SEARCH · Engineering Papers
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.
The Accelerator Controls Operations Research Network (ACORN) project will modernize the accelerator control system and upgrade power supplies to enable future operations of the Fermilab Accelerator Complex with megawatt proton beams. By focusing on MLOps, ACORN enables full integration of AI into the control system to allow safe and reliable improvements to accelerator beam operations.
The Fermilab Main Injector enclosure houses two accelerators, the Main Injector and Recycler. During normal operation, high intensity proton beams exist simultaneously in both. The two accelerators share the same beam loss monitors (BLM) and monitoring system. Beam losses in the Main Injector enclosure are monitored for tuning the accelerators and machine protection. Losses are currently attributed to a specific machine based on timing. However, this method alone is insufficient and often inaccurate, resulting in more difficult machine tuning and unnecessary machine downtime. Machine experts can often distinguish the correct source of beam loss. This suggests a machine learning (ML) model may be producible to help de-blend losses between machines. Work is underway as part of the Fermilab Real-time Edge AI for Distributed Systems Project (READS) to develop a ML empowered system that collects streamed BLM data and additional machine readings to infer in real-time, which machine generated beam loss.
This white paper addresses Focus Area II. We advocate developing a hybrid modeling system to improve the understanding of decadal- and longer-scale predictability of high impact water cycle components. This hybrid model combines a partial differential equation (PDE)-based dynamic core with AI/ML based emulators to represent many of the computationally expensive processes in Earth’s climate models. The hybrid modeling system has the potential to exploit emerging graphics processing unit (GPU)-accelerated architectures and allows for the generation of large ensemble (~1000’s) simulations to better characterize the model uncertainty and understand predictability.
The AI Bus software architecture was designed to support the construction of large-scale, production-quality applications in areas of high technology flux, running heterogeneous distributed environments, utilizing a mix of knowledge-based and conventional components. These goals led to its current development as a layered, object-oriented library for cooperative systems. This paper describes the concepts and design of the AI Bus and its implementation status as a library of reusable and customizable objects, structured by layers from operating system interfaces up to high-level knowledge-based agents. Each agent is a semi-autonomous process with specialized expertise, and consists of a number of knowledge sources (a knowledge base and inference engine). Inter-agent communication mechanisms are based on blackboards and Actors-style acquaintances. As a conservative first implementation, we used C++ on top of Unix, and wrapped an embedded Clips with methods for the knowledge source class. This involved designing standard protocols for communication and functions which use these protocols in rules. Embedding several CLIPS objects within a single process was an unexpected problem because of global variables, whose solution involved constructing and recompiling a C++ version of CLIPS. We are currently working on a more radical approach to incorporating CLIPS, by separating out its pattern matcher, rule and fact representations and other components as true object oriented modules.
Not Available
Over the last decade, Machine Learning (ML) technologies have slowly made their way into the accelerator community. Rapid advances in recent years in deep learning, particularly reinforcement learning for control system applications and the accessibility of deep learning in embedded hardware, have generated renewed interest and spawned a number of applications. The Fermilab Accelerator Complex, shown in Fig. 1, has provided High Energy Physics (HEP) experiments with proton beams for nearly fifty years. The current focus of the laboratory is its world-class experimental program at the intensity frontier. While increasing beam intensity certainly presents its own challenges, preserving beam size while minimizing beam losses – particles lost through interactions with the beam vacuum pipe – turns out to be, in many ways, the main challenge. The accelerator is controlled via a complex system of hundreds of thousands of devices. Enabling fine tuning and real-time optimization of their parameters using ML methods and stepping beyond experience-based reasoning of human operators are key to the success of future intensity upgrades. Our objective will be to integrate ML into accelerator operations and furthermore, provide an accessible framework, which can also be used by a broad range of other accelerator systems with dynamic tuning needs.
Reliable, high-intensity operation of the Fermilab Accelerator Complex is critical to the success of the Long-Baseline Neutrino Facility and Deep Underground Neutrino Experiment. We describe the requirements and infrastructure necessary to support routine use of artificial intelligence and machine learning (AI/ML) in the accelerator control system. Three capabilities are identified: a machine learning operations (MLOps) framework standardizing the lifecycle of AI/ML automation from data management through deployment and monitoring; a data quality framework defining and enforcing standards required to build trustworthy AI/ML applications; and workflow integration with large language models to assist physicists, engineers, and operators with information retrieval, code development, and routine analysis. Use cases spanning beam diagnostics, beam control, and support system automation illustrate the technical requirements across the complex.
Some current approaches to plausible reasoning in artificial intelligence are reviewed and discussed. Some of the most significant recent advances in plausible and approximate reasoning are examined. A synergism among the techniques of uncertainty management is advocated, and brief discussions on the certainty factor approach, probabilistic approach, Dempster-Shafer theory of evidence, possibility theory, linguistic variables, and fuzzy control are presented. Some extensions to these methods are described, and the applications of the methods are considered.
The Multimission VICAR Planner (MVP) system is an AI planning system which constructs.
The Multimission VICAR Planner (MVP) system is an AI planning system which constructs.
Advances in artificial intelligence (A.I.) over the last decade opens new opportunities for NASA to advance the state of the art in autonomous missions beyond low-Earth-orbit. Still, many challenges remain before these technologies can be deployed on manned spacecraft. This presentation summarizes many of these challenges, provides recommendations on how to address these problems, and provides an example of NASA’s Autonomous Power Controller's A.I.-based fault detection system.
Predictive mathematical modeling is an essential part of systems biology and is interconnected with information management. Systems biology information is often stored in specialized formats to facilitate data storage and analysis. These formats are not designed for easy human readability and thus require specialized software to visualize and interpret results. Therefore, comprehending modeling and underlying networks and pathways is contingent on mastering systems biology tools, which is particularly challenging for users with no or little background in data science or system biology. To address this challenge, we investigated the usage of public Artificial Intelligence (AI) tools in exploring systems biology resources in mathematical modeling. We tested public AI’s understanding of mathematics in models, related systems biology data, and the complexity of model structures. Our approach can enhance the accessibility of systems biology for non-system biologists and help them understand systems biology without a deep learning curve.
BACKGROUND Objective Structured Clinical Evaluations (OSCEs) have long been established as a robust methodology for summative assessment of clinical skills and decision-making during medical education. The recent integration of Artificial Intelligence (AI) into clinical decision-making processes has prompted the need for novel evaluation frameworks to assess the efficacy and reliability of AI clinical decision support system (CDSS) tools. This abstract outlines the process of quantitatively evaluating a novel CDSS (“Doc in a Box” Google 2024) trained on curated medical spaceflight data in the psychomotor domain as it interfaces with a human volunteer acting as the crew medical officer (CMO). PURPOSE The AI CDSS under review was developed as part of the Lunar Command and Control Interoperability (LuCCI) project, which is intended to address a gap in how Lunar Surface Systems (LSS) would interoperate across multiple programs, commercial partners, and international partners. The project objective is to define, prototype, integrate, and evaluate an interoperable lunar command, control, data, and software reference architecture to enable autonomy and informatics capability through common standards across LSS. A multi-modal AI-based CDSS compatible with Federated LSS will assist clinicians in diagnosing and managing complex medical conditions by providing evidence-based recommendations through predictive analytics. Given the critical role of decision-support as NASA continues to evolve its Earth-independent medical operations (EIMO), it is imperative to ensure that such AI tools perform reliably and align with clinical standards during progressive lunar and Martian exploration class missions. METHODS The OSCE framework, traditionally used for evaluating human clinicians, was adapted to assess the AI tool's decision-making capabilities in simulated clinical scenarios. In this adapted OSCE, the AI CDSS was tested across a series of structured clinical scenarios designed to mimic real-life spaceflight patient cases. These scenarios included a range of conditions and complexities, allowing for comprehensive assessment of the tool's performance. Key evaluation metrics included accuracy of diagnosis, timeliness of decision-making, and appropriate recommendations for therapies. The OSCE was scored by human physician evaluators who assessed the AI's recommendations in comparison with expert clinicians' medical decision making to ensure alignment with best practices and the standard of care. RESULTS Preliminary results indicate that the AI CDSS demonstrated high accuracy in diagnostic recommendations and decision support across various scenarios. However, certain limitations were noted, such as occasional discrepancies in handling complex or nuanced cases that required a more contextual understanding. Additionally, the tool scored higher on the diagnostic portion of the rubric, with lower scores in the therapeutic recommendations. These findings highlight the importance of continuous refinement and validation of AI tools through rigorous evaluation frameworks like the OSCE. The adaptation of OSCEs for AI tools presents several advantages, including a structured and reproducible approach to evaluation, the ability to test AI systems in diverse clinical scenarios, and the opportunity to benchmark AI performance against established clinical standards to permit charting of future progress as aerospace medicine evolves as a discipline. Remaining challenges include ensuring that these evaluations capture the full spectrum of clinical decision-making scenarios that will be confronted by CMOs during missions and adequately reflecting real-world variability of the austere spaceflight environment. CONCLUSION Employing OSCEs to evaluate AI clinical decision support tools offers a promising approach to validating their clinical utility and efficacy. This methodology not only provides insights into the tool's performance but also fosters ongoing improvement and alignment with standard of care practices. Future research should focus on refining these evaluation processes and addressing limitations to enhance the integration of AI tools in clinical spaceflight settings. REFERENCES Scott S, Hearns V, Barker MA. Testing Clinical Skills: A Look at the OSCE and USMLE Clinical Skills Exams. S D Med. 2019 Oct;72(10):451-453. Majumder MAA, Kumar A, Krishnamurthy K, Ojeh N, Adams OP, Sa B. An evaluative study of objective structured clinical examination (OSCE): students and examiners perspectives. Adv Med Educ Pract. 2019 Jun 5;10:387-397. Karam VY, Park YS, Tekian A, Youssef N. Evaluating the validity evidence of an OSCE: results from a new medical school. BMC Med Educ. 2018 Dec 20;18(1):313.
Recent advances in artificial intelligence (AI) and development of large language models (LLMs) present the opportunity to develop a new generation of power systems applications. In contrast with early power system AI applications based on structured numerical data, LLMs offer unique capabilities to perform logical reasoning using text documents, unstructured data, and application programming interface (API) calls to computational software. This paper seeks to bridge the knowledge gap between power systems engineers and LLM developers through a crosscutting explanation of use cases, characteristics, requirements, practical considerations from the perspectives of both LLM capabilities and industry needs. Specific focus is given to applications that can be realistically deployed by electric utilities. After introducing the architecture of LLMs and unique challenges of the power systems domain, this paper proposes twenty representative LLM applications grouped into categories of 1) power system operations, 2) asset management, 3) system planning and analytics, and 4) energy management and protection systems. Five use cases are presented within each category with descriptions of the motivation, objectives, approaches, example inputs / outputs, and benefits of each use case.
We develop AI-driven frameworks to characterize urban energy systems with the goal of transforming planning by reducing the labor of model generation, scaling scenario exploration, and improving accuracy for localized analysis. The approach integrates top-down and bottom-up data to train different AI models that predict missing information and generate inputs and targeted scenarios for district-scale models. The result is a scalable framework that provides actionable insights for reliable and efficient planning.
An overview is presented of the Hubble Space Telescope (HST) Service Mission Planning and Replanning Tool (SM/PART) and the approach that was employed to develop the system. SM/PART is an AI expert system utilized by NASA to rapidly build complex timelines and command plans to control HST activities. HST servicing missions are Space Shuttle/Orbiter missions expected to occur every three to five years to rendezvous with the HST in orbit and replace, upgrade, or repair HST components. SM/PART runs on a UNIX-based workstation and is implemented in the C language with a Motif-based graphical user interface.
This paper will discuss recent operational successes in implementing expert systems to support the complex functions of NASA mission control systems at the Goddard Space Flight Center, including fault detection and diagnosis for real time and engineering analysis functions in the Cosmic Background Explorer and Gamma Ray Observatory missions and automation of resource planning and scheduling functions for various missions. It will also discuss ongoing developments and prototypes that will lead to increasingly sophisticated applications of artificial intelligence. These include the use of neural networks to perform telemetry monitoring functions, the implementation of generic expert system shells that can be customized to telemetry handling functions specific to NASA control centers, the applications of AI in training and user support, the long-term potential of implementing systems based around distributed, cooperative problem solving, and the use of AI to control and assist system development activities.
In rule-based AI planning, expert, and learning systems, it is often the case that the left-hand-sides of the rules must be repeatedly compared to the contents of some 'working memory'. The traditional approach to solve such a 'match phase problem' for production systems is to use the Rete Match Algorithm. Here, a new technique using a multilayer perceptron, a particular artificial neural network model, is presented to solve the match phase problem for rule-based AI systems. A syntax for premise formulas (i.e., the left-hand-sides of the rules) is defined, and working memory is specified. From this, it is shown how to construct a multilayer perceptron that finds all of the rules which can be executed for the current situation in working memory. The complexity of the constructed multilayer perceptron is derived in terms of the maximum number of nodes and the required number of layers. A method for reducing the number of layers to at most three is also presented.