Klaus Jürgen Folz and Herbert Martins Gomes
The objective of this article is to evaluate and compare the performance of two machine learning (ML) algorithms, i.e. support vector machines (SVMs) and random forests (RFs)…
Abstract
Purpose
The objective of this article is to evaluate and compare the performance of two machine learning (ML) algorithms, i.e. support vector machines (SVMs) and random forests (RFs), when classifying seven states of operation of an electric motor using the Mel-frequency cepstral coefficients (MFCCs) as extracted representative features.
Design/methodology/approach
The extracted MFCCs are calculated using the motor’s vibration and audio signals separately.
Findings
After the training, the SVM model obtained a mean accuracy of 100% for the MFCCs obtained from database vibration signals and 69.6% for the database of audio signals.
Research limitations/implications
The ML strategies and results reported are limited to the well-known data for industrial electric motors used in the evaluations, although it was performed tests and cross-validations with unseen data and the information from the confusion matrix.
Practical implications
The success of these methodologies in defect classification, where the RF presented a mean accuracy of 99.15% for the vibration signals and 63.82% for the audio signal, enables the use of this ML and extracted features as a predictive tool for failure and anomaly detection, lifetime predictions and online real-time monitoring.
Originality/value
It is the first time that the MFCCs are being used for anomaly detection in vibration and audio signals for electrical motors, as this extracted feature is usually used for human speech identification in the literature.
Details
Keywords
Leonardo Valero Pereira, Walter Jesus Paucar Casas, Herbert Martins Gomes, Luis Roberto Centeno Drehmer and Emanuel Moutinho Cesconeto
In this paper, improvements in reducing transmitted accelerations in a full vehicle are obtained by optimizing the gain parameters of an active control in a roughness road…
Abstract
Purpose
In this paper, improvements in reducing transmitted accelerations in a full vehicle are obtained by optimizing the gain parameters of an active control in a roughness road profile.
Design/methodology/approach
For a classically designed linear quadratic regulator (LQR) control, the vibration attenuation performance will depend on weighting matrices Q and R. A methodology is proposed in this work to determine the optimal elements of these matrices by using a genetic algorithm method to get enhanced controller performance. The active control is implemented in an eight degrees of freedom (8-DOF) vehicle suspension model, subjected to a standard ISO road profile. The control performance is compared against a controlled system with few Q and R parameters, an active system without optimized gain matrices, and an optimized passive system.
Findings
The control with 12 optimized parameters for Q and R provided the best vibration attenuation, reducing significantly the Root Mean Square (RMS) accelerations at the driver’s seat and car body.
Research limitations/implications
The research has positive implications in a wide class of active control systems, especially those based on a LQR, which was verified by the multibody dynamic systems tested in the paper.
Practical implications
Better active control gains can be devised to improve performance in vibration attenuation.
Originality/value
The main contribution proposed in this work is the improvement of the Q and R parameters simultaneously, in a full 8-DOF vehicle model, which minimizes the driver’s seat acceleration and, at the same time, guarantees vehicle safety.
Details
Keywords
Herbert Martins Gomes and Armando Miguel Awruch
To research the feasibility in using artificial neural networks (ANN) and response surfaces (RS) techniques for reliability analysis of concrete structures.
Abstract
Purpose
To research the feasibility in using artificial neural networks (ANN) and response surfaces (RS) techniques for reliability analysis of concrete structures.
Design/methodology/approach
The evaluation of the failure probability and safety levels of structural systems is of extreme importance in structural design, mainly when the variables are eminently random. It is necessary to quantify and compare the importance of each one of these variables in the structural safety. RS and the ANN techniques have emerged attempting to solve complex and more elaborated problems. In this work, these two techniques are presented, and comparisons are carried out using the well‐known first‐order reliability method (FORM), with non‐linear limit state functions. The reliability analysis of reinforced concrete structure problems is specially considered taking into account the spatial variability of the material properties using random fields and the inherent non‐linearity.
Findings
It was observed that direct Monte Carlo simulation technique has a low performance in complex problems. FORM, RS and neural networks techniques are suitable alternatives, despite the loss of accuracy due to approximations characterizing these methods.
Research limitations/implications
The examples tested are limited to moderated large non‐linear reinforced concrete finite element models. Conclusions are drawn based on the examples.
Practical implications
Some remarks are outlined regarding the fact that RS and ANN techniques have presented equivalent precision levels. It is observed that in problems where the computational cost of structural evaluations (computing failure probability and safety levels) is high, these two techniques could improve the performance of the structural reliability analysis through simulation techniques.
Originality/value
This paper is important in the field of reliability analysis of concrete structures specially when neural networks or RS techniques are used.
Details
Keywords
The purpose of this paper is to investigate the optimum design of a quarter car passive suspension system using a particle swarm optimization algorithm in order to minimize the…
Abstract
Purpose
The purpose of this paper is to investigate the optimum design of a quarter car passive suspension system using a particle swarm optimization algorithm in order to minimize the applied loads and vibrations.
Design/methodology/approach
The road excitation is assumed as zero-mean random field and modeled by single-sided power spectral density (PSD) based on international standard ISO 8608. The variance of sprung mass displacements and variance of dynamic applied load are evaluated by PSD functions and used as cost function for the optimization.
Findings
The advantages in using this methodology are emphasized by an example of the multi-objective optimization design of suspension parameters and the results are compared with values reported in the literature and other gradient based and heuristic algorithms. The paper shows that the algorithm effectively leads to reliable results for suspension parameters with low computational effort.
Research limitations/implications
The procedure is applied to a quarter car passive suspension design.
Practical implications
The proposed procedure implies substantial time savings due to frequency domain analysis.
Social implications
The paper proposes a procedure that allows complex optimization designs to be feasible and cost effective.
Originality/value
The design optimization is performed in the frequency domain taking into account standard defined road profiles PSD without the need to simulate in the time domain.
Details
Keywords
Herbert Martins Gomes and Armando Miguel Awruch
In this paper, special emphasis is given to uncertainties in the evaluation of the structural behavior, looking for a better representation of the system characteristics and…
Abstract
In this paper, special emphasis is given to uncertainties in the evaluation of the structural behavior, looking for a better representation of the system characteristics and quantification of the significance of these uncertainties in structural design. The reliability analysis of reinforced concrete structures is performed taking into account the spatial variability of material properties. The finite element method is used to analyze reinforced concrete structures. A multidimensional non‐Gaussian stochastic field generation model (independent of the finite element mesh) is developed and used. The reliability analysis is carried out employing the first order reliability method. Numerical examples are presented to study how to generate correlated non‐Gaussian stochastic fields and determine the reliability of a reinforced concrete structure with respect to a limit state function.
Details
Keywords
M.Elena Gómez-Miranda, M.Carmen Pérez-López, Eva Argente-Linares and Lázaro Rodríguez-Ariza
The characteristics of a particular organizational culture may affect performance in achieving the objectives of international joint ventures (IJVs), a type of partnership that is…
Abstract
Purpose
The characteristics of a particular organizational culture may affect performance in achieving the objectives of international joint ventures (IJVs), a type of partnership that is often used in international business relations between developed and emerging countries. The purpose of this paper is to analyse whether the underlying dimensions that characterize organizational culture in these countries may affect firms’ performance, specifically their competitiveness, effectiveness and efficiency.
Design/methodology/approach
The survey conducted for this study was addressed to Spanish-Moroccan IJVs trading in Morocco. The research hypotheses were tested using multivariate analysis techniques (exploratory factor analysis and linear regression model).
Findings
Based on information provided by the CEOs of Spanish-Moroccan IJVs between small- to medium-sized firms, the present study shows that levels of competitiveness, effectiveness and/or efficiency in these organizations are influenced by the involvement of staff in management, the degree of centralization of decision taking and the firms’ emphasis on results or on procedures.
Practical implications
This research contributes to the knowledge of the main factors related to the organizational culture of joint ventures that influence competitiveness, effectiveness and efficiency achieved.
Originality/value
The value provided by this research lies in the sample examined, in its focus on a very common type of partnership between SMEs, which has been little studied previously, and in the fact that the results obtained are extensible to other realities, such as partnerships between European companies and those from countries with similar characteristics (located in Africa or in countries where an Arab culture prevails).
Details
Keywords
This narrative review explored the efficacy of school-based child sexual abuse prevention programmes between 1990 and 2002. There were 22 efficacy studies that met clear inclusion…
Abstract
Purpose
This narrative review explored the efficacy of school-based child sexual abuse prevention programmes between 1990 and 2002. There were 22 efficacy studies that met clear inclusion criteria. Results covered both methodological design and the range of outcome measures. Methodology was analysed through four dimensions (target population, prevention programme implementation, evaluation methodology and cost-effectiveness). Outcomes for children covered nine categories (knowledge, skills, emotion, perception of risk, touch discrimination, reported response to actual threat/abuse, disclosure, negative effects and maintenance of gains). The studies had many methodological limitations. Prevention programmes had a measure of effectiveness in increasing children ' s awareness of child sexual abuse as well as self-protective skills. Beyond minimal disclosure rates, there was no evidence to demonstrate that programmes protected children from intra-familial sexual abuse. For a small number of children prevention programmes produced minimal negative emotional effects. Recommendations for future research, policy and practice, include realistic outcomes for child participants and locating programmes within wider abuse prevention measures.
Details
Keywords
Norberto Santos, Claudete Oliveira Moreira and Luís Silveira
Tourism in Coimbra today is influenced by the fact that the Univer(s)city was distinguished as a World Heritage Site in 2013. The number of visits has grown very significantly in…
Abstract
Purpose
Tourism in Coimbra today is influenced by the fact that the Univer(s)city was distinguished as a World Heritage Site in 2013. The number of visits has grown very significantly in recent years, but the diversification of the tourist offer is still weak and unable to take advantage of existing resources. This paper aims to present genealogy tourism as an alternative urban cultural tourism in Coimbra.
Design/methodology/approach
Methodology involved mapping the Jewish culture elements in the city of Coimbra, and a route was outlined and proposed.
Findings
Genealogy tourism resources are identified in the historic centre of the city. These alternative spaces need urban rehabilitation and (re)functionalisation, which allowed the authors to rethink tourism in Coimbra. They are the motivation to visit for all urban cultural tourists, especially Israelis/Jews, and provide contact with places where the experiences of ancestors combine with the history and memory of places, with recent discoveries and the elements of Jewish culture in the city.
Originality/value
It is concluded that the quantity, diversity, authenticity and singularity of the heritage resources that bear witness to the Jewish presence in Coimbra are sufficient assets to create a route, to enrich the tourist experience in the city and to include the destination in the Sephardic routes.