that achieves the capacity of loss and amplification.