基于重训练高斯混合模型的语音转换方法

基于高斯混合模型的语音转换方法是语音转换中比较成功的方法之一,但基于高斯混合模型的转换方法训练过程复杂,训练时间长,需要大量的训练语音,这些都影响了它的实用性。对于传统高斯混合模型训练中的问题进行了分析研究,提出了训练过程中一个改进的方法（即二次训练法）,实验分析证明这个方法能有效提高模型训练速度,改善转换系统性能。     

Biogeography-based optimisation for flexible manufacturing system scheduling problem

Biogeography-based optimisation (BBO) algorithm is a new evolutionary optimisation algorithm based on geographic distribution of biological organisms. With probabilistic operators, this algorithm is able to share more information from good solutions to poor ones. BBO prevents the good solutions to be demolished during the evolution. This feature leads to find the better solutions in a short time rather than other metaheuristics. This paper provides a mathematical model which integrates machine loading, part routing, sequencing and scheduling decision in flexible manufacturing systems (FMS). Moreover, it tackles the scheduling problem when various constraints are imposed on the system. Since this problem is considered to be NP-hard, BBO algorithm is developed to find the optimum /near optimum solution based on various constraints. In the proposed algorithm, different types of mutation operators are employed to enhance the diversity among the population. The proposed BBO has been applied to the instances with different size and degrees of complexity of problem adopted from the FMS literature. The experimental results demonstrate the effectiveness of the proposed algorithm to find optimum /near optimum solutions within reasonable time. Therefore, BBO algorithm can be used as a useful solution for optimisation in various industrial applications within a reasonable computation time.     

Automatic breast cancer detection based on optimized neural network using whale optimization algorithm

Breast cancer is the second deadliest type of cancer. Early detection of breast cancer can considerably improve the effectiveness of treatment. A significant early sign of breast cancer is the mass. However, separating the cancerous masses from the normal portions of the breast tissue is usually a challenge for radiologists. Recently, because of the availability of high‐accuracy computing, computer‐aided detection systems based on image processing have become capable of accurately diagnosing the various types of cancers. The main purpose of this study is to utilize a powerful image segmentation method for the diagnosis of cancerous regions through mammography, based on a new configuration of the multilayer perceptron (MLP) neural network. The most popular method for minimizing the errors in an MLP neural network is backpropagation. However, this method has certain drawbacks, such as a low convergence speed and becoming trapped at the local minimum. In this study, a new training algorithm based on the whale optimization algorithm is proposed for the MLP network. This algorithm is capable of solving various problems toward the current algorithms for the analyzed systems. The proposed method is validated on the Mammographic Image Analysis Society database, which contains 322 digitized mammography images, and the Digital Database for Screening Mammography, which contains approximately 2500 digitized mammography images. To assess the detection performance of the proposed system, the correct detection rate, percentage of identification with false acceptance, and percentage of identification with false rejection were evaluated and compared using various methods. The results indicate that the proposed method is highly efficient and yields significantly better accuracy compared with other methods.     

基于变分贝叶斯算法和MLP网络的后非线性混合盲源分离方法研究

传统的后非线性模型往往要求其后非线性函数是可逆的,否则无法进行源信号的分离。然而在实际中,这一要求并不完全满足。针对此不足,结合变分贝叶斯推论和多层感知器网络,提出一种改进的多层感知器后非线性模型,它通过多层感知器来模拟后非线性函数,实现对不可逆后非线性函数混合的盲分离。仿真和实验结果表明该方法是有效的。     

Applying ANN,ANFIS and LSSVM Models for Estimation of Acid Solvent Solubility in Supercritical CO₂

In the present work, a novel machine learning computational investigation is carried out to accurately predict the solubility of different acids in supercritical carbon dioxide. Four different machine learning algorithms of radial basis function, multi-layer perceptron (MLP), artificial neural networks (ANN), least squares support vector machine (LSSVM) and adaptive neuro-fuzzy inference system (ANFIS) are used to model the solubility of different acids in carbon dioxide based on the temperature, pressure, hydrogen number, carbon number, molecular weight, and the dissociation constant of acid. To evaluate the proposed models, different graphical and statistical analyses, along with novel sensitivity analysis, are carried out. The present study proposes an efficient tool for acid solubility estimation in supercritical carbon dioxide, which can be highly beneficial for engineers and chemists to predict operational conditions in industries.     

Wavelet neural network (WNN) approach for calibration model building based on gasoline near infrared (NIR) spectra

In this paper we have compared the abilities of two types of artificial neural networks (ANN): multilayer perceptron (MLP) and wavelet neural network (WNN) — for prediction of three gasoline properties (density, benzene content and ethanol content). Three sets of near infrared (NIR) spectra (285, 285 and 375 gasoline spectra) were used for calibration models building. Cross-validation errors and structures of optimized MLP and WNN were compared for each sample set. Four different transfer functions (Morlet wavelet and Gaussian derivative – for WNN; logistic and hyperbolic tangent – for MLP) were also compared. Wavelet neural network was found to be more effective and robust than multilayer perceptron.     

Development and analyses of data-driven models for predicting the bed depth profile of solids flowing in a rotary kiln

Soft computing data-driven modeling (DDM) techniques have attracted the attention of many researchers across the globe as they do not require deep knowledge of the complex physical process. In the present research, data-driven based models have been developed using support vector regression (SVR), multilayer perceptron neural network (MLP), radial basis function neural network (RBFNN) and general regression neural networks (GRNN) techniques for predicting the bed depth profile of solids flowing in a rotary kiln. The performances of the developed models were compared and evaluated against the experimental results in terms of statistical measures such as coefficient of determination (R²), and average absolute relative error (AARE). The obtained results and findings from this research have revealed that data-driven models can predict the bed depth profile of solids flowing in a rotary kiln quite accurately. The SVR-based model exhibited the lowest AARE value of 1.72% and highest R² value of 0.9981 while GRNN, RBFNN, and MLP models gave corresponding values of AARE as 3.69%, 55.13%, 98.15% and those of R² as 0.9898, 0.0052 and 0.0081, respectively. Moreover, the developed DDM-based models i.e. GRNN, RBFNN, and MLP models overcame the limitations of the existing solutions which involved iterative numerical procedure entailing high degree of computational complexity.     

基于DBLSTM-DCNN的骨导和气导语音转换

为了解决人们在强噪声环境下,通过空气途径传递的语音信号会严重失真的问题,提出了一种基于深层双向长短期记忆-深度卷积神经网络(Deep Bidirectional Long and Short Term Memory-Deep Convolutional Neural Network,DBLSTM-DCNN)的骨导语音转...     

基于GMM模型和LPC-MFCC联合特征的声道谱转换研究

声道谱转换是语音转换中的关键技术。目前,大多数语音转换方法对声道谱的转换都是先提取语音中的某一种声道特征参数,然后对其进行训练转换,进而合成转换语音。由于不同的声道特征参数表征着不同的物理和声学意义,因此这些方法通常忽略了不同声道特征参数之间可能存在的互补性。针对这一问题,研究了不同声道特征参数之间进行联合建模的方法,引入了一种由线性预测系数(LinearPredictionCoefficient,LPC)和梅尔频率倒谱系数(Mel-Frequency Cepstral Coefficient, MFCC)联合构成的LPC-MFCC特征参数,提出了一种基于高斯混合模型(Gaussian Mixture Model, GMM)和LPC-MFCC联合特征参数的语音转换方法。为验证文中方法的有效性,仿真实验选取了基于GMM和LPC的语音转换方法进行对比,对多组实验数据进行主观和客观测试,结果表明,文中提出的语音转换方法可以获得相似度更高的转换语音。     

Absolute measurement of the effective nonlinearities of KTP and BBO crystals by optical parametric amplification

Absolute magnitudes of the effective nonlinearity, deff, were measured for seven KTP and six BBO crystals. The d(eff), were derived from the parametric gain of an 800-nm signal wave in the sample crystals when they were pumped by the frequency-doubled, spatially filtered light from an injectionseeded, Q-switched Nd:YAG laser. The KTP crystals, all type II phase matched with propagation in the X-Z plane, had d(eff) values ranging from 1.97 to 3.50 pm/V. Measurements of gain as a function of phase velocity mismatch indicate that two of the KTP crystals clearly contain multiple ferroelectric domains. For five type I phase-matched BBO crystals, d(eff) ranged from 1.76 to 1.83 pm/V, and a single type II phase-matched BBO crystal had a d(eff) of 1.56 pm/V. The uncertainty in our measurements of d(eff) values is ±5% for KTP and ±10% for BBO.     

A neural network approach for damage detection and identification of structures

This study examines the feasibility of using artificial neural network in conjunction with system identification techniques to detect the existence and to identify the characteristics of damage in composite structures. The methodology proposed here includes a training phase and a recognition phase. In the training phase, candidate models for structures with various types of damage are designated as the patterns. These patterns are organized into pattern classes according to the location and the severity of the damage. Then system identifications are performed to extract the transfer functions as the features of the structural systems. These transfer functions are fed into a multi-layer perceptron (MLP) as the input patterns for training. The MLP serves as a nearest neighborhood classifier. In the pattern recognition phase, a structure with unforeseen damage is classified within the closest class in the training set and the damage in the structure is identified as that of the class. The results of numerical tests demonstrate the feasibility of the proposed method.     

On-line tuning system of multivariate dEWMA control based on a neural network approach

The double exponentially weighted moving average (EWMA) controller is a popular algorithm for on-line quality control of semiconductor manufacturing processes. The performance of the closed-loop system hinges on the adequacy of the two weight parameters of the double EWMA equations. In 2004, Su and Hsu presented an approach based on the neural technique for ‘on-line’ tuning the weight of the single EWMA equation in the single-input single-output (SISO) system. The present paper extends the neural network on-line tuning scheme to the double EWMA controller for the non-squared multiple-input multiple-output (MIMO) system, and validates the control performance by means of a simulated chemical–mechanical planarization (CMP) process in semiconductor manufacturing. Both linear and non-linear equipment models are considered to evaluate the proposed controller, coupling with the deterministic drift, the Gaussian noise and the first-order integrated moving average (IMA) disturbance. It has been shown from a variety of simulation studies that the proposed method exhibits quite competitive control performance as compared with the previous control system. The other merit of the proposed approach is that the tuning system, if sufficient training in a neural network is available, can be practicably applied to complex semiconductor processes without undue difficulty.     

Conflict detection and resolution for goal formation in the fractal manufacturing system

The fractal manufacturing system (FrMS) is based on the concept of autonomously cooperating agents referred to as fractals. A fractal is a set of self-similar agents whose goal can be achieved through cooperation, coordination, and negotiation among the agents for themselves. A fractal has fractal-specific characteristics (e.g. self-similarity, self-organization, self-optimization, goal-orientation, and dynamics), and it also has the characteristics of an agent (e.g. autonomy, mobility, intelligence, cooperation, and adaptability) at the same time. In the FrMS, a goal can be regarded as the status which the system aspires to be in. The goal-formation process (GFP) in the FrMS is a process of generating goals and modifying them by coordination between agents. In the GFP, conflicts may occur between goals, which can drive a system to become inefficient. In this paper, a conflict resolution mechanism via agent-based negotiation is proposed for facilitating the GFP. The scheme deals with non-fixed goals. The mobile agent-based negotiation process (MANPro), in which a mobile agent is used for information-exchanging and problem-solving, is used for negotiations in this scheme. The proposed mechanism is illustrated with a goal formation scenario in an exemplary FrMS.     

Computer simulation-based method to predict packing density of aggregates mixture

A key theory in concrete mix design is maximizing aggregate packing density (PD) of aggregate mixture. Different methods have been presented by researchers to estimate PD of aggregate mixture. One such method is computer simulation that has become increasingly common over the last decade; however, it is usually a time-consuming procedure. In the current study, a method based on computer simulation is proposed for estimating aggregate PD. In this method, aggregates with specific shapes, grading and PDs are substituted by monosized spherical aggregates. An equation is also presented for determining the diameter of equivalent monosized aggregates. The coefficient of friction between the equivalent monosized aggregates is determined in a way that the monosized aggregates will have a PD equal to that of actual aggregates. The proposed method is also used to simulate laboratory experiments conducted by the present authors and other researchers. Comparisons reveal the high accuracy of the proposed simple method in predicting the PD of aggregate mixtures.     

Biogeography-based particle swarm optimization with fuzzy elitism and its applications to constrained engineering problems

In evolutionary algorithms, elites are crucial to maintain good features in solutions. However, too many elites can make the evolutionary process stagnate and cannot enhance the performance. This article employs particle swarm optimization (PSO) and biogeography-based optimization (BBO) to propose a hybrid algorithm termed biogeography-based particle swarm optimization (BPSO) which could make a large number of elites effective in searching optima. In this algorithm, the whole population is split into several subgroups; BBO is employed to search within each subgroup and PSO for the global search. Since not all the population is used in PSO, this structure overcomes the premature convergence in the original PSO. Time complexity analysis shows that the novel algorithm does not increase the time consumption. Fourteen numerical benchmarks and four engineering problems with constraints are used to test the BPSO. To better deal with constraints, a fuzzy strategy for the number of elites is investigated. The simulation results validate the feasibility and effectiveness of the proposed algorithm.     

Load torque estimation problem for electric machinery servo systems by an inverse method

Abstract

This study analyzes a load torque estimation problem for an electric machinery servo system using the recursive input estimation (RIE) algorithm. This study also presents four novel estimation models, which have different numbers of sensors, considers the process and the measurement noise to estimate load torque without a torque sensor. The proposed algorithm is a novel estimation technique for solving force feedback and state estimation problems for an electric machinery servo control system. In this work, the DC servomotor system is utilized as an electric machinery servo system, and four important RIE characteristics were verified by numerical simulation results: (i) Adaptive forgetting factor in the RIE algorithm can estimate load torque more effectively than constant forgetting factor; (ii) The significant variation of control input and/or load torque impact affects the estimation precision; (iii) The high‐performance load torque estimation model can be established based only on the given control input and techogenerator sensor; (iv) Using different degrees of model errors, the estimation of the performance tendency toward good or bad for control input and load torque are the same. This characteristic shows that good estimated performance for an unknown load torque can be confirmed based on the strength of good control input estimated performance. In other words, model error can be identified via the deterministic control input and its estimation results. The proposed models may have practical applications for solving disturbance compensation problems in electric machinery servo systems.     

Certain investigations on the reduction of side lobe level of an uniform linear antenna array using biogeography based optimization technique with sinusoidal migration model and simplified-BBO

In this paper, we propose biogeography based optimization technique, with linear and sinusoidal migration models and simplified biogeography based optimization (S-BBO), for uniformly spaced linear antenna array synthesis to maximize the reduction of side lobe level (SLL). This paper explores biogeography theory. It generalizes two migration models in BBO namely, linear migration model and sinusoidal migration model. The performance of SLL reduction in ULA is investigated. Our performance study shows that among the two, sinusoidal migration model is a promising candidate for optimization. In our work, simplified – BBO algorithm is also deployed. This determines an optimum set value for amplitude excitations of antenna array elements that generate a radiation pattern with maximum side lobe level reduction. Our detailed investigation also shows that sinusoidal migration model of BBO performs better compared to the other evolutionary algorithms discussed in this paper.     

Metaheuristics with Deep Learning Empowered Biomedical Atherosclerosis Disease Diagnosis and Classification

Atherosclerosis diagnosis is an inarticulate and complicated cognitive process. Researches on medical diagnosis necessitate maximum accuracy and performance to make optimal clinical decisions. Since the medical diagnostic outcomes need to be prompt and accurate, the recently developed artificial intelligence (AI) and deep learning (DL) models have received considerable attention among research communities. This study develops a novel Metaheuristics with Deep Learning Empowered Biomedical Atherosclerosis Disease Diagnosis and Classification (MDL-BADDC) model. The proposed MDL-BADDC technique encompasses several stages of operations such as pre-processing, feature selection, classification, and parameter tuning. Besides, the proposed MDL-BADDC technique designs a novel Quasi-Oppositional Barnacles Mating Optimizer (QOBMO) based feature selection technique. Moreover, the deep stacked autoencoder (DSAE) based classification model is designed for the detection and classification of atherosclerosis disease. Furthermore, the krill herd algorithm (KHA) based parameter tuning technique is applied to properly adjust the parameter values. In order to showcase the enhanced classification performance of the MDL-BADDC technique, a wide range of simulations take place on three benchmarks biomedical datasets. The comparative result analysis reported the better performance of the MDL-BADDC technique over the compared methods.     

IoMT-Based Smart Monitoring Hierarchical Fuzzy Inference System for Diagnosis of COVID-19

The prediction of human diseases, particularly COVID-19, is an extremely challenging task not only for medical experts but also for the technologists supporting them in diagnosis and treatment. To deal with the prediction and diagnosis of COVID-19, we propose an Internet of Medical Things-based Smart Monitoring Hierarchical Mamdani Fuzzy Inference System (IoMTSM-HMFIS). The proposed system determines the various factors like fever, cough, complete blood count, respiratory rate, Ct-chest, Erythrocyte sedimentation rate and C-reactive protein, family history, and antibody detection (lgG) that are directly involved in COVID-19. The expert system has two input variables in layer 1, and seven input variables in layer 2. In layer 1, the initial identification for COVID-19 is considered, whereas in layer 2, the different factors involved are studied. Finally, advanced lab tests are conducted to identify the actual current status of the disease. The major focus of this study is to build an IoMT-based smart monitoring system that can be used by anyone exposed to COVID-19; the system would evaluate the user’s health condition and inform them if they need consultation with a specialist for quarantining. MATLAB-2019a tool is used to conduct the simulation. The COVID-19 IoMTSM-HMFIS system has an overall accuracy of approximately 83%. Finally, to achieve improved performance, the analysis results of the system were shared with experts of the Lahore General Hospital, Lahore, Pakistan.     

E-mail Spam Classification Using Grasshopper Optimization Algorithm and Neural Networks

Spam has turned into a big predicament these days, due to the increase in the number of spam emails, as the recipient regularly receives piles of emails. Not only is spam wasting users’ time and bandwidth. In addition, it limits the storage space of the email box as well as the disk space. Thus, spam detection is a challenge for individuals and organizations alike. To advance spam email detection, this work proposes a new spam detection approach, using the grasshopper optimization algorithm (GOA) in training a multilayer perceptron (MLP) classifier for categorizing emails as ham and spam. Hence, MLP and GOA produce an artificial neural network (ANN) model, referred to (GOAMLP). Two corpora are applied Spam Base and UK-2011 Web spam for this approach. Finally, the finding represents evidence that the proposed spam detection approach has achieved a better level in spam detection than the status of the art.