Fault Detection and Diagnosis in dynamic systems using Weightless Neural Networks

Oliveira, Jose C. M.; Pontes, Karen Valverde; Sartori, Isabel; Embiruçu, Marcelo

doi:10.1016/j.eswa.2017.05.020

Cited by 43 publications

(19 citation statements)

References 59 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Moreover, many studies have approved the superiority of ML classifiers in improving the performance of SFP models [15]. Some of used ML algorithms for dealing with SFP include OneR (One Rule), DTs, NB, SVM and ANN classifiers [16]- [18].…”

Section: Doamentioning

confidence: 99%

Boosted Whale Optimization Algorithm With Natural Selection Operators for Software Fault Prediction

et al. 2021

View full text Add to dashboard Cite

Software fault prediction (SFP) is a challenging process that any successful software should go through it to make sure that all software components are free of faults. In general, soft computing and machine learning methods are useful in tackling this problem. The size of fault data is usually huge since it is obtained from mining software historical repositories. This data consists of a large number of features (metrics). Determining the most valuable features (i.e., Feature Selection (FS) is an excellent solution to reduce data dimensionality. In this paper, we proposed an enhanced version of the Whale Optimization Algorithm (WOA) by combining it with a single point crossover method. The proposed enhancement helps the WOA to escape from local optima by enhancing the exploration process. Five different selection methods are employed: Tournament, Roulette wheel, Linear rank, Stochastic universal sampling, and random-based. To evaluate the performance of the proposed enhancement, 17 available SFP datasets are adopted from the PROMISE repository. The deep analysis shows that the proposed approach outperformed the original WOA and the other six state-of-the-art methods, as well as enhanced the overall performance of the machine learning classifier.

show abstract

Section: Doamentioning

confidence: 99%

Boosted Whale Optimization Algorithm With Natural Selection Operators for Software Fault Prediction

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Numerous approaches are available to find anomalies in univariate/multivariate sequences. We group these methods into four categories: (1) Statistics-based methods [19] (2) Intelligent-computing methods [20] (3) Bayesian networks [21] and (4) Model-based approaches [22]. Statistical based methods come from techniques that detect abnormal changes.…”

Section: Sequential Seriesmentioning

confidence: 99%

An incremental malware detection model for meta-feature API and system call sequence

Kishore

Barisal

Mohapatra

2020

Proceedings of the 2020 Federated Conference on Computer Science and Information Systems

View full text Add to dashboard Cite

In this technical world, the detection of malware variants is getting cumbersome day by day. Newer variants of malware make it even tougher to detect them. The enormous amount of diversified malware enforced us to stumble on new techniques like machine learning. In this work, we propose an incremental malware detection model for meta-feature API and system call sequence. We represent the host behaviour using a sequence of API calls and system calls. For the creation of sequential system calls, we use NITRSCT (NITR System call Tracer) and for sequential API calls, we generate a list of anomaly scores for each API call sequence using Numenta Hierarchical Temporal Memory (N-HTM). We have converted the API call sequence into six meta-features that narrates its influence. We do the feature selection using a correlation matrix with a heatmap to select the best meta-features. An incremental malware detection model is proposed to decide the label of the binary executable under study. We classify malware samples into their respective types and demonstrated via a case study that, our proposed model can reduce the effort required in STS-Tool(Socio-Technical Security Tool) approach and Abuse case. Theoretical analysis and real-life experiments show that our model is efficient and achieves 95.2% accuracy. The detection speed of our proposed model is 0.03s. We resolve the issue of limited precision and recall while detecting malware. User's requirement is also met by fixing the trade-off between accuracy and speed.

show abstract

“…To ensure that the operation of any process is correct, there must be a set of actions that must comply with three fundamental stages, failure detection, diagnosis and restoration of operating conditions as specified by the process, ie process monitoring and is applicable to any application [3], [4]. The industry has adopted methods and procedures that have allowed to automatically detect failures in generation and electric motors, extending their life cycle, improving their safety and providing financial savings [5], [6], these methods are based on the identification of highly probabilistic parameters, calculation of equations, estimation of state variables and multivalent statistical methods that help to determine the causes of failure [7], [8] and other classical methods used is the rule of diffuse equations or neural network approaches [9], [10]. In general, the traditional methods of failure detection and diagnosis are based on dynamic and mathematical models [11], [12], having important aspects in failure detection as the expertise of the system operator and knowing the functions of the individual components and their respective connections, qualitative and/or quantitative in the process of nominal operation against abnormal operation, in relation to the presence of an abnormal event or called failure [13] in a machine, for example an engine, can influence its operation, efficiency and even the interruption of the main functions in the process [14].…”

Section: Introductionmentioning

confidence: 99%

Fault Detection using Principal Component Analysis and Mean Value Modeling in a 2 MW gas engine

2020

View full text Add to dashboard Cite

This paper describes the combination of statistical techniques and mathematical modeling in order to developed a fault detection system in a 2 MW natural gas engine under actual operation conditions. The Mixing chamber, turbochargers, intake and exhaust manifolds, cylinders, throttle and bypass valves, and the electric generator, which are the main components of the gas engine, were studied under a mean value engine to complement the statistical analysis. Objective: The main objective of this paper is to integrate two approaches in order to relate the faults with the changes of mean thermodynamic values of the system, helping to sustain the engine in optimal operating conditions in terms of reliability. The Principal Component Analysis (PCA), a multivariate statistical fault detection technique, was used to analyze the historical data from the gas engine to detect abnormal operation conditions, by means of statistical measures such as Square Prediction Error (SPE) and T2. These abnormal operation conditions are categorized using cluster techniques and contributions plots, to later examine its causes with the support of the results of a mean value mathematical model proposed for the system. The integration of the proposed methods allowed successfully identify which component or components of the engine might be malfunctioning. Once combined, these two methods were able to accurately predict and identify faults as well as shut downs of the gas engine during a month of operation. Statistical analysis was used to detect faults on a 2 MW industrial gas engine, also the result were compared with a mean value model in order to detect variations of the thermodynamic properties of the system at abnormal conditions.

show abstract

Fault Detection and Diagnosis in dynamic systems using Weightless Neural Networks

Cited by 43 publications

References 59 publications

Boosted Whale Optimization Algorithm With Natural Selection Operators for Software Fault Prediction

Boosted Whale Optimization Algorithm With Natural Selection Operators for Software Fault Prediction

An incremental malware detection model for meta-feature API and system call sequence

Fault Detection using Principal Component Analysis and Mean Value Modeling in a 2 MW gas engine

Contact Info

Product

Resources

About