Simulation and the monte carlo method

Mathematical Handbook for Scientists and Engineers

Linear Systems, Random Data, Spectra Zero-Memory Nonlinear Systems Bilinear and Trilinear Systems Nonlinear System Input/Output Relationships Square-Law and Cubic Nonlinear Systems Statistical Errors in Nonlinear Estimates Parallel Linear and Nonlinear Systems.

Nonlinear System Analysis and Identification from Random Data

Leukemia diagnosis in blood slides using transfer learning in CNNs and SVM for classification

Learning-based fault localization has been intensively studied recently. Prior studies have shown that traditional Learning-to-Rank techniques can help precisely diagnose fault locations using various dimensions of fault-diagnosis features, such as suspiciousness values computed by various off-the-shelf fault localization techniques. However, with the increasing dimensions of features considered by advanced fault localization techniques, it can be quite challenging for the traditional Learning-to-Rank algorithms to automatically identify effective existing/latent features. In this work, we propose DeepFL, a deep learning approach to automatically learn the most effective existing/latent features for precise fault localization. Although the approach is general, in this work, we collect various suspiciousness-value-based, fault-proneness-based and textual-similarity-based features from the fault localization, defect prediction and information retrieval areas, respectively. DeepFL has been studied on 395 real bugs from the widely used Defects4J benchmark. The experimental results show DeepFL can significantly outperform state-of-the-art TraPT/FLUCCS (e.g., localizing 50+ more faults within Top-1). We also investigate the impacts of deep model configurations (e.g., loss functions and epoch settings) and features. Furthermore, DeepFL is also surprisingly effective for cross-project prediction.

https://dl.acm.org/doi/pdf/10.1145/3293882.3330574

DeepFL: integrating multiple fault diagnosis dimensions for deep fault localization

The attempts for solving linear inseparable problems have led to different variations on the number of layers of neurons and activation functions used. The backpropagation algorithm is the most known and used supervised learning algorithm. Also called the generalized delta algorithm because it expands the training way of the adaline network, it is based on minimizing the difference between the desired output and the actual output, through the downward gradient method (the gradient tells us how a function varies in different directions). Training a multilayer perceptron is often quite slow, requiring thousands or tens of thousands of epochs for complex problems. The best known methods to accelerate learning are: the momentum method and applying a variable learning rate. The paper presents the possibility to control the induction driving using neural systems.

/pdf/multilayer-perceptron-and-neural-networks-3f5tyef0nr.pdf

Multilayer perceptron and neural networks

The study of predictive control of temperature in oil-filled transformers is performed in this work. In order to study transformers thermal loss of life, complex models taking into account electrical and thermal characteristics are required. Moreover, the precision of thermal models is dependent upon the exactitude of the parameters. The work presented in this article shows that, through electromagnetic similitude laws, for an homogeneous series of transformers with different rated powers, the main parameters required for the thermal model can be obtained. Different methodologies to estimate thermal parameters with data from standardised heat-run tests are compared. The foremost advantage of our alternative methodology is its compactness, since parameters are obtained only from the knowledge of transformer rated power. Theoretical results were compared with data from transformer manufacturers and the good agreement between both validates theoretical results is accomplished.

/pdf/electromagnetic-and-thermal-model-parameters-of-oil-filled-sdcdoc9fx9.pdf

Electromagnetic and thermal model parameters of oil-filled transformers

http://www.wseas.us/e-library/conferences/2009/istanbul/ACMOS/ACMOS-21.pdf

Model for predictive control of temperature in oil-filled transformers

This article begins with a study of transformer additional losses due to the presence of non-sinusoidal loads, presents existing models to represent these losses and proposes an original model. Consequent increment in transformer loss of life is studied by means of a sensitivity analysis of the model to transformer rated power, load amplitude, thermal inertia, profiles correlation and harmonic spectrum content.

http://www.wseas.us/e-library/conferences/2009/istanbul/ACMOS/ACMOS-22.pdf

Transformer model extension for variation of additional losses with frequency

http://www.wseas.us/e-library/transactions/circuits/2009/29-337.pdf

Liliana Perescu-Popescu

Papers

Multilayer perceptron and neural networks

Electromagnetic and thermal model parameters of oil-filled transformers

Model for predictive control of temperature in oil-filled transformers

Transformer model extension for variation of additional losses with frequency

Non-linear thermal model for transformers study