فصلنامه مدل سازی در مهندسی
پیاپی 61 (تابستان 1399)

Pipeline systems are widely used in gas refineries, oil factories chemical power plants. It is important to measure the fluid temperature in this pipeline. Numerous cases, however, exist where the internal fluid temperature is not available and only an outside wall temperature measurement is feasible. Recently inverse heat transfer problems have been analyzed for the situations where a direct measurement of boundary/initial conditions, or thermophisical properties are difficult, or indeed impossible, to carry out. Various numerical approaches have been proposed to estimate fluid temperature in the pipes. In this study, the analytical method is applied to solve the one-dimensional inverse heat conduction problem, in order to estimate fluid temperature in the pipes, based on the exact outer wall temperature. The fluid temperature is assumed to be a polynomial. The unknown polynomial coefficients are found by combining the method of Duhamel's theorem and the least squares method. The results show that the fluid temperature in the pipes can be accurately estimated by using the inverse heat transfer method.

In this paper, an ultra-wideband (UWB) bandpass filter is designed by using the stepped impedance method. A stepped impedance structure with low order is proposed, then some of them are used in series. First only one short circuit stub is used then some of them are combined according to frequency response. The result of simulation depicts that the passband is 3.356 to 22.81 GHz, it means bandwidth is over than 19 GHz, therefore the fractional bandwidth is 148.7%, which it is good compared to papers of recent years. Ripple is less than 1dB in band. The insertion loss is over than 25 dB in out of band. The presented filter size is 16×16 mm2 that it is compact. The designed filter is fabricated on RO4003 substrate with 3.38 dielectric constant, 32 mil thickness, and 0.0026 loss tangent. The simulation is performed with numerical software and they confirm each other's results.

Due to real-time nature of multimedia data, the 802.11ac standard applies the EDCA method to prioritize different types of traffics. The EDCA method supports four different access categories each with its own queue so that the highest priority is assigned to voice packets. However, the EDCA method only works well when the amount of voice packets is small compared to other types of traffics. As the density of the voice packets increases, the collision among them increases which results in delay and loss ratio increasing. This in turn can highly degrade the quality of the VoIP service. To address this EDCA issue, a new dynamic model called D-EDCA is proposed in this work. The new D-EDCA model provides a deterministic QoS by continually monitoring the voice packets and their corresponding queue to dynamically adjust the corresponding parameters when the density of the voice packets is above a threshold. The D-EDCA model is designed and implemented using NS3 simulator tool. Furthermore, in order to validate and compare the performance of the proposed D-EDCA with the conventional EDCA, a learning-based framework is designed that operates in three phases in the uplink mode of 802.11ac network. The results prove that the D-EDCA model can highly improve VoIP communication compared to the conventional EDCA in terms of network metrics.

The unwanted operation of the distance relay during the power swing conditions can lead to increased disturbances and exacerbate of the power grid. Therefore, the rapid and accurate detection of the power swing and blocking of the distance relay after a power swing is necessary to maintain the security and reliability of the power grid. On the other hand, for a fault condition during the power swing, in order to maintain the dependability index of the protective system, it is necessary to identify the fault. This paper presents an intelligent and time adaptive algorithm for detecting symmetric and asymmetric faults in series compensated transmission lines through the long short-term memory (LSTM) recurrent neural network. This method uses three-phase currents in the distance relay point as input. In order to investigate the proposed algorithm, the reference power system for transmission-line relay testing introduced by the IEEE Power System Relaying Committee (PSRC), was considered. Different fault types in different conditions such as fault location, fault resistance, load angle and fault inception time were modeled and simulated in PSCAD software. The results show that the proposed method has an average response time (ART) and an average accuracy (AA) of 0.1004 ms and 99.04%, respectively.

Today, due to the importance of transportation in developing countries, addressing the issues related to transportation safety planning has also become increasingly important. One of the important issues in the field of safety planning is the synchronization of safety planning with the stages of transportation planning and predicting large-scale accidents is one of its components. The purpose of this research is to present a mathematical model for predicting the number of severe road traffic accidents along with transport planning and determine the most appropriate planning stage for building macro level accident prediction models. In this paper, by examining the factors affecting accidents, the macro model is presented using genetic programming method. For this purpose, data from the years 2011 to 2014 were used as baseline data and statistics from year 2015 were used to validate the model. Based on the data of each planning step, separate models were developed and after comparison between them, the model based on travel production variables was identified as the optimal model. The results show that among the data corresponding to the transport planning stages, the variables of the trip generation stage are the most appropriate data set for modelling of predicting road freight accidents.

In this paper, a single-product, single-machine system under Markovian deterioration of machine condition and demand uncertainty is studied. The objective is to find the optimal intervals for inspection and preventive maintenance (PM) activities in a condition-based maintenance planning with discrete monitoring (CBMDM) framework. At first, a stochastic dynamic programming model whose state variable is the machine status is presented. This model whose objective is minimizing inspection, preventive maintenance and lost production costs due to the difference in actual capacity with nominal capacity, does not take into account demand. Then, demand is appended to the state variable in the second model and the average cost of lost production which is due to the difference in actual production capacity with demand is replaced in the first model correspondingly. Finally, to validate and analyze the proposed models, an application of the models in wind turbines is prepared. The numerical results show that replacing demand by nominal capacity in simultaneous inspection and preventive maintenance planning, the average total costs consist of inspection and preventive maintenance in the planning horizon are reduced in the planning horizon.

One of the main challenges in speech recognition is noise resistant feature extraction. In this paper, a new feature extraction algorithm, called Fractional and Adaptive Power Normalized Cepstral Coefficients Algorithm, has been proposed as a noise-resistant method for speech recognition. This proposed feature extraction method is based on a fractional short-term Fourier Transform. The selection of fractional conversion coefficient is important for proper analysis of multi-component signals like speech. Therefore, the proposed method obtains the optimum parameter of α for fractional Fourier Transform based on the noise class in the environment, adaptively by the Differential Evolution meta-heuristic algorithm. Moreover, TI Digit and Noisex-92 are used for evaluation of the resistance and accuracy of the recognition of the automatic speech recognition system. Simulation results show more resistance and higher recognition accuracy of the proposed feature extraction method rather than other methods in noisy and without noise environments. In the proposed ASR system, the Support Vector Machine (SVM) classifier with a nonlinear kernel has been used. Also, all the simulations are performed in MATLAB.

In this paper, two novel and simple approaches are presented for the evaluation of friction in the metal forming processed. By employing the data obtained from a cylinder compression test in addition to the flow curve, the friction coefficient of the process can also be evaluated. Hence, no extra measurement is required. Two new exponential parameters are introduced, geometrical and force barreling factors which are defined based on the change in geometry and forging force of the deformed cylinder, respectively. Numerical simulations are carried out to determine friction calibration curve diagram of cylindrical samples under various friction conditions. Then, by plotting the experimental data on the calibration curve diagram, the coefficient of friction of the process is determined. The results of the friction coefficients obtained from the geometrical and force barreling approaches are the same. The conventional ring compression tests are also performed for the corresponding lubricants to validate the proposed approaches. It is observed that the friction coefficients predicted by the presented approaches are identical to those obtained from the ring compression test. Also, friction coefficients that are greater than 0.2 can be determined more precisely than the ring compression test due to the increase in the distance of the curves.

In this study, we investigated the flow of blood plasma in which white blood cell, erythrocyte and platelet nanoparticles were suspended with a percentage of actual concentrations inside a three-dimensional rectangular micro channel. The chamber rotates around the vertical axis at constant angular velocity and is affected by electrical double layer .The numerical results on white blood cell, erythrocyte and platelet nanoparticles were extracted in plasma at different angular velocities and the movement of suspended particles in the blood was studied that was observed that in the angular velocity increases, the pressure decreases along the channel and along the axis of rotation of course the velocity has the maximum amount in the center of channel. Numerical studies of the velocity and distribution of particles also show that the velocity variations are the same in all particles and the velocities of white blood cells, erythrocytes and platelets at the end of the channel are more than another place and the streamlines more compact.

A seamless graphene inverter including graphene nanoribbon field effect transistor (GNRFET) and graphene interconnect is proposed. The seamless structure is suggested to eliminate the ohmic, schottky, and parasitic resistances in the junction of the traditional interconnects with the Gate, Source and Drain of GNRFET. After that, using the circuit models of the graphene devices that are used in the proposed structure, transfer matrix model of the proposed seamless graphene inverter is calculated and extracted. All of the capacitive, inductive and scattering effects are included in the assumed circuit models of the GNRFET - graphene interconnect and consequently in the overall matrix model of the seamless graphene inverter. Elimination of the ohmic, schottky and parasitic resistances causes to improve in the working speed of the proposed inverter. Extraction of the transfer matrix model of the seamless graphene inverter and calculation of its step time response, relative stability and frequency bandwidth confirms this improvement. The advantage of the transfer matrix model of the proposed inverter is that any change in the physical parameters of the graphene nanoribbons that are used in the structure can be included in the model and one can analyze the effect of it in all of the technology nodes. Using the circuit model and the extracted transfer matrix, anyone can evaluates various stability analyses such as Nyquist, Bode and Nichols together with the time-frequency responses of the graphene seamless inverter used in very large scale integrated (VLSI) circuits.

This research was carried out to optimize energy consumption in Isfahan Steel P-type boilers by numerical modeling of heat transfer in combustion chamber at different angles of natural gas fuel burner. Using the Ansys-Fluent combustion chamber modeling software, four natural gas burners, 187 firefighting tubes and its refractory walls were used. The results show that changing the angle of the burner is a good way to control the flame. By decreasing the angle of the burner, the heat flux on the firebox tubes the temperature and fluid velocity inside the combustion chamber increases and thus improves the boiler efficiency. However, reducing the angle of the burners results in an increase in the surface temperature of the fire pipes and can even pierce and damage the equipment. The results also showed that with constant amount of gas consumed, the ratio of radiation heat transfer to total heat transfer of firewall tubes at different angles is 90% on average. Also the agreement between numerical and experimental results including the existing boiler condition in which the burner at 30 ° angle confirms the modeling accuracy.

Prediction of cutting forces by analytical modeling methods in the machining process has a considerable importance. In this study, mechanistic modeling was used to predict cutting forces in milling process by two-cutting edge end mill tool with a straight cutting edge κ = 90°. This modeling approach due to no need for costly experiments to obtain material behavior information in the high strain rate cutting process and due to no need for time-consuming analysis by numerical software as well as the application of material characteristics and cutting conditions with coefficients of cutting and cutting edge have acceptable performance and accuracy. In this research, for the first time, a special design for the work piece was considered to eliminate the effect of round insert tip on modeling error. The results of comparing the predicted and experimental cutting force curves obtained from the dynamometer for the three cutting force components〖 F〗_x, F_yand F_z show acceptable agreement.〖 F〗_x, F_yand F_z show acceptable agreement.