Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Anomaly Detection”

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 1,045 records · Page 58

A Distributed Trust Model Simulator for Energy Grid of Things Distributed Energy Resource Management System

The evolution of networks into more distributed, self-reliant nodes has mitigated single-point failures that plagued traditional centralized networks. Applied to power grids, distributed systems can increase the integrity and availability of grid services while also offering a power management solution. However, while distributed networks provide scalability, security, and sustainability compared to centralized networks, their distributed nature makes them harder for anomaly detection and prevention. Incorporating a Distributed Trust Model (DTM) System into an Energy Grid of Things Distributed Energy Resource Management System (EGOT DERMS) allows grid participants to be characterized and their communication to be analyzed for possible attacks. A Trust Model simulator is needed to evaluate and improve the DTM System.Trustworthiness is calculated using a Trust Model. While many trust models exist, most only consider 2-3 matrices to evaluate trust. The TM proposed in this thesis uses a Metric Vector of Trust (MVoT) monitoring 17 parameters when assessing trust. Moreover, unlike standard trust models, the proposed trust model establishes a method to test the trust between various actors within the network and probe the trust model itself. Using a Trust Model Simulator, MVoT calaculations, initial values, and parameters are fine-tuned to achieve high-confidence message classifications and minimize false positives. The DTM System and Trust Mode Simulation Suite allow for distributed trust evaluation with a real-time classification of EGOT DERMS actors, providing additional security for distributed systems.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Open-source FPGA-ML codesign for the MLPerf Tiny Benchmark

We present our development experience and recent results for the MLPerf Tiny Inference Benchmark on field-programmable gate array (FPGA) platforms. We use the open-source hls4ml and FINN workflows, which aim to democratize AI-hardware codesign of optimized neural networks on FPGAs. We present the design and implementation process for the keyword spotting, anomaly detection, and image classification benchmark tasks. The resulting hardware implementations are quantized, configurable, spatial dataflow architectures tailored for speed and efficiency and introduce new generic optimizations and common workflows developed as a part of this work. The full workflow is presented from quantization-aware training to FPGA implementation. The solutions are deployed on system-on-chip (Pynq-Z2) and pure FPGA (Arty A7-100T) platforms. The resulting submissions achieve latencies as low as 20 $\mu$s and energy consumption as low as 30 $\mu$J per inference. We demonstrate how emerging ML benchmarks on heterogeneous hardware platforms can catalyze collaboration and the development of new techniques and more accessible tools.

Borras, Hendrik↗

Multi-Kernel Adaptive Support Vector Machine for Scalable Predictive Maintenance

Application of data-driven solutions across an industry is challenging, since the data are often stored locally, and increasing privacy and security concerns restrict access to the data. In addition, it is highly unlikely that all potential data patterns are captured in a single data source. Because it is highly unlikely that all potential data patterns are captured in a single data source, machine learning (ML) models developed from a single source cannot be robust enough. An alternative is to train the ML model at each source and develop a distributed knowledge discovery and aggregation approach to build global knowledge. In this paper, we develop and demonstrate a distributed ML model, federated transfer learning (FTL), using a multi-kernel-based adaptive support vector machine (MK-A-SVM). For federated learning (FL), the multi-kernel (MK) approach enables feature-specific model aggregation under data heterogeneity; whereas for transfer learning (TL) the adaptive model enables utilization of an aggregated model from a different task. The proposed approach is validated using nuclear power plant (NPP) vertical motor-driven pump data to predict the health condition of vertical motor-driven pumps as an anomaly detection. The efficiency of the proposed approach is also quantified and compared with neural network.

42 ENGINEERING↗

Progress on Machine Learning for the SNS High Voltage Converter Modulators

The High-Voltage Converter Modulators (HVCM) used to power the klystrons in the Spallation Neutron Source (SNS) linac were selected as one area to explore machine learning due to reliability issues in the past and the availability of large sets of archived waveforms. Progress in the past two years has resulted in generating a significant amount of simulated and measured data for training neural network models such as recurrent neural networks, convolutional neural networks, and variational autoencoders. Applications in anomaly detection, fault classification, and prognostics of capacitor degradation were pursued in collaboration with the Jefferson Laboratory, and early promising results were achieved. This paper will discuss the progress to date and present results from these efforts.

Pappas, Chris↗

Situational Awareness of Grid Anomalies (SAGA)

The modern power industry becomes more vulnerable to cyber events due to the growing interconnectivity, interdependence, and complexity of the electric power grid. High-fidelity modeling and simulation tools that support the preventative risk analysis on potential cyber-relevant events is essential for ensuring the situational awareness of the system operator as it provides an inexpensive and risk-free environment to test the system responses under various cyber-relevant events and hereby can support research on cyber anomaly detection, optimal protective resource allocation, and mitigation measures. In this webinar, we will share NREL's cybersecurity research capabilities by highlighting the development of a scalable cyber-physical event test bed and demonstration with real hardware in the loop. The developed cyber-physical event test bed is backboned by an integrated transmission, distribution, and communication dynamic co-simulation framework and a plug-and-play cyber event generation module. It is designed to be modular and compatible with parallel computing, and thereby supports large-scale system simulations at an affordable computation cost. The test bed can capture millisecond-to-minutes dynamic frequency and voltage responses under cyber events from the bulk transmission system to the active distribution systems and distributed energy resources at the grid edge.

co-simulation↗

Machine Learning for Improved Availability of the SNS Klystron High Voltage Converter Modulators

Beam availability has increased at the SNS, however, the targeted availability is greater than 95 %, while the SNS has failed to meet lower targets in the past. The HVCM used to power the linac klystrons have been one source of lost beam time and was chosen to explore using AI/ML techniques to improve reliability. Among the possibilities being explored are automating the tuning of HVCMs and predicting component failures such as capacitor aging, rectifier assemblies containing hundreds of diodes, and insulating oil degradation. The methodology pursued includes data cleaning, de-noising, post-analysis data labeling, and machine learning model development. We explore using Long Short-Term Memory and autoencoders for anomaly detection and prognostication used to schedule maintenance. We evaluate the use of model regularizers and constraints to improve the performance of the model and investigate methods to estimate the uncertainty of the models to provide a robust prediction with statistical interoperability. This paper describes the operational experience and known failures of the HVCMs and the proposed ML methodology and the preliminary results of training the AI/ML algorithms.

Pappas, Chris↗

Application of Margin-Based Methods to Assess System Health

Health management of complex systems such as nuclear power plants is an essential task to guarantee system reliability. This task can be greatly enhanced by constantly monitoring asset status and performances and process such data (through anomaly detection, diagnostic, and prognostic computational algorithms) to identify asset degradation trends and faulty states. While such information and data are typically available for many of the assets, they are not propagated from the asset to the system level in order to identify the most critical assets and prioritize maintenance and surveillance activities. The main reason is driven by the fact that current reliability modeling techniques are inadequate to process such information/data. This is due to the nature of these techniques which are based on the concept of failure rate/probability that do not serve an operational context where quantitative asset health information is available. Simply stated, current reliability techniques serve a run-to-failure operational setting and not a predictive maintenance one where the goal is to perform maintenance and surveillance activities only when they are needed based on asset health. The risk informed asset management (RIAM) project is focusing on the development of a different kind of reliability modeling techniques designed to adequately serve a predictive operational setting. Such reliability techniques move aways from a failure rate/probability to a margin-based mindset where margin is here used as a metric to quantify asset health based only on current and past operational experience of the asset under consideration. In addition, margin-based reliability techniques are able to propagate asset health information from the component to system level and provide importance measure to each asset. This report summarizes a recent activity performed in collaboration with plant modernization pathway designed to integrate monitoring data into margin-based reliability models. Such activity focuses on a specific system of an existing nuclear power plant where large amount of historic monitoring data is used to monitor asset and system health.

97 MATHEMATICS AND COMPUTING↗

Artificial Intelligence Application to D and D - 20492

As aging facilities across the DOE complex await decommissioning, there is an ongoing need to understand any changes in the structural conditions. Many of these facilities were built over 50 years ago and, in some cases, these facilities have gone beyond the expected operational lifetime. Many facilities have been placed in a state of 'cold and dark,' sitting unused and awaiting decommissioning. Especially challenging are the aging facilities that provide unique operational/production capabilities to support critical DOE missions and cannot be shut down. In any of these scenarios, the structural integrity of these facilities may become compromised as time passes. It is critical that adequate inspections be performed on a continual basis and that the data collected undergoes sufficient analysis to support timely identification of any new or worsening structural issues as well as prompt needed maintenance and repairs to maintain the facilities in a safe condition. In recent days, Artificial Intelligence (AI) [1] and its application to various domains are growing at fast speed. FIU is performing research in this area and exploring the associated technologies to solve nuclear decommissioning problems. Artificial intelligence refers to the capability of a program to autonomously act, react and adapt to the working environment. AI enables the machine to behave like humans and perform the cognitive functions such as 'learning' and 'problem solving'. AI systems gradually moving from traditional approaches (algorithms and expert systems) towards more efficient and advanced technologies (machine learning [1] and deep learning [2] [3]). AI is the study of algorithms and statistical models that is being used by computers to perform specific tasks without using explicit instructions. FIU is working to develop a pilot-scale infrastructure to implement structural health monitoring using AI technologies with focus on machine learning, deep learning. This research is focused on Computer Vision/Image Classification area of AI applications. This can also be expanded to other areas of AI related to Object Recognition and Character Recognition in images. In addition to utilizing existing data sets, FIU will collect and investigate image and video data using FIU test-bed mockups to monitor structural health of the facility. Resulting data will be processed and analyzed using machine learning/deep learning technologies. The proposed pilot system is intended to serve as a starting point to engage the DOE field sites on related data sets and their decision making needs. It is anticipated that proposed machine learning/deep learning technologies can be effectively employed using anomaly detection to solve EM challenges in surveillance and maintenance of the D and D facilities. FIU will work with research stakeholders to identify applications at various sites and other DOE facilities. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Big Data and AI at DoE's Legacy Sites - 20546

More than 30 years have passed since DOE started the decommissioning of nuclear weapon complexes and the clean-up of soil and groundwater. All the sites have been collecting and archiving soil and groundwater monitoring datasets; particularly contaminant concentration time-series. These datasets provide unparalleled opportunities to understand the system behavior (including more fundamental hydrological and geochemical processes, the response to various perturbations, the long-term trend and environmental decay rate towards the regulatory limit). This understanding is critical for providing multiple lines of evidences that can support site closure. In this study, we explore the machine learning (ML) and artificial intelligence (AI) applications to the long-term soil and groundwater management at DoE's legacy sites. ML can improve our understanding of the subsurface systems, which is critical for long-term monitoring and management of the sites, while AI can automate or support some of decision-making processes (e.g., anomaly detection, monitoring well placements). The particular focuses are to develop general algorithms to: (1) to identify distinct spatiotemporal patterns and to identify several groups that have similar temporal behaviors, using unsupervised clustering methods, (2) identify the different temporal scales of hydrological responses to climate perturbations by time-series analysis, and (3) reduce the number of monitoring wells by identifying the minimum sufficient number of wells to capture the heterogeneity of the groundwater contaminant plume and concentration distribution, using the Gaussian Process model. We demonstrate our methodology at the Savannah River Site F-Area. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Determining appropriate data analytics for transformer health monitoring - 107

Transformers are vital assets for the safe, reliable and cost-effective operation of nuclear power plants. The unexpected failure of a transformer can lead to different consequences ranging from a lack of export capability, with the corresponding economic penalties, to catastrophic failure, with the associated health, safety and economic effects. Condition monitoring techniques examine the health of the transformer periodically, with the aim to identify early indicators of anomalies. However, many transformer failures occur because diagnostic and monitoring models do not identify degraded conditions in time. Therefore, health monitoring is an essential component to transformer life cycle management. Existing tools for transformer health monitoring use traditional dissolved gas analysis based diagnostics techniques. With the advance of prognostics and health management (PHM) applications, we can enhance traditional transformer health monitoring techniques using PHM analytics. The design of an appropriate data analytics system requires a multi-stage design process including: (i) specification of engineering requirements; (ii) characterization of existing data sources and analytics to identify complementary techniques; (iii) development of the functional specification of the analytics suite to formalize its behavior, and finally (iv) deployment, validation, and verification of the functional requirements in the final platform. Accordingly, in this paper we propose a transformer analytics suite which incorporates anomaly detection, diagnostics, and prognostics modules in order to complement existing tools for transformer health monitoring. (authors)

42 ENGINEERING↗

System and method for in situ inspection of defects in additively manufactured parts using high speed melt pool pyrometry

A system and method is disclosed for detecting anomalies in an additively manufactured part. An energy source generates a signal forming an optical beam for creating a melt pool in a layer of feedstock material being selectively fused to make a part in an additive manufacturing operation. A sensor is configured to receive a signal reflected from the melt pool. The reflected signal forms a thermal signal indicative of a temperature of the feedstock material at a known location on a layer of the feedstock material while the feedstock material is being fused at the known location. A controller receives and analyzes data relating to the received signal to determine if an anomaly exists at the known location.

Forien, Jean-Baptiste↗

Cybersecurity for Distributed Energy Resources: All Hazards Approach for Grid Modernization

Distributed energy resources (DER) sit at the intersection of IoT and critical infrastructure. As we work towards clean energy and decarbonization targets, high rates of growth of DER are expected, making them an important component of generation and grid services. They have been a part of the explosion of internet connected devices developed to solve real-world problems and make life easier by providing clean, local energy and enable the smart management of grid services. However, the quick-to-market, low-cost drivers for DER, combined with a historically relative low-impact has meant that cybersecurity has not been a top priority. So as DER penetration increases, will DER be a part of the growing challenge of securing the grid, or part of the solution for necessary grid modernization? In this talk, we will discuss the factors that have led to our current position and what makes cybersecurity for DER unique. A wide range of research is conducted at national labs, in partnership with industry and academia, to inform everything from standards and regulation, to the development of cyber-physical anomaly detection through use of digital twins, and even advanced threat analytics to SBOM and HBOM tracking. Even with this multi-faceted approach to the challenges, the question remains: is this too little too late or are we doing enough to secure the grid for the next 30 years?

14 SOLAR ENERGY↗

Generating and Analyzing Program Call Graphs using Ontology

Call graph or caller-callee relationships have been used for various kinds of static program analysis, performance analysis and profiling, and for program safety or security analysis such as detecting anomalies of program execution or code injection attacks. However, different tools generate call graphs in different formats, which prevents efficient reuse of call graph results. In this paper, we present an approach of using ontology and resource description framework (RDF) to create knowledge graphs for specifying call graphs to facilitate the construction of full-fledged and complex call graphs of computer programs, realizing more interoperable and scalable program analyses than conventional approaches. We create a formal ontology-based specification of call graph information to capture concepts and properties of both static and dynamic call graphs so different tools can collaboratively contribute to more comprehensive analysis results. Our experiments show that ontology enables merging of call graphs generated from different tools and flexible queries using a standard query interface. Index Terms—Callgraph, ontology, knowl

Dorta, E.↗

Digital Twin Framework for PIP-II Linac: AI-Driven Multi-Scale Modeling from Ion Source to 800 MeV

The PIP-II linac will enable >1.2 MW beam power for DUNE, requiring unprecedented operational reliability across its warm front-end (RFQ, MEBT) and five distinct SRF sections operating at 162.5/325/650 MHz. We present a comprehensive digital twin framework uniquely combining a fully differentiable fast beam transport code with neural network surrogates trained on high-fidelity PIC simulations, capturing space charge and nonlinear dynamics beyond traditional envelope codes while achieving 10⁴× speedup at <1% accuracy. End-to-end differentiability enables gradient-based optimization across 500+ parameters simultaneously—previously impossible with conventional tools—while the model incorporates static/dynamic errors and serves as a virtual commissioning platform for diverse hardware integration. The framework facilitates reinforcement learning for pulsed/CW mode transitions, predictive maintenance through anomaly detection, and autonomous tuning algorithm development with real-time execution capability. Validation against physics simulations shows excellent agreement for the front-end, with initial results demonstrating potential for 30% commissioning time reduction and proactive fault mitigation, providing a scalable blueprint for operating next-generation high-intensity accelerators.

Pathak, Abhishek [Fermilab] (ORCID:000000021704208↗

Artificial Intelligence and Digital Engineering as Enablers for System Engineering in the Energy Sector

Systems engineering is of utmost importance for the success of high-cost, high-complexity megaprojects, which are common in the energy sector. However, the traditional document-centric systems engineering approach tends to be labor-intensive and time-consuming, which has inhibited its full adoption despite proven metrics on its return on investment. However, with the modern approach of digital engineering and technological advancements in artificial intelligence (AI) technologies, the barriers to systems engineering adoption can finally be broken. This paper goes through the systems engineering V-model for lifecycle management and assesses the current state of implementation of digital engineering (especially, mod-el-based systems engineering, digital twins, and digital threads) and AI for each step. It was observed that a combination of digital engineering and AI is being used across different industries to accelerate and optimize systems engineering processes such as concept development, requirements management, architecture definition, system development, verification and validation, operations, and maintenance. Specifically in the energy sector, AI-augmented digital engineering has shown initial potential in accelerated development and deployment, performance optimization, anomaly detection, predictive maintenance, and configuration management. However, challenges remain in integrating DE and AI into an end-to-end system lifecycle management ecosystem safely and reliably. Addressing these challenges and continuously developing impactful tools will enable fast, efficient, and high-frequency deployment of power generation capabilities to keep up with the world’s energy demands and build energy security.

42 - ENGINEERING↗

Artificial Intelligence and Digital Engineering as Enablers for Systems Engineering in the Energy Sector

Systems engineering is of utmost importance for the success of high-cost, high-complexity megaprojects, which are common in the energy sector. However, the traditional document-centric systems engineering approach tends to be labor-intensive and time-consuming, which has inhibited its full adoption despite proven metrics on its return on investment. However, with the modern approach of digital engineering and technological advancements in artificial intelligence (AI) technologies, the barriers to systems engineering adoption can finally be broken. This paper goes through the systems engineering V-model for lifecycle management and assesses the current state of implementation of digital engineering (especially, mod-el-based systems engineering, digital twins, and digital threads) and AI for each step. It was observed that a combination of digital engineering and AI is being used across different industries to accelerate and optimize systems engineering processes such as concept development, requirements management, architecture definition, system development, verification and validation, operations, and maintenance. Specifically in the energy sector, AI-augmented digital engineering has shown initial potential in accelerated development and deployment, performance optimization, anomaly detection, predictive maintenance, and configuration management. However, challenges remain in integrating DE and AI into an end-to-end system lifecycle management ecosystem safely and reliably. Addressing these challenges and continuously developing impactful tools will enable fast, efficient, and high-frequency deployment of power generation capabilities to keep up with the world?s energy demands and build energy security.

42 - ENGINEERING↗

Learning to Trigger: Reinforcement Learning at the Large Hadron Collider

High-throughput scientific facilities such as the Large Hadron Collider depend on real-time event filtering (\textit{triggering}) under tight constraints on bandwidth, latency, and storage. In practice, trigger menus are largely static and hand-tuned and can become suboptimal as detector conditions, pileup, and background composition drift over time. We cast online threshold tuning as a sequential decision-making problem: a reinforcement learning agent ingests streaming summaries of recent rates and signal-sensitive features and updates trigger thresholds to maximize signal efficiency while tracking a target background rate within a tolerance band. We adapt Group-Filtered Policy Optimization (GFPO) to streaming control and introduce two variants (GFPO-F, GFPO-FR) that enforce background rate feasibility during training. On a benchmark that emulates realistic collider operation, we study two representative triggers: a total transverse energy ($H_{T}$) trigger sensitive to pileup variation, and an anomaly-detection (AD) trigger based on reconstruction loss for rare or non-standard signatures. On Monte Carlo streams, our agent increases the fraction of in-tolerance time intervals by 48% ($H_T$) and 28% (AD), with a cumulative gain of up to 2% in signal efficiency on those in-tolerance intervals. Transferring from simulation to \emph{real} collision data (CMS Run 283408), the same agent, without fine-tuning, achieves a 56% ($H_T$) and 28% (AD) in-tolerance improvement over baselines, with further signal-efficiency gain on both triggers. To our knowledge, this is the \emph{first} demonstration of RL-based trigger control on real Large Hadron Collider collision data. Code is available at https://github.com/Zixind/GFPO_LHC (see repo for details).

Ding, Zixin [Chicago U.]↗

Database-Agnostic Log Analysis and Monitoring Framework

Prior to my internship, I was informed that a previous intern had built a tool to analyse MongoDB logs and look for invalid access attempts, which served as a great reference point for my project. I was initially tasked with expanding on her prototype and filling in the gaps such as integrating it with the main monitoring tool the lab uses. Eventually, the scope grew, expanding to support other databases and a growing collection of tools. I organized the framework around an observer pattern, meaning one point in the program sending updates to the rest of the framework. Every time a log was read and parsed, it was sent to be processed by the tools, using the type of event as a means to determine which tools should get a chance to act on the log. This decouples the tools from the log reader, making future updates and additions much easier. The framework processes MongoDB logs at ~135,000 entries per second and PostgreSQL logs at ~170,500 entries per second, accurately detecting anomalies such as slow queries and connections from unknown addresses. This framework serves to fill gaps in database monitoring tools currently implemented at the lab, such as tracking failed authentication for PostgreSQL and MongoDB which had very minimal or none before this framework. National labs such as Fermilab hold sensitive data and valuable computing resources, making them attractive targets. Monitoring intrusion attempts on databases is made much easier by this comprehensive monitoring suite.

Clark, Dylan [Unlisted, US, IL; Fermilab]↗