فصلنامه مدل سازی در مهندسی
پیاپی 58 (پاییز 1398)

Nowadays, with the progresses in technology to solve problems where there is no exact mathematical relationship between input and output, neural networks are efficiently proposed and used. In the shadow of its unique features, in this study, two multilayer perceptron neural networks including feedforward artificial neural network (FFANN) and cascade artificial neural network (CANN) were proposed to predict the surface tension of imidazolium-based ionic liquids. To verify the validity of the proposed models, 1251 experimental data points were collected from various previously published literature including the surface tension of 40 ionic liquids in a wide range of temperatures (from 263.61 to 533.2 K). The results showed that the proposed CANN consists of three inputs including molecular weights of anionic and cationic part of ionic liquid and temperature with a hidden layer containing 8 neurons with a hyperbolic tangent activation function and trained with Levenberg–Marquardt algorithm has the best correlative capability for surface tension of ionic liquids. In addition, error analysis of test data set with an average absolute relative deviation percent of 1.07 indicates the appropriate performance of the nonlinear CANN model in the linking between network inputs and surface tensions. Also, comparing the accuracy of the proposed model with existing models, including the corresponding states principle, Parachor, the group method of data handling (GMDH) and the model based on least-squared supported vector machine (LSSVM) indicate the superiority of the proposed model.

Fossil fuels are going to either be gone or they are going to become too expensive to realistically use. This issue along with the air pollution caused by these kinds of fuels provides specific concentrations for renewable energies. Most of the common methods for electricity generation by use of solar energy are not economic. However, among the methods, solar plants equipped with parabolic collectors are more acceptable. A parabolic trough collector (PTC) is a system in which reflective parts (like mirror or aluminum plates) are installed in parabolic form upon steel bases. Parallel arrays of such collectors are commonly installed subsequently in vast open fields to form a power plant, called solar farm. These power plants are usually subjected to high wind speeds without much shelter or protection. The aerodynamic loads due to wind blow should be reduced as far as possible. Designing the collector arrangement as well as collectors structures is of most important sections of establishing such power plants due to the wind force effects. In the present study, methods of calculating wind force and the effective parameters on the drag force are discussed. Wind force calculation is accomplished using CFD. Aerodynamic factors are calculated in the 3D model and in consequence shear analysis is performed for the collector and its structure. Simulating the problem in three-dimension provides the condition for more precise modeling of flow in between collectors distance, pressure distribution in collectors and the exerted moment on collectors structures.

In this paper, numerical analysis of flow and heat transfer in a two-dimensional channel with porous medium and based on the lattice Boltzmann method is performed. The main reason for placing the porous medium inside the channel is to reduce the effective cross section of the channel and increase the localized mean velocity to increase the rate of heat transfer and control the maximum temperature of the hot plate. Also, due to passage of a part of the flow from inside of the porous medium, a sudden drop in pressure is prevented by reducing the cross-section. For this purpose, a porous medium with a random structure is considered in the upper part of the channel. The results include the effects of effective parameters such as Reynolds numbers, Prandtl number, aspect ratio and porosity of the porous medium on the maximum temperature, average Nusselt number and pressure drop. Results show that increase of the porosity increases the difference between the maximum temperature of simple channel and channel with porous medium. It is also observed that the use of a porous medium in the low Reynolds numbers creates a higher decrease in the hot plate temperature. In addition, increasing the Prandtl number initially increases the maximum temperature of the hot plate and its further increase will reduce the maximum temperature.

Due to the development of metro lines, always the maintenance of buildings of injuries in urban environments is important, therefore, investigation of the surface settlement of the tunnel is undeniable. Several researchers have yet studied the ground surface settlement. With increasing in number of metro lines, interaction between them will be important. So, in the follow-up to this researches, in this paper the interaction between the tunnels on the surface settlement using numerical modeling software FLAC3D is investigated. For this purpose intersection between metro lines in Tabriz (Line 1, 2 and 3 around of Daneshsara square), that the most important issue is the interaction between metro tunnels in Tabriz city, is considered. In this study, the EPB machine system used for excavation of tunnel. The modellings indicates that passing the line 3 metro tunnels of Tabriz under the line 1 makes a 139.4 % increase in surface settlement. Studies have also shown that passing the line 3 tunnels above the line 2 tunnel makes a 56 % increase in surface settlement.

Laminar mixed convection of a nanofluid consists of water and Al2O3 in a horizontal tube has been studied numerically. Continuity, momentum and energy equations are discreted with the volume control approach and the second-order accuracy and coupling pressure and velocity with the SIMPLE algorithm is done. To include free convection, variation of fluid density with temperature varies according to the Boussinesq equation. Two-phase mixture model has been used by using of effective properties for modeling of two-phase nanofluid flow. Meshing has been done with the organization, and next to the walls, boundary layer mesh has been created. Investigating the results obtained in laboratory conditions indicates the acceptable accuracy of numerical simulations. Simulation has been performed in two states of constant the Richardson number and heat flux, steady state and the variation of the heat transfer coefficient and the Nusselt number are presented in graphical form in various conditions. The results show that, in the case of constant heat flux, the choice of suitable formula for the effective thermal conductivity coefficient will have a significant effect on the accuracy of the obtained results.

This paper investigates the effect of the area of the horizontal stabilizer of a single-main-rotor helicopter on its dynamic responses, stability properties, and flight handling quality during forward flights. Taking linear aero-dynamical relations into account, and assuming a linear induced velocity for the rotor and two degrees of freedom of flapping and feathering for each blade, we have derived the six degrees of freedom nonlinear dynamical model of the helicopter consisting of a main rotor, tail rotor, horizontal stabilizer, vertical stabilizer, fuselage, and the thrust system. The paper covers our findings on the stability derivatives, the effect of the area of the horizontal stabilizer on the forward-flight stability of the helicopter. The results indicate that the trim flight conditions do not considerably change when the horizontal stabilizer area increases. Moreover, increasing the area of the horizontal stabilizer threefold has been found to improve the longitudinal stability of the helicopter—the poles of the open-loop system are transferred to the stability region. However, such stability is accompanied by a favorable overshoot and the settling time required for flight at level 1(ADS-33E); (overshoot less than 30% and the settling time less than 5 seconds.)

The complexity of engineering problems and the existence of various constraints on these issues, encourage the researchers to use of innovative methods based on a heuristic algorithm to find the optimal solution for practical problems at a cost-effective time and non-consistency tolerance. A distinction has been made between various issues and, therefore, extensive research has been done to improve heuristic algorithms in order to enhance their ability to solve engineering and practical problems. In this paper, due to the ability to global search some of the metaheuristic search patterns (such as the EMA algorithm) and the ability to local search for some meta-heuristic search patterns (such as the FPA algorithm), a novel combination method is proposed to use the ability of both types of algorithms. Then, using the proposed method, a hybrid search pattern with new abilities is presented, whose abilities are proven on standard benchmark testing functions as well as solving engineering problems.

Deep learning is one of the most important scopes of the Machine Learning that includes some important architectures. Deep Convolutional Neural Network is one of the attractive architectures that uses in digital image processing. In this paper, we use the Alexnet model for face recognition from input images. We fine-tune the Alexnet model by converting one or two fully connected layers to convolutional layers as well as using the suitable filters. To improve the robustness of the model in coping with the situations that some parts of the input images damaged, we use five crops of the input images including five pixel areas. Furthermore, to visualize the output of each layer, we use the Deconvolution technique in our method. The output of some convolutional and activation layers has been shown. Using this technique, we obtain the Heat-map of the image. To show the results, we use the LFW and Caltech faces datasets. After pre-processing the images of datasets, we compare the results of the Alexnet model in two states: before fine-tuning and after fine-tuning. The results show the recognition accuracy improvement of the fine-tuned models on input images.

Using multi transmission line (MTL) model, a compact circuit model for mixed multiwall carbon nanotube (MMWCNT) interconnects is proposed. Using the proposed circuit model, an algorithmic model is proposed for calculating the transfer matrix of these interconnects. In the proposed circuit model and also algorithmic model, capacitive, inductive and tunneling effects between the carbon layers is considered. Moreover the concept of distributed circuit block parameters is proposed for this type of interconnects. This model is compact because it can be used in wide range and technology of interconnects. Also using this model, any change in the physical parameters of the nanotubes can be considered in the circuit model and algorithmic model. Using the circuit model and algorithmic model one can calculate various stability and time domain responses for MMWCNT interconnects in VLSI circuits. In this paper after proposing the circuit model and algorithmic model, we have calculated step time response and Nyquist response of MMWCNT interconnects. The results show that by increasing the length and diameter of the tubes, the delay of the interconnects is increased and its stability is increased too.

In this paper, a CMOS operational amplifier (op-amp) for applications requiring a bandwidth several hundreds of MHz will be designed and optimized. The op-amp is two-stage and compensated by current buffer and a Miller capacitor. In order to reduce the occupied silicon area, the compensation capacitor has been replaced with a capacitance transistor (MOSCAP). The most important issue in designing compensation networks for amplifiers and particularly operational amplifiers is the calculation and selection of the optimal size for the circuit elements. After selecting op-amps designated by the user, the size of the circuit elements including transistors of the op-amps, along with the size of MOSCAPs, is determined by optimization algorithm, and the bandwidth, DC gain, power consumption and chip area will be optimized with this algorithm. Using the proposed technique, analytical relationship from the optimized solutions can be obtained. The obtained relationship indicates what is the best trade-off between phase margin, power and unity gain bandwidth which helps to achieve the desirable properties .A two-stage amplifier based on CBMC techniques has been designed by a commercial 0.18-µm CMOS process. When driving a 1-pF capacitive load, the CBMC amplifier acheives over 70-dB dc gain, 680-MHz gain-bandwidth product (GBW), 65o phase margin, and 350-V/µs average slew rate, while it consumes only 900-µW at a 1.8-V supply. Nearly 34% improvement in bandwidth is obtained while the compensation capacitor size can be reduced to 1/3 of its original size.

In Recent years, it has been mentioned that the Transmission Expansion Planning is an effective way to mitigate the cascading failure impacts in transmission network. Also, solving transmission expansion planning with cascading failure issues via analytical and assessment approach instead of optimization approach causes that the impacts of cascading failure are more effectively mitigated. The Analytical and assessment approach offers a set of critical lines as a suitable candidate placement of new lines. However, installing new lines for all of elements of the set of critical line is an impractical and uneconomical way. In order to overcome this drawback, a new method is proposed in this paper. The proposed method is based on introducing a vulnerable area in transmission network instead of providing a set of critical lines. The vulnerable area is provided via finding the most vulnerable line and considering a bus farther way (adjacent bus) strategy. All of possible transmission expansion plan at this area are considered to find the best transmission expansion plan. Moreover, a transmission network solidification index based on transmission network vulnerability index is proposed in this paper to rank the possible transmission expansion plans at the vulnerable area. The efficiency and feasibility of the proposed method is investigated at IEEE 39-Bus test system. The numerical results show that using the proposed approach is able to mitigate cascading failure impacts in transmission network.

In recent years, using power electronic equipment in residential and office buildings, commercial complexes, manufacturing unit's loads such as heating, lighting, domestic appliances have significantly increased. Since, most of them are single-phase with variable load pattern, LV power systems are imbalanced and harmonic polluted. The presence of harmonics in imbalance power systems can generate vital problems such as excessive neutral current losses and costs. Therefore, a detailed study into these systems is essential. In this paper, instantaneous neutral power is used as an indication of identifying imbalances in low voltage systems. By applying Matrix Pencil Method to instantaneous neutral power, and based on the rules introduced, the harmonics that have contributions in unbalancing of the system are found. According to this result and according to another set of proposed rules, the loads that affect power losses are identified in order to be switched between phases and, further, to optimize the loss in neutral conductor.

One of the most important challenges for the adaptive filtering is the slow convergence rate of adaptive algorithm against highly correlated input signals. The affine projection adaptive algorithm (APA) is an extension of the well-known normalized least mean square (NLMS) algorithm which achieves a higher convergence rate in both full-band and sub-band structures. In this paper, to further improve the convergence rate of the algorithm against high-correlation input signals in the application of sparse linear system modeling, a sub-band APA is proposed in which the number of projection vectors is determined as a function of the estimated sub-band mean square error (MSE). In addition, variable sub-band step-sizes are proposed as a function of filter weights and the estimated MSE such that at the initial convergence stage, bigger weights make an increased contribution to the adaptation process while during convergence, the contributions approach the same amount. The proposed idea improves the convergence rate and lowers the steady-state MSE. Simulation results for the sparse linear system modeling and for the application of acoustic echo cancellation, verify the superiority of the proposed algorithm in the convergence performance and estimation accuracy of the acoustic path coefficients over its counterparts.

The present research deals with the free vibration analysis of functionally graded (FG) frames with semi-rigid connection and elastic supports. Flexibility of elastic supports and semi-rigid connection are modeled with linear rotational springs. Moreover, inhomogeneous members (FGM) are modeled as a power function. Accordingly, based on the Euler–Bernoulli beam theory an exact formulation for free vibration analysis of FGM frames with various boundary conditions is obtained in terms of the Bessel functions. In other words, explicit relations are available to find the natural frequency and vibrational modes of semi-rigid frames with inhomogeneous members and elastic rotational restraints. In this way, dynamic properties such as the natural frequency and vibrational modes of these structures with various boundary conditions can be found. Comparing the outcomes with other research demonstrating the accuracy and efficiency of the proposed formulation. Finally, the effects of various parameters such as flexibility of connection and elastic supports, and the functionally graded material on the free vibration response of the structures are investigated.

Since software packages, which predict the charge motion of single ball in the mineral processing tumbling mills, use the Powell (1991) relationships, the accuracy of these relationships was investigated in this research. The test results indicated that the difference between the impact points for two balls with different materials was lower than 2o. A change in friction coefficient had no significant effect on the ball impact point, whereas in the Powell relationships changing the friction coefficient to 0.4 increased the impact point difference 15o. The friction coefficient in the Powell relationships considers to be more than zero at the point of equilibrium. The impact point difference between the results from the Powell relationships and the relationships that considers static friction coefficient (µs) equal to zero at the point of equilibrium were compared. The results indicated that the difference is negligible when µs is near to zero and is high when µs is more than 0.1.

One of the major issues in construction project management is selection of the best alternative for execution of each activity, so that the project progresses with the lowest possible total cost and total time. In the real world, different sources of uncertainty usually adversely affect on activity duration and activity cost. Therefore, it is important to present methods for robust construction project scheduling that are less sensitive to disruptions caused by noise factors. In addition, existing approaches in literature seldom focus on environmentally conscious construction planning. However, obtaining of this aim requires to consider multiple objectives and their relationships, simultaneously. In this study, a roust optimization based method for solving discrete time-cost-enviroment trade-off problem is presented, in which time and cost of activities as uncertain factors are considered. To investigate the performance of the suggested approach, a computational analysis on construction project of protective wall of the east coast of port in the economic zone of Amirabad is performed. the obtained results indicate that the robust model has beter performance in comparison with the corresponding deterministic model.

Scheduling of jobs and maintenance activities is a vital problem in a production system. In the current business environment, the competitiveness of manufacturing companies is based on their ability to response the rapid changes in demand with high quality and low costs. Therefore. in this paper, the Scheduling of jobs and maintenance activities in an unrelated parallel machines environment is studied. By considering the controllable processing times, a multi-objective mathematical programming model is presented to minimize the sum of the compression and expansion of processing times and maintenance costs, makespan, tardiness, and earliness. The multi-choice goal programming (MCGP) is applied. The problem is NP-hard, therefore, the NSGA-II algorithm is developed to solve the medium and large size instances. The obtained results from solving the proposed model with NSGA-II and comparing with the results from MCGP demonstrates that the proposed algorithm is presented a non-dominated solution set in a short time.

In this paper, a five-level pharmaceutical supply chain network is modeled under uncertainty. Levels of this pharmaceutical supply chain network include raw material supplier, production center, distribution center, Health centers and customers. Also, the objectives of this paper include minimizing the costs of the total supply chain network, minimizing the maximum unmet demand, and maximizing reliability in timely delivery of drugs, by considering the perishable time and the discount on the purchase of raw materials. Decision-making variables are divided into two strategic and tactical categories and, include determining the optimal number and location of potential facilities and determining the optimal amount of drug flow between the selected facilities, respectively. A robust-possibilistic optimization method has been used to control the uncertain parameters, and the hybrid LP-metric method and Monte Carlo simulation approaches have been used to solve the multi-objective model. Finally, by providing a numerical example, the outputs obtained from the solution of the model have been discussed.

Lack of environmental and economical resources leads to a quest for new alternatives in civil construction. In the recent decades, gypsum was used for plastering material on walls and ceilings. Many researches have been dedicated to natural fibers in order to be used as a reinforcement in cement and gypsum and some of them have been managed to decrease the density while enhancing the mechanical properties of the cement and gypsum composites. In this research, a lingo cellulose fibers which is rice straw in ten composition percentages are used in gypsum accordingly. In addition, Methyl cellulose, lime, starch, alkali lime saturated solution and one molar potassium hydroxide as a corrector of straw are used. From each composite three samples were made, based on the related standards, for assessment of axial and bending strengths However, Rice straw illustrated weak performances in axial and bending strengths it reduces density by 12 percent. While having the latter improvements, processed rice straw increased bending strength by 12 percent. Studying Scanning Electron Microscope (SEM) outputs confirmed the obtained results. Furthermore, utilizing ANFIS in Matlab software a model for predicting properties of new composites and reaching best composites is obtained.

One of the main techniques in the enhancement of heat transfer during recent decades is utilization of nanofluids. In this study, the forced convective turbulent flow of CuO-H2O and SiO2-H2O nanofluids have been numerically investigated. The results obtained for two different conditions in a horizontal tube under constant wall heat flux. In order to investigate the performance of turbulence models, various two-equation turbulence models as well as one Reynolds stress model, exist in ANSYS-FLUENT package, are used to modeling the turbulent flow. The single phase approach also, is used to estimte the thermo physical properties of the nanofluids. The Nusselt number or equivalently the heat transfer coefficient, and the friction factor are calculated in fully developed region of the flow. The experimental data are also used for the sake of comparison and to find which turbulence models has the best performance. To accomplish this, more than 500 runs have been carried out for various Reynolds number as well as different volume fractions up to 4 percent for two dimensionless number of Nusselt and friction factor.

AAbstract: Transfer function of the InGaAs/GaAs SAQD laser is calculated using its rate equations. Using the calculated transfer function, time and frequency analysis can be implemented completely. In the time domain, the calculated transfer function can show the transient and steady state response of the laser. Also this transfer function can be used in circuit simulator such as SPICE in order for analyzing the electro-optical VLSI circuits. After the calculation of the transfer function, time domain together with Nyquist responses has been calculated. The effect of carrier dynamics on the output response of the laser is analyzed in this case. Nyquist diagram is selected in this analysis because one can predict the closed loop system using the open loop system. Unlike the systems such as interconnects and transistors that are analyzed before, the results reveal that in SAQDLs, the ratio of the transient peak overshoot in the output power to stable output power is very large. Accordingly in integrated circuits that use SAQDLs, amplification of each overshoots in output power of the laser can deteriorate the input of the next circuit module on the chip or shock to it. This increases the importance of the stability analysis in SAQDLs for more reliability.

Structural health monitoring systems, by evaluating several parameters of structural health, attempt to detect the faults in early stages, and prevent their expansion. One of the most important components of a building is its beams in which any crack or damage can result in irreparable financial and psychological damage. Using motion sensors and analyzing their data is one of the methods for monitoring the structural health and detecting its cracks and damages. Meanwhile, wavelet transformation is one of the most commonly used diagnostic tools. In current research, a method based on combination of a discrete wavelet transform and a digital filter is proposed to detect the reinforced concrete beam defects. Various fault modes have been simulated using Abaqus software and Mode shapes have been analyzed using Matlab software under discrete wavelet transform and digital filter. The results indicate a very accurate performance of the proposed method in the detection of millimeter cracks, which shows about 10-times improvement in the resolution, compared with the previous works.

Increasing the electrical resistance and reducing the conductivity of nanotubes due to the impact forces have attracted a lot of their use as power sensors. In this paper, the impact of a nanoparticle on simply supported and clamped double-walled nanotubes has been investigated by using elasticity nonlocal theory. The spring constant for each nanobeam, which is influenced by the geometry of the structure and its shear and bending deformations, has been obtained using mass-spring system. The results of this analytical method have been compared with the results in the background research. According to the results, by increasing the speed, the dynamic deformation in the middle of the double-walled nanobeam increases. The dynamic deformation in the double-walled nanotube is smaller than single-walled nanotube and the maximum dynamic deformation of the double-walled nanotubes is increased by increasing the mass of nanoparticle. Also by considering the van der Waals force between the layers of double-walled nanotube, maximum dynamic deformation in the middle of nanobeam reduces.

At present, spectrometric chromatography are used to methane in the fermentation process, but the use of off-line mass spectrometry requires the sampling and maintenance of the solution. Silicon membrane-based gas sensors are designed and used to detect methane based on this principle. The simulation of this sensor was carried out using COMSOL Multhiphysics (CM) 5.2 A software in 3D. Input gas helium was applied at three values of 40, 50 and 60 ml/min. Also, silicon tubes with lengths of 5, 10 and 15 cm and in three diameters of 0.05, 0.05, and 0.25 cm were investigated. The shortest response time from the sensor simulation is 10% compared to the results of the experimental at 1.6 minutes. The shear stresses, the effect of inlet gas velocity, flow pressure and fluid concentration in this study were compared. All result showed that the silicon membrane-based have been best absorbent for gas specially Methane.