Engineering Papers⌕ Search

Engineering topics

Li, Yeni

Publications and source records attributed to Li, Yeni.

Feature extraction for subtle anomaly detection using semi-supervised learning

The demand for automated and effective monitoring techniques has soared with the increased digitization of industrial monitoring systems. State-of-the-art machine learning methods are effectively detecting abrupt changes in system states. However, these methods lack comparable maturity in detecting subtle changes that may be signs of incipient faults. This manuscript argues that the current anomaly detection methods can be enhanced by exploring weak patterns to enable subtle variation detection. Specifically, the concept of semi-supervised learning is employed, with labels representing knowledge about some anomalous conditions of a system. The basic idea is to extract a candidate set of weak patterns discarded by state-of-the-art baselining algorithms. With few labeled anomalous data, the algorithm selects the weak patterns and allows for their possible fusion using the highest sensitivity to the labeled anomalies. Here, the method’s applicability is demonstrated using a representative pressurized water reactor (PWR) model simulated by Dymola.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Big Data For Operation and Maintenance Cost Reduction

The purpose of this research is to develop a first-of-a-kind framework for integrating Big Data capability into the daily activities of our current fleet of nuclear power plants. Big Data is traditionally defined as data sets with high volume, velocity, and heterogeneity, and the existing Big Data analytics capabilities are now widely popular in fields such as finance, weather, e-commerce, healthcare and sports. In the nuclear industry, while the volume and velocity of data may present computational challenges for existing analytics capabilities, data heterogeneity are seen to present the major challenge. This research project mainly focuses on incorporating the wide range of data heterogeneities in nuclear power plants into an integrated Big Data Analytics capability. The primary end-product of this project is a Big Data framework that is capable of dealing with the large volume and heterogeneity of the data found in nuclear power plants to extract timely and valuable information on equipment performance. The framework can generate system insights that are actionable relations between measurable impacts and the corresponding maintenance action plans and enable optimization of plant operation and maintenance based on the extracted information. The developed framework is capable of handling heterogeneous data including both image data and time-series sensor data. Specifically, this developed framework includes the following components. The first component is an overarching maintenance ontology which includes system insights required by maintenance optimization. The maintenance ontology interacts with other components in the developed framework. The second component handles Piping & Instrumentation Diagram (P&ID) data. It can be used to extract system components and their relations automatically from the P&IDs. This extracted information is stored in the first component, i.e., maintenance ontology, and is also used as input to the third component, i.e., a tool for generating the fault tree for the corresponding system. The generated fault tree in turn is stored in the ontology for assessing risk that is used as a criterion in maintenance policy optimization. The fourth component is a tool for inferring the parameters in the Markov degradation model for a nuclear system. It uses basic information from the ontology. The fifth component is a tool for assessing the degradation level using sensor measurement data, for example, pressure, flowrate. This tool can be used for determining corrective maintenance actions. The results obtained from components four and five are returned to the ontology. The sixth component of the framework is a tool for optimizing the maintenance policy for a nuclear system of interest. It takes certain basic information from the ontology, e.g., costs of maintenance actions and system failures, as input, and returns the optimal maintenance policy to the ontology. This tool can be used for determining predictive maintenance actions. A set of experiments have also been conducted to verify the algorithms developed in this project for nuclear system degradation monitoring. The experiments are based on four solenoid valves, similar to the ones used in nuclear power plants. The analyses based on the experimental data using two algorithms, i.e., the Randomized Window Decomposition (RWD) algorithm and the particle filtering algorithm, and the results are introduced in the report. The Big Data framework developed in this project can be used as a support tool in daily activities of plant operation and maintenance and will reduce current costs while maintaining or improving safety levels. Overall, the project will not only benefit existing reactors, however it will open new frontiers to realize the long overdue value of Big Data Analytics in the nuclear sphere.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Process Anomaly Detection for Sparsely Labeled Events in Nuclear Power Plants

An essential aspect of online monitoring, subtle anomaly detection increases the detection lead time for equipment failure and enables a nuclear power plant (NPP) to mitigate unexpected partial or full outages, resulting in significant cost saving to the plant. Once an anomaly is detected by plant staff, its cause and severity are investigated. Because the vast majority of anomalies require some level of investigation, including some that require time-consuming examination, before they are passed over to the engineering organization for further analysis, plants are often equipped with tools to assist the staff in performing anomaly detection. Those tools operate as a black box and are often based on statistical methods that establish sensor correlations using preconfigured mathematical models and flag correlation deviations as anomalies. Due to the number of anomalies detected at a given NPP on a daily basis, a significant number of flagged anomalies usually await examination for days or weeks. A primary cause of this backlog is that the methods used by the tools generate many false positives. Though this is usually attributed to oversensitive model settings due to very narrow normal operation bands, it can also be associated with the model development being inadequate for the process being monitored, or with missing model inputs that could have explained misclassified positives. The performance of anomaly detection tools impacts their plant acceptance and utilization, especially when the effort to address false positives generated by the tool depletes the value or cost saved by using that tool. Thus, means to advance anomaly detection performance have been investigated by the Department of Energy’s Light Water Reactor Sustainability program. Previous and ongoing efforts have targeted unsupervised machine-learning (ML) methods, which do not require the labeling of any data fed into the ML model. By contrast, in supervised anomaly detection methods, every data point is labeled as either a normal or abnormal process condition, and the model is trained to replicate the classification process. Supervised methods usually outperform unsupervised methods, due to the added value in differentiating normal from anomalous states of the monitored process. An NPP’s corrective action program requires it to track and document, via a dedicated report, the resolution of any issues that occur within the plant. Once created, each report is reviewed by a plant screening committee, and several classifications and decisions are made. Recently, a collaborating NPP developed an artificial intelligence and ML-based classifier to categorize a condition report (CR) into classes that can serve to label the data as normal or anomalous. Applying CRs as labels represents a semi-supervised use case. Semi-supervised ML assumes that labels exist for some data points (i.e., labeled anomalies, in this case) but not for the rest. In this effort, semi-supervised ML methods were used to fuse data from CRs with anomaly detection methods in order to test the hypothesis that partially labeled anomalies would improve the accuracy of the anomaly detection methods. Specifically, two methods were used. The first is the deep Semi-supervised Anomaly Detection (deep SAD) method, which can handle labels ranging from fully unsupervised to fully supervised cases. The second is a newly designed ML method developed specifically for this effort and referred to as the high-order feature (HOF)-based method. To evaluate these two methods in controlled environments, synthetic data generators were developed and used. The first datasets used a spring-mass-damper (SMD) system simulator commonly found in mechanical engineering references. This was used to create two use cases: a one- and a three-mass system. Anomalies were introduced by changing the spring and damper coefficients while the system was actuated by random forces. The second datasets used the commercial Dymola-Modelica software to build a simplified nuclear reactor model. Anomalies were added in the form of corrupted sensor readings and/or control commands. The deep SAD method was tested using the SMD system, while the HOF method was tested using both datasets. Application of the deep SAD semi-supervised ML method demonstrated that labels can generate increased confidence in detecting true anomalies. This helped increase the number of true positives and decrease the number of false negatives—something that would aid in addressing the backlog of possible anomalies. Application of the HOF method demonstrated that labels can aid in down selecting from a candidate set of features to a more optimal subset in order to better differentiate between normal and anomalous conditions.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Data trustworthiness signatures for nuclear reactor dynamics simulation

With the increased reliance on digitization in industrial control systems, the need for effective monitoring techniques has risen dramatically. Specifically, there is now a growing concern about the so-called false data injection (FDI) attacks. These attacks aim to alter the raw sensors’ data to cause malicious outcomes. Any serious FDI algorithm is based on an intimate knowledge of the system and its associated physics models, which renders conventional outlier/anomaly detection techniques almost obsolete in the face of such attacks. Thus, a critical need has emerged to develop a new class of defense methods that are capable of detecting FDI attacks under the assumption that the attacker has a strong familiarity with the system and its physics modeling. This class of defense methods are denoted by model-based defenses which are premised on the assumption that the attacker, while having a good understanding of the system, does not have full privileged access to all proprietary data and historical records of operation. However, (s)he is assumed to be capable of learning system behavior using self-learning techniques during an initial lie-in-wait period. To defend against this scenario, we propose a new model-based randomized window algorithm that searches time-series data for signatures that can serve as classifiers between normal and FDI scenarios. The classifiers are based on the correlations between the dominant degrees of freedom (DOFs) and the less-dominant DOFs (expected to be very sensitive to the system details that are unknown to the attacker). For demonstration, RELAP5 models are employed to calculate representative nuclear reactor behavior during a number of transient scenarios. Finally, falsified data are injected into the RELAP5-simulated behavior, and the proposed signature-identification algorithm is employed to detect the injected data.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

ROM-Based Surrogate Systems Modeling of EBR-II

We report the System Analysis Module (SAM), developed and maintained by Argonne National Laboratory, is designed to provide whole-plant transient safety analysis capabilities for a number of advanced non-light water reactors, including sodium-cooled fast reactor (SFR), lead-cooled fast reactor (LFR), and molten salt reactor (MSR)/fluoride-salt-cooled high-temperature reactor (FHR) designs. SAM is primarily constructed as a systems-level analysis tool, with the potential to incorporate reduced order models from three-dimensional computational fluid dynamics (CFD) simulations to improve characterization of complex, multidimensional physics. It is recognized that the computational expense associated with CFD can be intractable for various engineering analyses, such as uncertainty quantification, inference, and design optimization. This paper explores the reducibility of a SAM model using recent advances in randomized linear algebra techniques, which attempt to find recurring patterns in the various realizations generated by a model after randomly perturbing all its input parameters. The reduction is described in terms of fewer degrees of freedom (DOFs), referred to as the active DOFs, for the model variables such as input model parameters and model responses. The results indicate that there is significant room for additional reduction that may be leveraged for additional computational gains when employing SAM for engineering-intensive analyses that require repeated model executions. Different from physics-based reduction approaches, the proposed approach allows one to estimate upper bounds on the reduction errors, which are rigorously developed in this work. Finally, different methods for surrogate model construction, such as regression and neural network-based training, are employed to correlate the input and output active DOFs, which are related back to the original variables using matrix-based linear transformations.

42 ENGINEERING↗

Development of Defenses against False Data Injection Attacks for Nuclear Power Plants

With the recent successful attempts against the digital control systems of critical infrastructures, there is a need to develop new defense strategies that recognize two important realities, 1) state- sponsored attackers can rely on a number of techniques including espionage, social engineering, and brute force techniques, etc. to gain access to the raw data used to control system behavior, 2) attackers can falsify operational data in manners that do not trigger conventional outlier/anomaly detection techniques in order to go undetected, which is referred to as false data injection attacks. Therefore, there is a strong need to explore another class of defense measures, referred to as physical process defense, serving as a new line of defense in the event existing defenses relying on information protection measures are breached. This physical process defenses utilize the physics and engineering models of the system to build unique signatures for genuine system behavior. If successful, the signatures would be able to detect attacks that falsify the operational data and render them harmless before they can inflict physical damage on the system. This report is focused on exploring the feasibility of physical process defenses for nuclear reactors, and their associated functional requirements to maximize their resiliency against state-sponsored, or equivalent, attackers.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

ROM-based Subset Selection Algorithm for Efficient Surrogate Modeling

This summary discusses a new methodology to support the construction of surrogate models for computationally expensive models with many input parameters and output responses. In realistic engineering calculations, such as thermal-hydraulics simulation, the cost of the simulation is typically expensive, especially when detailed models are employed to describe system behavior, thereby justifying the need for surrogate models. The cost of surrogate model construction is typically a function of the number of model parameters and the degree of model nonlinearity. For sufficiently complex models, the cost of constructing the surrogate model could be by itself overwhelmingly expensive, thereby diminishing its value. One key requirement of surrogate model construction is the ability to preselect the best parametric form to describe the model behavior over the expected range of parameter variations. This summary presents a new algorithm to help select the best approximation out of a template of functions using recent advances in reduced order modeling (ROM) techniques. The idea of selecting functions out of a template is typically referred to as subset selection in the statistical and applied mathematics community. Different from existing subset selection algorithms, we rely on ROM to help identify the most influential functions for the surrogate model. Previous work has shown that ROM can help reduce the effective dimensionality of the parameter space, which can help reduce the cost of surrogate model construction. In this summary, we adopt a different philosophy, where the fitting functions are treated as pseudo parameters and ROM is applied on the pseudo parameter space, with the result now being a ranked list of the pseudo parameters along with their contribution to the quality of the surrogate model fit. This approach allows the analyst to discard all terms that have negligible impact on the quantity of interest. A relap5 model is employed to demonstrate the application of the proposed methodology. (authors)

42 ENGINEERING↗

Dimensionality Reduction Algorithms for Big Nuclear Data Streams with Application to SCALE Cross Sections

A central problem in performing accurate nuclear reactor simulations is the ability to capture the changes in the few group microscopic cross sections as functions of system state, e.g. moderator density, fuel temperature, or neutron spectrum, and time and/or burnup of the fuel. For most reactor simulation scenarios, this results in huge data stream, which must be interpolated as needed by downstream calculations (e.g., core-wide simulation). This is typically accomplished by using sophisticated interpolating functions to capture the few-group variations over the wide range of conditions expected during operation. To address handling this big data, this paper explores the use of dimensionality reduction (DR) techniques to identify the true dimensionality of the data prior to its use in the calculations. This could significantly reduce the computational cost required for data processing/manipulation, thus reducing a significant portion of the total calculations time. By way of an example, in the current version of the ORIGEN-ARP module, the assembly-averaged microscopic cross sections at different conditions, e.g., fuel enrichments, irradiation history and water density, which are calculated using the TRITON sequence and interpolated on-demand by the ARP module to model assembly depletion for a wide range of assembly types (e.g. Westinghouse 17x17 and GE 9x9) over a wide range of conditions. This approach typically results in ∼1 GB per assembly type, which although not insurmountable, can be impractical for distribution. and it is the premise of this paper that most of this data is heavily correlated and hence can take great benefit of condensation via DR. To demonstrate DR, the proposed algorithm is applied to the nuclear data streams generated by SCALE 6.1. Using the depletion sequence (TRITON), microscopic cross sections are calculated for a mixed oxide (MOX) fuel assembly design at a representative range of fuel enrichments and fuel burnup. In conventional core simulators, multi dimensional interpolation of these cross sections is used to provides neutronics feedback due to local conditions without returning to expensive TRITON calculations for every change. This paper shows that dimensionality reduction can help compress the size of the cross section data without sacrificing the accuracy of their reconstruction at all conditions of interest. This work explores the application of dimensionality reduction algorithms to nuclear data streams typically employed in support of conventional full-core reactor calculations which rely on interpolation of pre-calculated microscopic cross sections. In particular, analysis of the few group MOX cross sections revealed that a large degree of dependencies exist between the few-group cross sections, which can be revealed via range-finding algorithms. Specifically, correlations between different reaction types for the same isotope have been identified, and correlations between different enrichments and correlations across different isotopes have been identified. This paper demonstrates that the reduction errors can be quantified, and they are uncorrelated with the range of variations for the various cross sections. In future work, we will focus on identifying the smallest rank that takes into account all types of correlations across the range of conditions used to model cross section variations. (authors)

22 GENERAL STUDIES OF NUCLEAR REACTORS↗