---

Aim: The objective of this paper is to design nutrient-adequate, varied and cost-efficient diets for diabetes patients.
Methods: A new multi-objective mixed integer linear programming model under uncertainty is developed to design diet plans for diabetes patients.
Findings: The analysis is conducted on the population of 30 years old men and women in 24.99 and 18.5 body mass index, 1.50, 1.65 and 1.80 (m) height categorized in 4 physical activity levels (sedentary, low, active and very active). The objectives of the model are the minimization of the total amount of saturated fat, sugar and cholesterol and the total cost of the diet plans. The constraints of the model are fulfilling the body's nutrient requirements and the diversity control of each patient’s diet. In order to get closer to the real world, fuzzy parameters are considered in the model. To solve the model, a new hybrid solution methodology (Jimenez and epsilon-constraint method) is used to offer the optimal Pareto of non-dominated solutions. Each optimal Pareto of the model consists of diet plans that each patient can choose the proper food based on the taste, availability and cost.
Conclusion: Mathematical modeling of diet planning and study of its optimal solutions can be considered as a decision support tool for the professionals to design the most proper diet plans.

---

Design of a suitable Control Strategy for operating a hybrid propulsion system in different types of roads and driving cycles is one of the most challenging subjects in hybrid vehicle research areas. Intelligent Control Strategies have been designed to meet the above requirement. The control signals in an intelligent control strategy of the hybrid vehicles are generated according to the driving cycle type. This is done by using a driving cycle identification unit. In this paper, design of a fuzzy based driving cycle identifier has been presented. The main idea in this unit is that any arbitrary driving cycle is similar to a group of standard driving cycles according to some degrees of similarity. As a result, the control strategy of the hybrid powertrain in the arbitrary driving cycle is affected by the optimized control strategy of the standard diving cycle based on the degree of similarity. Here, the subset of sufficient features is determined by using the floating search method as a useful feature selection algorithm. Also, the fuzzy clustering method is used to generate the values of similarity degrees to each standard driving cycle. Finally, the performance of the fuzzy driving cycle identification unit is assessed.

---

 

---

In this paper, a real time simulator of the engine-generator for a series hybrid electric bus is designed which can be used for hardware in the loop testing of the hybrid bus control unit. As an important step for designing a hardware in the loop simulator, messages that the engine and generator receive from the vehicle control unit are identified. Design of the simulator is based on these received messages and dynamic behavior of the components. Since the engine and generator receive speed or torque commands from the vehicle control unit, two PID controllers are designed to bring the engine or generator to the desired speed. By applying appropriate inputs, different modes of operation of the diesel-generator is simulated including the soft start process and steady state operation. For a simulator, the ability to interface with a real hardware requires a real time simulation. To do so, the design process is implemented in LabVIEW environment and results show that the designed simulator gives responses close to real test results and it is suitable for developing the hybrid vehicle control unit.

---

Hydraulic engine mounts are applied to the automotive applications to isolate the frame from the high frequency noise and vibration produced by the engine. It also designs to reduce the engine shake motions from the road distribution usually occurred at low frequencies. This implies that the stiffness and damping properties of the engine mount should be amplitude- and frequency- dependent. In the semi-active engine mounts this task will be done by changing the mount parameters such as stiffness and damping. Magneto-rheological fluids are used in the mounts to change their damping by applying the magnetic field. When the current is applied to the electromagnet and the magnetic field is present, the behavior of the magneto-rheological mount is changed by the magneto-rheological effects. In this paper, a prototype magneto-rheological mount was built and experimentally evaluated. Also, the mathematical model of the mount was developed to represent the dynamic behavior of the engine mount system. The model was numerically solved based on the prototype parameters and simulated in MATLAB. The experimental results were used to verify the model in predicting the mount characteristics.

---

Due to the complexity in the control system of hybrid vehicles, the electronic control units should go through extensive testing before getting installed in a prototype vehicle. In the first stages of the development process, Hardware-in-the-Loop (HiL) simulation can not be performed because the hardware components of the vehicle are not yet available. In this case, Model-in-the-loop (MiL) simulation is used which couples the designed control software with an environment simulation with no need for a special hardware. In this paper, the MiL simulation is introduced for verification of the vehicle control software in a series hybrid electric bus. To do so, considering the dynamic behavior of various components of the vehicle, a simulator of the hybrid bus is designed by going through with the input/outputs of the vehicle control software. Using the designed test bench, the user can act as a real driver and experience different driving regimes and analyze the control software commands. In the designed simulator, all subsystems are simulated separately in LabVIEW environment and real-time simulation is achieved with an acceptable error. Therefore, it can be used in a HiL test bench for testing each of the vehicle components. The designed simulation model has been validated using real test results. Using that, results show that all control functions in the vehicle control software can be tested and verified with no cost and in the shortest possible time.

---

The aim of this study is to feasibility study and design of hydraulic hybrid power train system for refuse truck in order to regenerate and store kinetic energy and reuse it for supplying propulsion power of vehicle. The hydraulic hybrid propulsion system includes a conventional internal combustion engine, a hydraulic pump/motor and also the accumulators as the energy storage device. Here, the parallel configuration has been chosen for implementing this powertrain. At first part of the paper, regarding the unique driving trends of refuse trucks, a driving cycle for refuse truck in Tehran has been extracted to improve the reliability of the designed powertrain. Also, AXOR 1828, one of the trucks used as the refuse vehicles in Tehran, has been chosen as the based vehicle. The driving cycle is extracted by performing observations on the based vehicle operation during several days. In the second part of the paper, the components of hydraulic hybrid powertrain have been designed to recuperate as much kinetic energy as possible in the refuse truck driving cycle. The initial computations show 17 percent reduction in fuel consumption of the refuse truck.

---

Transition control is of the great significance in laminar flows since determination of aerodynamics coefficients as well as heat transfer magnitude is strongly affected by accurate prediction and control of this phenomenon. Transition is severely dependent on space and time such that various microscopic and macroscopic scales can convert to each other rapidly. In one side, available uncertainties in RANS turbulence models can lead to inappropriate, or at least expensive, designs. In the other side, considering the growing rate of computational resources along with development of more efficient numerical methods in CFD applications, Direct Numerical Simulation, DNS, has found an applicable role even in industrial applications. In present study, a robust computational code is developed for Direct Numerical Simulation aimed at fundamental purposes. To this end, high-order compact finite-difference for spatial derivatives and high-order Runge-Kutta time integration are used in the present code as well as a low-pass filter to elucidate spurious oscillations. Also, non-reflecting boundary condition is employed to keep the domain size as small as possible and to improve the numerical accuracy at the boundaries. In present study, Direct Numerical Simulation investigates controlled transition scenarios for flow over a flat plate. Results are in a good agreement with those of previous researches both qualitatively and quantitatively which verify the various parts of the developed solver.

---

Non-Newtonian fluid flows experience turbulent regime in some industrial applications. Several approaches have been proposed for numerical simulation of turbulent flows that each one has specific features. RANS turbulence models have reasonable computational costs, while include several sources of uncertainties affecting simulation results. In addition, developed RANS models for non-Newtonian fluids are modified versions of available models for Newtonian fluids, therefore, they cannot provide reliable estimation for viscoplastic stress term. On the contrary, DNS delivers accurate results but with high computational costs. Consequently, use of DNS data for estimation of uncertainty in RANS models can provide better decision making for engineers based on RANS results. In the present study, a turbulence model based on for power-law non-Newtonian fluid is developed and employed for simulation of flow in a pipe. Then, an efficient method is proposed for quantification of available model-form uncertainty. Moreover, it is assumed that uncertainties originating from various sources are combined together in calculation of Reynolds stress as well as viscoplastic stress. Deviation of the stresses, computed using RANS turbulence model, from DNS data are modeled through Gaussian Random Field. Thereafter, Karhunen-Loeve expansion is employed for uncertainty propagation in simulation process. Finally, the effects of these uncertainties on RANS results are shown in velocity field demonstrating the fact that the presented approach is accurate enough for statistical modeling of model-form uncertainty in RANS turbulence models.

---

In this paper, one-dimensional numerical optimization of lead-acid battery with finite-volume method is performed using the governing equations of battery dynamics. For validation, the present results are compared with previous studies which show good agreement. The demand for batteries with high energy and power has increased due to their use in hybrid vehicles.The major shortcoming of lead-acid batteries in industry is low energy and high weight; therefore, a cell with higher energy and lower thickness is designed by using particle swarm optimization based on developed simulation code which is less time consuming and much faster than experimental method. The results of optimization show that an optimal battery that has 85 percent higher energy can be made with the same cell length. The results also show that an optimum cell battery can be obtained with a decrease of 25 percent in weight and 23 percent in dimensions while keeping the energy content constant.

---

Modeling and simulation are useful tools to optimize and analyze the dynamic behavior of lead-acid batteries. One of the main problems is that the governing equations of lead-acid batteries are highly coupled which significantly increases the computational time of numerical methods in simulations. Using reduced order models (ROM) is one of the best ways to overcome this difficulty. In the present study, the one-dimensional electrochemical governing equations of lead-acid battery are solved using model order reduction based on proper orthogonal decomposition (POD). To show the capability of this method, the governing equations including conservation of charge in solid and liquid phases and conservation of species are solved simultaneously for a lead-acid cell during discharge, rest and charge process. The results of reduced order model including cell voltage, acid concentration and state of charge (SoC) are compared to the results of finite-volume method (FVM). The obtained numerical results show that not only the POD-based ROM of lead-acid battery significantly decreases the computational time (speed-up factor of 15) but also there is an excellent agreement with the results of previous computational fluid dynamic (CFD) models.

---

According to the importance of the engine emission and because of the cost of the laboratory tests, it is necessary to simulate the engine via numerical methods. In this study a numerical simulation of single cylinder SI engine has been carried out to predict the internal combustion engine emission with the AVL BOOST software. The engine calibration has been performed in 2000 rpm engine speed and three loads (part load, mean load and WOT) and three compression ratios (12, 14, 16) with stoichiometric air fuel ratio. After the calibration of engine, the Lambda value is changed in range of 0.8 to 1.25 and the NOx and CO values are calculated. the results show that emission of NOx is dependent extermly to load of combustion engine but compression ratio has not very influence on this emission.unlike the NOx, the both compression ratio and load don't have any effects on CO values.

---

One of the methods for solving Navier-Stokes equations in order to analyse aerodynamic flows is using finite volume method. Since aerodynamic flows are mostly in the range of compressible flows, here one of the density based algorithm (CUSP) have been studied to connecting equations. So here by adding LD (Low diffusion) part to the CUSP method a new method LDE (Low diffusion E-CUSP) have been created which containing new improved discretizations and it has been extended for a unstructured two dimensional mesh. Because of using edge-based data structure it gives the ability to solve the unstructured and structured meshes. Also the discretization of time section is done explicitly by Runge-Kutta method. It has acceptable stability range in compare with the amount of calculation utilized. Then, the results of new improved method (LDE) have been studied for a unstructured 2d mesh and compared with old method which it has been improved for unstructured mesh. The results show that the convergence time and the number of iterations to reach desired error are reduced. Also error percentage of numerical results like pressure coefficient is reduced. Moreover, dissipation of this new method does better than first method in terms of capturing shock location in a proper way.

---

Thermal run-away of lead acid batteries is one of the destruction modes of lead-acid battery. This phenomenon is a thermal-fluid dynamics instability problem that needs to be solved via direct numerical procedures. High-order simulation of lead-acid battery is the first step of direct numerical simulation (DNS) methods for research on thermal run-away phenomenon. In this study, due to simple geometry of lead-acid-cell, spectral methods which are very common in DNS simulation of thermal and fluid dynamics instability problems is implemented on lead-acid cell. A full cycle of discharge, rest and recharge process of a lead-acid cell is simulated by Chebyshev spectral collocation method combined with fourth order Runge-Kutta time integration. Due to complexities, the simulations are performed to find the possible numerical difficulties of this method as the first step. Two coarse and fine grids with Chebyshev polynomials of 8 and 12 order are selected to perform numerical simulations. Comparison of the error shows that the accuracy will be decreased up to 200 times just by adding two points to grid. Also numerical results show that this method is sufficiently able to predict the cell behavior in the high rates of cell current. The results indicate that spectral methods and Runge-Kutta time integrations are promising tool for direct numerical simulation of lead-acid batteries to study complex physical phenomena such as thermal run-away problem.

---

In researches on ducted wind turbines, in order to consider the effects of the duct, the solution process is dependent on parameters which arise from experimental tests or computational fluid dynamics. In the present study, our goal is to present a method for considering the effects of the duct and hub on the wind turbine enclosed in a duct without needing to costly experimental tests or time-consuming numerical simulations. For this purpose, the potential flow method which requires only lift and drag coefficients as input parameters is used. The surface vorticity method and the lifting line theory based on the Biot-Savart law are implemented as a numerical method to analyze the performance of the ducted horizontal axis wind turbine. The proposed method is programmed in the MATLAB software. The validation is carried out with experimental result of the DONQI horizontal axis wind turbine. The results are in good agreement with experimental data in the literature. The output power of the ducted wind turbine is compared to the same bare wind turbine to show the effect of the duct on the performance of the wind turbine. The power curve is illustrated that the ducted wind turbine produces more power than an unducted wind turbine in the same condition.

---

In current research, surface reaction phenomena in several packed bed reactors have been considered. Flow field through several fractal Sierpinski carpet porous media have been simulated by LBM. The endothermic Isopropanol dehydrogenization reaction has been considered as basic reaction mechanism and two major parameters of non-homogeneity and specific area in catalytic surface reaction have been investigated. To validate our numerical method, the obtained results have been compared to a recent benchmark study and adopted very well. Also, in both cases the porosity factor retained constant (ε=0.79). The results shown that, by three times increase in specific area, the reactant conversion rate is increased significantly (approximately one order of magnitude), and the pressure drop increased (nearly 5 times), too. Also, to consider non-homogeneity arrangement, the particle arrangements from small to large and large to small have been considered. In both, the pressure drop is approximately the same. At low Re, reactant conversion of both arrangement are the same but by increasing of Re, the packed bed reactor with large to small arrangement has a little more reactant conversion.

---

Increasing of capacity in lead-acid batteries and reducing charging time in lower temperature are considered as some main challenges of designers and manufacturers. Geometrical properties of battery plates such as thickness and maximum activated area are some of effective parameters on battery performance. Thus, determining of optimum values for independent variables is an important problem for battery industry. In the present study, a numerical solution code is developed using computational fluid dynamic method to simulate battery behavior. Numbers of 50 runs are suggested using response surface method. For each response one empirical model is extracted as a function of independent variables and from these models the optimization process is done. The results shows that in positive electrode thickness of 0.078 cm, negative electrode thickness of 0.53 cm, separator thickness of 0.04 cm and maximum activated areas for positive and negative electrode of 80 cm-1 is an optimum condition to get maximum capacity, minimum charging time and temperature. A confirmation test is done and it demonstrates that the results are in good agreement to predicted optimum results. In conclusion, the present study shows that by changing geometrical properties of the battery one can improve its performance.

---

The main approach in the study of fluid flow instabilities is the theory of linear stability, which is based on linearizing the governing equations and finding unstable eigenvalues. In many flows, like shear flows, the results of linear stability theory fail to match most experiments. In a linear system, even if all the eigenvalues are stable, the perturbations can lead to instability, if the eigenfunctions are not orthogonal. The transient features of these non-normal dynamical systems, can be described with low-dimensional structures, i.e. a few modes. It is possible to suppress the asymptotic and transient growth by identification of time-dependent modes. In this paper, a method of order reduction based on optimally time-dependent modes has been implemented. This method identifies the growth behavior of disturbances in short and long times. Also, a control algorithm based on the above method has been implemented to stabilize the growth of disturbances. The DNS solution of the flow and the implementation of the reduction and control algorithms is based on the NEKTAR++ open-source solver. At first problem, to validate the solution method, the order reduction and control algorithm has been implemented on the flow over a cylinder with Re=50. At second problem, for the first time, the control algorithm is implemented on the flow over a cylinder subjected to persistent time-varying disturbances. The results show that by applying a control force, the Von-Karman vortices are stabilized and a constant lift is obtained and body vibrations are cancelled.