Sharif Digital Repository

One solution to reduce fuel consumption (FC) of vehicles is to use infinitely variable transmission (IVT) as the power train. However, inappropriate design of IVT leads to an inadequate decrease in the vehicle FC. Therefore, in order to reach a maximum fuel economy, an optimization on the power train proves to be necessary. The present article aims to optimize IVT to decrease the vehicle FC in a driving cycle. The results have revealed that the FC of the vehicle equipped with the optimized IVT is approximately 7% and 4.7% compared to the cases of using the optimized six-speed and nine-speed manual transmissions, respectively  

A control strategy on a hybrid vehicle can be implemented through different methods. In this paper, the Cerebellar Model Articulation Controller (CMAC) and Radial Basis Function (RBF) neural networks were applied to develop an optimal control strategy for a split parallel hydraulic hybrid vehicle. These networks contain a nonlinear mapping, and, also, the fast learning procedure has made them desirable for online control. The RBF network was constructed with the use of the K-mean clustering method, and the CMAC network was investigated for different association factors. Results show that the binary CMAC has better performance over the RBF network. Also, the hybridization of the vehicle... 

Intelligent deep-drawing is an instrumental research field in sheet metal forming. A set of 28 different experimental data have been employed in this paper, investigating the roles of die radius, punch radius, friction coefficients and drawing ratios for axisymmetric workpieces deep drawing. This paper focuses an evolutionary neural network, specifically, error back propagation in collaboration with genetic algorithm. The neural network encompasses a number of different functional nodes defined through the established principles. The input parameters, i.e., punch radii, die radii, friction coefficients and drawing ratios are set to the network; thereafter, the material outputs at two... 

The aim of this article is design, fabricate, and control a conceptual prototype of a flexible wearable robot to increase fingers’ force using a novel cable mechanism. In this robot, force and position are controlled in separate phases by a fuzzy system equipped with emotional learning. A Flexible structure has been chosen for this robot, because flexible gloves are lighter and much more suitable for use in daily tasks compared to rigid structures. Since this system is supposed to be substituted for weak or damaged limb, it should have a behavior similar to the body organs as much as possible. Therefore, a novel mechanism, inspired by the natural movement mechanism of the human hand, has... 

The aim of this article is design, fabricate, and control a conceptual prototype of a flexible wearable robot to increase fingers’ force using a novel cable mechanism. In this robot, force and position are controlled in separate phases by a fuzzy system equipped with emotional learning. A Flexible structure has been chosen for this robot, because flexible gloves are lighter and much more suitable for use in daily tasks compared to rigid structures. Since this system is supposed to be substituted for weak or damaged limb, it should have a behavior similar to the body organs as much as possible. Therefore, a novel mechanism, inspired by the natural movement mechanism of the human hand, has... 

The aim of this article is design, fabricate, and control a conceptual prototype of a flexible wearable robot to increase fingers’ force using a novel cable mechanism. In this robot, force and position are controlled in separate phases by a fuzzy system equipped with emotional learning. A Flexible structure has been chosen for this robot, because flexible gloves are lighter and much more suitable for use in daily tasks compared to rigid structures. Since this system is supposed to be substituted for weak or damaged limb, it should have a behavior similar to the body organs as much as possible. Therefore, a novel mechanism, inspired by the natural movement mechanism of the human hand, has... 

This article presented an efficient deep learning technique to predict the compressive strength of high-performance concrete (HPC). This technique combined the convolutional neural network (CNN) and genetic algorithm (GA). Six CNN architectures were considered with different hyper-parameters. GA was employed to determine the optimum number of filters in each convolutional layer of the CNN architectures. The resulted CNN architectures were then compared to each other to find the best architecture in terms of accuracy and capability of generalization. It was shown that all of the proposed CNN models are capable of predicting the HPC compressive strength with high accuracy. Finally, the best of... 

Despite development of accurate musculoskeletal models for human lumbar spine, the methods for prediction of muscle activity patterns in movements lack proper association with corresponding sensorimotor integrations. This paper uses the directional information of the Jacobian of the musculoskeletal system to orchestrate adaptive critic-based fuzzy neural controller modules for controlling a complex nonlinear redundant musculoskeletal system. The proposed controller is used to control a 3D 3-degree of freedom (DOF) musculoskeletal model of trunk, actuated by 18 muscles. The controller is capable of learning to control from sensory information, without relying on pre-assumed model parameters.... 

The objective of this research is geometrical optimization of half toroidal Continuously Variable Transmission (CVT) in order to achieve high power transmission efficiency. The dynamic analysis of CVT is implemented and contact between the disk and the roller is modeled viaelastohydrodynamic (EHL) lubrication principles. Computer model is created using geometrical, thermal and kinetic parameters to determine the efficiency of CVT. Results are compared by other models to confirm the model validity. Geometrical parameters are obtained by means of Particle Swarm Optimization (PSO) algorithm, while the optimization objective is to maximize the power transmission efficiency. Optimization was... 

Metal matrix composites (MMCs) have received considerable attention due to their excellent engineering properties, but their poor machinability has been the main deterrent to their substitution for metal parts. Optimization of machining parameters such as cutting speed, feed rate and depth of cut will improve the machinability of this material. This paper represents application of artificial neural network (ANN) model and genetic algorithm to study the machinability aspects of Al/SiC-15% produced by powder metallurgy process and to obtain optimum machining conditions. A multilayer feed forward ANN has been employed to study the effect of machining parameters on three aspects of machinablity,... 

A six-degree-of-freedom musculoskeletal model of the lumbar spine was developed to predict the activity of trunk muscles during light, moderate and heavy lifting tasks in standing posture. The model was formulated into a multi-objective optimization problem, minimizing the sum of the cubed muscle stresses and maximizing the spinal stability index. Two intelligent optimization algorithms, i.e., the vector evaluated particle swarm optimization (VEPSO) and nondominated sorting genetic algorithm (NSGA), were employed to solve the optimization problem. The optimal solution for each task was then found in the way that the corresponding in vivo intradiscal pressure could be reproduced. Results... 

In this research, the design of a solar reverse osmosis desalination plant was investigated by integrating various components using TRNSYS and ROSA software. To this goal, a two-stage reverse osmosis system with 50% recovery in the city of Chabahar was modeled. The calculations were performed in three different case studies, i.e., a photovoltaic power plant, a solar collector power plant with Organic Rankine Cycles, and a photovoltaic thermal power plant with Organic Rankine Cycles, with the reverse osmosis desalination plant being a novel investigation. Water production and electrical energy generation of each case study were evaluated both on a daily and yearly bases. The simulation... 

Wind speed is the most important parameter in the design and study of wind energy conversion systems. The weibull distribution is commonly used for wind energy analysis as it can represent the wind variations with an acceptable level of accuracy. In this study, the wind data for 11 cities in Iran have been analysed over a period of one year. The Goodness of fit test is used for testing data fit to weibull distribution. The results show that this data fit to weibull function very well. The scale and shape factors are two parameters of the weibull distribution that depend on the area under study. The kinds of numerical methods commonly used for estimating weibull parameters are reviewed. Their... 

This research aims to optimize a multi-zone Air Handling Unit's (AHU) energy consumption by using a Non-linear Model Predictive Control (NMPC) approach. In this paper, Genetic Algorithm (GA) and Non-linear autoregressive network with exogenous inputs (NARX) have been utilized to design NMPC for a multi-zone AHU. The NMPC problem could be divided into two main sections: internal model and the optimizer. NARX serves as the controller's internal model to predict the building's thermal dynamics. GA is then used to solve the NMPC problem and find the optimal value of the control signals at each time step. The proposed NMPC jointly minimizes energy consumption of the AHU and the deviation from the... 

In this paper, we describe an innovative method to model and solve spatio-temporal behavior of nuclear reactor cores via three-dimensional multilayer cellular neural networks. This method uses electrical elements and the existing duality between neutronic and thermal-hydraulic parameters of nuclear reactors. The relevant electrical circuit can be simulated by existing professional electrical circuit software. This research goes beyond our previous efforts to use a neural computing approach in the nuclear field. Modeling and solving simple nuclear reactor kinetic equations is now expanded to a complete dynamic calculation, integrating the core thermal-hydraulic models and the relevant... 

Recurrent neural networks (RNNs) in identification of complex nonlinear plants like nuclear reactor core, have difficulty in learning long-term dynamics. Therefore, in most papers in this area, the reactor core is used to identify just the short-term dynamics. In this paper we used a multi-NARX (nonlinear autoregressive with exogenous inputs) structure, including neural networks with different time steps and a heuristic compound learning method, consisting of off-line and on-line batch learnings. This multi-NARX was trained by an accurate 3-dimensional core calculation code. Network responses show that this procedure solves the difficulty in identification of complex nonlinear dynamic MIMO... 

Different nonelectrical problems can be effectively modeled by their equivalent electrical circuit, using cellular neural network (CNN). Dynamics of such large scale systems with partial differential state equations can be simulated by this technique in real-time. In this paper, we described an originally derived method to model and solve nuclear reactor kinetic equations via multilayer CNN. We proposed an innovative method for online calculation of spatio-temporal distribution of the reactor core neutron flux. One of the main applications of the proposed approach can be development of a new hardware for online simulation and control of nuclear reactor core via very large scale integration... 

In pressurized water reactors, the fuel reloading problem has significant meaning in terms of both safety and economics. An optimal loading pattern is defined as a pattern in which the local power peaking factor (Pq) is lower than a predetermined value during one cycle and the effective multiplication factor (keff) is maximized to extract the maximum energy. This article presents a specialized genetic algorithm for loading pattern design. The tests on well-researched cases have shown that the genetic algorithm is capable of finding better loading patterns than solutions found by direct search methods. However, most of the previous researchers have considered simple fitness (cost) functions;... 

Various methods have been used for solving the neutron transport equation in the past, and a number of computer codes have been developed based on these solution methods. This paper describes a novel method for the solution of the steady-state and time-dependent neutron transport equation using the duality between neutronic parameters in the method of characteristic (MOC) and the electrical parameters in the cellular neural networks (CNN). The relevant electrical circuit can be simulated by professional electrical circuit simulator software, HSPICE. This software is used for numerical solution of the transport equation only by preparation of appropriate inputs. This method does not need... 

Improving the methods of identification and control of nuclear power reactors core is an important area in nuclear engineering. Controlling the nuclear reactor core during load following operation encounters some difficulties in control of core thermal power while considering the core limitations in local power peaking and safety margins. In this paper, a nuclear power reactor core (VVER) is identified using a multi nonlinear autoregressive with exogenous inputs (NARX) structure, including neural networks with different time steps and a heuristic compound learning method, consisting of off- and on-line batch learning. An intelligent nuclear reactor core controller, is designed which...