Search

The lower hook mechanism is one of the key mechanisms in netting machine. The weft reel diameter of the new generation netting machine has doubled to the original one, resulting in a large route change of lower hook, thus the lower hook mechanism needs to be redesigned. By comparing the fitting results under different methods, the hook tip trajectory with smaller acceleration is determined. By using the closed vector method, the inverse and forward kinematic models of the mechanism is established and solved. By using NSGA-II genetic algorithm, the transmission performance parameters of the mechanism are effectively optimized. The kinematic simulation of the optimized mechanism is completed in Adams to verify the correctness of kinematic models and the effectiveness of optimization results. The comparison result of the simulated and initial fitted trajectory shows that the lower hook mechanism designed in this paper could meet the technical requirements of the position and posture of the lower hook. [ABSTRACT FROM AUTHOR]

Ultra-high-pressure (UHP) water jetting is commonly used to remove rust from ship hulls. However, few studies have been undertaken to optimize the topology of reaction rotating sprinklers (RRS) using UHP. This work proposes two contributions aimed at optimizing the RRS topology and maximizing the efficiency of rust removal. The first contribution is a novel accumulative impinging-duration model, which can accurately quantify the spiral trajectory density of multiple water jets and provide a practical means to evaluate the feasibility of RRS topology. The second contribution is the development of a new meta-heuristic algorithm called adaptive genetic algorithm (AGA), which offers high-quality solutions for the combinatorial topology optimization problem associated with RRS. The AGA converges significantly faster than the conventional genetic algorithm (CGA) and avoids being trappped at a local optimum. The effectiveness of the proposed topology optimization method is demonstrated through two case studies involving RRS with different model specifications. The AGA significantly increases the utilization ratio of water-jet energy by more than 50.6 %, outperforming CGA. The rust-removal efficiency of AGA-optimized RRS is theoretically predicted by analyzing the maximal translation speed and further validated through field experiments. Results show that the rust-removal efficiency of AGA-optimized RRS for ship hulls is greatly improved, with improvements of over 50 % and 43 % in theoretical and practical verification, respectively. This research introduces a unique quantization method and an efficient optimization algorithm for practical RRS design. [ABSTRACT FROM AUTHOR]

In order to improve the operating performance for cycloid pinwheel reducer in fields of transmission efficiency, compacted structure, load-bearing capacity, and tooth profile modification. A double-stage multi-objective optimized method is proposed that considers for parameters of structure and modification. Firstly, taking the high transmission efficiency and small volume of reducer as objectives, the constraint conditions included structural parameter size and strength limitations were analyzed. Based on genetic algorithm, the mathematical model for structural parameters of cycloid pinwheel reducer was established as the first-stage optimization; secondly, the algorithm of meshing force and the deviation between rotated angle modification and equidistance and moved distance combination modification were derived, taking the low meshing force and deviation as objectives, the theoretical model for modification parameters of equidistant and moved distance amount was proposed as the second-stage optimization. The results indicate that the comprehensive performance for cycloid pinwheel reducer can be improved effectively after optimization. [ABSTRACT FROM AUTHOR]

This study has developed new features necessary for condition monitoring and diagnosis of rotating machinery. These features are developed using the phase change of vibration signal, which is characteristic of blade rubbing fault. These developed features are intended to identify the fault's correct condition and severity of the rotating machinery. The difference between normal and blade rubbing fault was compared through experiments. The experimental model was produced to simulate a blade rubbing fault. The data were acquired through the experimental model and calculated using the developed features. Fault detection was confirmed by using genetic algorithm and machine learning that failure detection was possible using the developed features, it is expected that such study can evaluate the health of the rotating machinery. [ABSTRACT FROM AUTHOR]

Accuracy and efficiency are two important performances of model predictive control in trajectory tracking and they are seriously affected by the control parameters of model predictive control. To make the model predictive control with high accuracy and efficiency simultaneous, the paper proposed a strategy to tune the control parameters for model predictive control. The proposed strategy converts the tuning problem to a multi-objective optimization problem and employs non-dominated sorting genetic algorithm (NSGA-II) to solve it. The proposed strategy is employed to tune the control parameters for a classical model predictive control in a typical trajectory tracking condition. The simulation results show that the comprehensive performances of model predictive controller tuned by the proposed method are better than other tuning methods. The proposed tuning strategy is validated and it can be applied to tune the control parameters for model predictive control in trajectory tracking. [ABSTRACT FROM AUTHOR]

Instrumented indentation is a versatile method of extracting hyper-elastic material parameters, particularly useful for applications where stress-strain data are difficult to be in-situ measured. Because the analytical force-displacement relation is still unavailable for the indentation of hyper-elastic materials, identifying hyper-elastic parameters often requires an iterative optimization strategy that fits finite element simulations with experimental data. However, the optimization strategy is burdened by heavy computation and its prediction accuracy is greatly influenced by the choice of optimization algorithm. To address these challenges in this study, a bidirectional long short-term memory (BLSTM) neural network is presented that directly predicts hyper-elastic material parameters from indentation load-displacement data, focusing on Mooney-Rivlin hyper-elasticity as an example. To improve the predication accuracy, the condition numbers for the inverse identification of the hyper-elastic parameters are investigated. And, a normalization procedure is proposed to treat the input data, which can guarantee the BLSTM network is well-conditioned. During evaluation, the trained BLSTM network significantly outperforms the iterative optimization strategy using a genetic algorithm. Furthermore, the effect of the normalization procedure is demonstrated. [ABSTRACT FROM AUTHOR]

In the production process of the giant tire, the tire unloader needs to clamp and flip the tire, and the tire has a large inertia force, easy to cause vibration and lead to the failure of bearing parts, etc., then the flange is needed to play a fixed role. therefore, the rationality of flange design for the use of safety and service life are very important. For this reason, this paper simulates the flange connected to the tire unloader to obtain its stress history data, at the same time through genetic algorithm structure optimization analysis was carried out on the flange, it is concluded that the influence of structure parameters on the life of flange, and flange for structure optimization, get the optimized flange fatigue life 71 % increase compared to the original plan. The weight is reduced by 0.9 %, and the mechanical properties of the optimized flange are better. [ABSTRACT FROM AUTHOR]

The positive characteristics observed in terms of compact structural sizes and high torque-transmitting capacities have allowed co-rotating twin-screw (CRTS) gear transmissions to be extensively applied to instances requiring small central distances and large power densities. In this study, a multi-objective optimization model for the quick determination of structural parameters of the CRTS gear transmission system is proposed based on the heuristic search and non-dominant sorting genetic algorithm. Results reveal that the optimized scheme with 17 design variables and 18 constrained parameters can be achieved with this model. The system reduces mass by 7.74 % from the original mass of 199.06 kg, while the safety factor variance of the transmission system decreases by 29.3 %. This method applies to other transmission systems with different structures and thus provides general theoretical support for modern gear transmission optimizations. [ABSTRACT FROM AUTHOR]

A double-layer optimization method for the forging process parameters of a hinge beam structure and the design of intermediate billet structure is proposed. First, a primary optimization system for the forging process parameters of hinge beam was established based on orthogonal experiments and range analysis method, and the optimal combination of forging process parameters was selected. On this basis, a mathematical model of the secondary optimization of the final forging of the hinge beam was established. Through radial basis function (RBF) agent model and multi-island genetic algorithm, the secondary optimization of intermediate billet structure of hinge beam was carried out, which improved the grain refinement degree of final forging and reduced the forming load required for forging. Through numerical simulation and optimization design of forging process, the quality and performance of final forging of hinge beam were improved, which has certain guiding significance for the actual production of hinge beam. [ABSTRACT FROM AUTHOR]

Research on the vibration characteristics of the permanent magnet synchronous motor (PMSM) is hindered by the harmonic characteristics of the unbalanced magnetic pull force when air-gap eccentricity is neglected. Thus, the effects of both air-gap eccentricity and mass eccentricity on the rotor vibration have not yet been fully elucidated. To address this, a unified expression for the unbalanced magnetic pull force of the motor was derived under various pole pairs and air-gap eccentricities. The distribution of the radial electromagnetic force under different eccentricities is revealed. An electromagnetic simulation model of the fan motor was established for three eccentric cases and the effect of the unbalanced magnetic pull distribution and harmonic frequency characteristics on the rotor were analyzed. Furthermore, the influence of the gyroscopic effects on the modal frequency of the rotor structure was studied. The harmonic response of the rotor shaft system was first established using multi-physical field coupling, then the vibration amplitude-frequency laws of the rotor-shaft system under the influence of air-gap eccentricity and mass eccentricity were investigated. To reduce the vibration of the unbalanced PMSM rotor-shaft system, a multi-objective optimization was performed using the genetic algorithm in ANSYS Workbench. Research on the vibration characteristics of the permanent magnet synchronous motor (PMSM) is hindered by the harmonic characteristics of the unbalanced magnetic pull force when air-gap eccentricity is neglected. Thus, the effects of both 3-gap eccentricity and mass eccentricity on the rotor vibration have not yet been fully elucidated. To address this, a unified expression for the unbalanced magnetic pull force of the motor was derived under various pole pairs and air-gap eccentricities. The distribution of the radial electromagnetic force under different eccentricities is revealed. An electromagnetic simulation model of the fan motor was established for three eccentric cases and the effect of the unbalanced magnetic pull distribution and harmonic frequency characteristics on the rotor were analyzed. Furthermore, the influence of the gyroscopic effects on the modal frequency of the rotor structure was studied. The harmonic response of the rotor shaft system was first established using multi-physical field coupling, then the vibration amplitude-frequency laws of the rotor-shaft system under the influence of air-gap eccentricity and mass eccentricity were investigated. To reduce the vibration of the unbalanced PMSM rotor-shaft system, a multi-objective optimization was performed using the genetic algorithm in ANSYS Workbench. The optimization results show that when the motor rotor core is shifted 2 mm in the positive X direction and the distance between the double-row bearing and rotor core is 7 mm, the vibration acceleration amplitude is reduced by 54.9 % compared with the original model. [ABSTRACT FROM AUTHOR]

Most of the kinematic calibration methods of parallel mechanisms only consider the geometric error of the mechanism, and the calibration effect is poor. To improve the calibration effect, this paper takes the 3-PTT (3-prismatic hook hook) parallel mechanism as the research object and proposes a kinematics calibration method based on the normalized representation model of the geometric error and non-geometric error of the mechanism. First, the degree of freedom of the mechanism is analyzed, its kinematics positive and negative solutions are solved, and its singularity is analyzed. Secondly, a normalized characterization method of mechanism geometric errors and non-geometric errors is proposed, and an error model without redundant parameters is constructed. The end motion error of the machine is measured by laser tracker, and the objective function is constructed. The genetic algorithm is designed to solve the minimum value of the objective function, and the normalized error of the mechanism is identified. By comparing with the recent methods, the better identification performance of the algorithm in this paper is verified. Finally, the kinematics model was corrected according to the identification results. After the calibration was completed, the movement errors of the end of the mechanism along each coordinate axis were reduced by more than 99 %. Compared with other calibration methods, the better calibration performance of the method in this paper is verified. [ABSTRACT FROM AUTHOR]

In order to solve the problem of lack of flexibility in direct teaching of industrial robots under complex working conditions, dynamic inertia term compensation is combined with the traditional force-free control algorithm to reduce the traction in the teaching process in this paper. Firstly, considering the multi-dimensional, strongly nonlinear and multi-parameter characteristics of the 6-DOF manipulator dynamic model, a step-by-step parameters identification method based on genetic algorithm is proposed. This method can effectively reduce the computational complexity and improve the optimization speed of industrial robot dynamic model identification. Secondly, based on the internal torque analysis of industrial robot joints, a force-free compensation control algorithm based on the torque control is designed. Finally, the effectiveness of the step-by-step robot dynamic model parameter identification method is verified by a traction teaching experiment. In the verification experiment, the relative error rate between the predicted torque and the actual torque calculated from the identification results is 1.64 % to 5.99 %. After increasing the inertia term compensation, the traction force of the teaching process is reduced by 28.6 %. The experimental results show that the identification result is more accurate, and the proposed force-free control algorithm significantly improves the compliance of the teaching process. [ABSTRACT FROM AUTHOR]

To reduce energy consumption and improve service stability and security, various design methods of cutterhead based on lightweight, modal optimization and life prediction were adopted. In the present work, a multi-disciplinary design optimization (MDO) methodology is presented and subsequently applied to the development of cutterhead. According to the requirements of the cutterhead process and the results of sensitivity analysis, the design parameters of each discipline were determined. To alleviate the computational cost caused by calling the results of finite element method (FEM) in design optimization, the Kriging model was employed to approximate the highly nonlinear implicit relationship between the mode frequency, fatigue life and the design parameters. Then, the system-level optimization model and the subsystem-level optimization model of mass, mode frequency and fatigue life were established based on collaborative optimization (CO). Finally, genetic algorithm (GA) was used for iterative optimization between the system-level and subsystem-level model to obtain the overall optimization results. The results showed that the mass of the cutterhead was finally reduced from 90.4 t to 88.5 t, the first-order modal frequency raised from 32 HZ to 37 HZ, and the fatigue life raised from 3.0E5 cycles to 5.5E5 cycles. [ABSTRACT FROM AUTHOR]

Mathematical validation and comparison of coal milling operations of a horizontal ball tube mill and a vertical bowl mill located in two distinct coal based power plants in India was carried out. The coal mill model was developed and the unknown model coefficients were obtained numerically by genetic algorithm (GA). The input variables of mass flow rates of raw coal and air and temperature of inlet primary air were obtained from the on-site data. The output variables of mill current and mill outlet temperature as obtained by GA were compared with on-site data for the two mills and show good agreement. The ball tube mill requires less mass flow rate of primary air and raw coal per MW of power produced than the bowl mill. It contains larger mass of raw and pulverized coal but less grinding heat is generated in this mill. [ABSTRACT FROM AUTHOR]

An asymmetric modification method for the roller profile was proposed to address the bias load effect of rollers in the main drive bearing of a large roadheader. A contact mechanics model of the bearing was solved, and line contact theory was used to explore the effect of combined load and convexity on the bias load effect. The contact stress was found to be asymmetrically distributed along the contact line. This trend tended to be remarkable with the decrease in axial load or the increase in overturning moment. Symmetrically modified rollers can be used to improve the bias load effect by increasing convexity. However, this process led to the contact stress concentrating in the roller center. The influence of the design parameters for asymmetric modification on the roller profile and the contact stress was considered. The ideal design parameters were determined by using a genetic algorithm. The contact condition and fatigue life were remarkably increased. [ABSTRACT FROM AUTHOR]

This paper presents a double-mass dynamic vibration absorber (DVA) that rapidly reduces the motion of the object to be damped (main system) and converts it to the motion of the DVA. In this way, the DVA attenuates the vibration of the main system. A DVA is usually designed to improve the resonance amplitude and convergence time of the main system vibration, and the kinetic energy of the main system is dissipated as heat in the damping element of the DVA. We describe a novel configuration consisting of two DVAs connected by a spring and an inerter. This new structure enables a rapid response reduction of the main system and residual vibration in the DVA. By tuning the responses of both the main system and the DVA, two vibration modes can be realized: a damping mode and a residual vibration mode. The proposed device is optimized using a genetic algorithm and applied to a bridge to demonstrate its performance. [ABSTRACT FROM AUTHOR]

In this study, the results of genetic algorithm and sequential quadratic programming are discussed for NACA four-digit airfoil series optimization. The lift-to-drag ratio is defined as the objective function. The aerodynamic constraints related to the lift, drag, moment values of the airfoil at high and low angle of attack, and the geometric constraints related to the wetted length and internal volume are set. Results show that the sequential use of the former and latter methods and the stand-alone use of the latter method yield an increment of approximately 20 % in the objective function. The change in the geometric design variables reveals that the optimum airfoils are obtained by the variation in the maximum camber and thickness. Sequential quadratic programming is proven to converge to the global optimum in a significantly shorter solution time than genetic algorithm when the practical constraints are defined into the optimization problems. [ABSTRACT FROM AUTHOR]

As a core component of a fan, the blade has a decisive impact on the aerodynamic performance of a low-pressure axial flow fan. Due to the limitations of the classical fan design theory on considering the complex 3D internal flow, the spanwise distribution of blade stacking line and section profiles are usually hard to reach the best state. This paper builds a surrogate-assistant multi-objective optimization flow combined with CFD method to explore the optimum blade shape under two typical working conditions. A total of 16 parameters were selected for demonstrating the blade stacking line and section profiles, according to Morris one-at-a-time sensitivity analysis. The objective and constraint functions were the fan's total-to-static efficiency and static pressure rise, respectively. During the optimization, the surrogate models of all response functions were built using kriging models, on which the multi-objective genetic algorithm takes the exploration. The optimization results indicate that the maximum improvement of the efficiency is 1.26 % for low mass flow working condition and 5.47 % for high mass flow working condition. The optimized models tend to make the low-pressure zone distributed along the blade leading edge in the meridian view and to reduce the tip leakage vortex intensity. This paper provides a good practical demonstration of multi-objective fine optimization on the blade shape of a low-pressure axial flow fan. [ABSTRACT FROM AUTHOR]

Aiming at the problem that the output displacement of the piezoelectric actuator is small and the coupling motion of the positioning platform affects the positioning accuracy, we designed a 2-DOF decoupling large-stroke precision positioning platform. First, a dynamic model based on the lumped mass method was established and verified by simulation. The multi-objective genetic optimization algorithm was used to optimize the structural size parameters of the positioning platform, and the optimal solution set of the structural size parameters of the positioning platform was obtained. Finally, according to the theoretical and simulation results, a prototype was fabricated and the working stroke, decoupling performance and natural frequency were tested experimentally. The results show the coupling rates between axes of the positioning platform in the x-direction and y-direction are 1.31 % and 1.62 %, respectively, the natural frequency is 337.2 Hz, and the positioning stroke is 89.2 µm×85.9 µm. The positioning platform designed in this paper is decoupled and has a large output stroke. [ABSTRACT FROM AUTHOR]

Nonlinear energy sinks (NESs) have received increasing attention for their ability to passively inhibit a large amount of vibration energy over a wide range of frequencies. However, although many studies have focused on dynamics of NES, few have addressed the optimal design method of NES, especially for the non-smooth NES (NSNES). Therefore, the parameter optimization method for an NSNES which can be applied to supress torsional vibration of rotor system was developed. First, the design variables were reduced by using piece-wise linear torsional stiffness to equivalently fit cubic torsional stiffness; then, taking the torsional vibration of single-disk rotor system as an example, the genetic algorithm (GA) was used to solve the optimization problem of the NSNES in torsional vibration suppression of rotor system. Finally, the optimized NSNES was verified by numerical simulation and experiment to suppress the torsional vibration of a rotor system. The results show that the optimally designed NSNES can effectively suppress torsional vibration of a rotor system. In transient vibration suppression, the optimal percentage of accumulated energy dissipation of NSNES can reach 95.3 %. For steady-state vibration suppression, the peak vibration suppression of NSNES can reach 82.2 % in the simulation and 81.9 % in the test. [ABSTRACT FROM AUTHOR]

In the present study, the aerodynamic performance and flight stability of a two-dimensional (2D) canopy in a paraglider are optimized using a combination of response surface methodology (RSM) and a multi-objective genetic algorithm (MOGA) coupled with the unsteady Reynolds-averaged Navier-Stokes (URANS) equations solver. Compared to a 2D base case, an optimized canopy, featured by reduced airfoil thickness, shows an increase in the aerodynamic performance up to 18.9 % based on lift-to-drag ratio, while the flight stability is similar between them. An optimized three-dimensional (3D) canopy is constructed by duplicating the 2D canopy along the arc direction to identify the effects of the optimization on an actual 3D canopy. Based on large-eddy simulation (LES) data of the optimized 3D canopy and base 3D canopy, we show an improvement of the aerodynamic performance and stability of the optimized 3D canopy, consistent with our results from the 2D canopies. [ABSTRACT FROM AUTHOR]

This paper presents an optimization method based on the Kriging and multi-objective genetic algorithm. First, taking the hinged beam structure of the cubic diamond press as the design object, the optimization design mathematical model was established with the mass as the objective function, the stress and displacement peak as the constraint conditions. Second, in combination with SolidWorks and ANSYS Workbench, parametric modeling analysis was conducted to obtain a large number of sample points sparing less time, and the agent model constructed by Kriging was trained and verified. Finally, taking advantage of global search of the multi-objective genetic algorithm, a lightweight design was realized and the mass of the hinge beam structure was effectively reduced, which would be a guiding significance for the lightweight design of other mechanical parts. [ABSTRACT FROM AUTHOR]

In order to get the optimal profile of cycloid gear after sectional modification, a multi-objective optimization design method is proposed that considers both the modification parameters and macro-parameters of the cycloid gear. An algorithm of meshing force, meshing efficiency and anti-gluing ability between the cycloid gear and the pin gear was derived, and the related independent parameters were extracted as optimization variables. Taking high efficiency, high strength and light weight as the objective, the mathematical models of double-objective and three-objective optimization were established, and the influence of key design variable on the objective function was analyzed, and NSGA-II multi-objective genetic algorithm was used to solve the Pareto optimal solution of the optimization mathematical model. Results show that the optimized parameters can significantly improve the meshing efficiency, reduce the volume and meet the design requirements of high strength, high efficiency and lightweight on the premise of ensuring the strength of cycloid gear surface. [ABSTRACT FROM AUTHOR]

Traditional method evaluates operating reliability by counting life-test results of a small number of samples, which ignores variability between individuals and cannot accurately reflect the true state. This paper proposes a model for predicting operational reliability of mechanism in a batch of micro-switches based on manufacturing parameters, in view of complex structure and diverse manufacturing parameters. A multi-body kinetic model based on manufacturing parameters and principle of action considering contact fatigue and wear was established, and the operating characteristics in the degradation process with sliding wear of joints were obtained. Finally, the linkage with revolute clearance in miniature circuit breaker (MCB) was used as numerical example application to perform investigation. Prediction results are consistent with batch experiments, which verifies accuracy of model. In addition, the optimal combination of clearances was obtained by using genetic algorithm based on predicted model, which increases locking reliability to 99.3% after 10000 operation cycles. [ABSTRACT FROM AUTHOR]

To accurately predict fatigue life and reliability of the forage crusher rotor, the stress load spectrum of the rotor is obtained via the two-way fluid-structure coupling method, which calculates the coupled flow field in the forage crusher and rotor structure. The fatigue life of the rotor is predicted, and its reliability is analyzed. Finally, the rotor is optimized by a multiisland genetic algorithm. The results indicate that the fatigue life prediction based on fracture mechanics theory and the Gerber average stress correction method is more suitable for the prediction of the rotor's fatigue life. After optimization, the fatigue life cycles of the rotor are increased from 1.229×108 to 7.67×108 and the reliability is increased from 89.4 % to 99.7 %. This meets the reliability requirement of key parts of agricultural and animal husbandry machinery. This study provides a reference for the prediction and optimization of fatigue life and reliability of forage crushers. [ABSTRACT FROM AUTHOR]

Compared with ordinary injection-molded parts, the slender, cantilevered, and thin-walled plastic parts are harsh on the injection molding process conditions. For complexity and particularity, it is difficult to form such parts. It is also more likely to cause excessive warpage deformation, affecting the molding quality and performance. The automobile audio shell is a typical slender, cantilevered, thin-walled plastic part. When the mold structure and material are determined, optimizing its injection molding process is the most economical and effective method to manufacture the products with the optimum properties. In order to minimize the warpage deformation, the adaptive network based fuzzy inference system (ANFIS) and genetic algorithm (GA) were adopted to optimize the injection molding process parameters. In particular, considering the high-dimensional nonlinear relationship between the process parameters and the warpage, the ANFIS is constructed as the prediction model of the warpage. Then, the GA is used to globally optimize the prediction model to determine the optimal process parameters. The results show that the optimization method based on ANFIS-GA has a good performance. The warpage is reduced to 0.0925 mm while reduced by 88.25 %. The optimal injection molding process parameters are used for simulation and manufacture, verifying the effectiveness and reliability of the optimization method. [ABSTRACT FROM AUTHOR]

This paper presents the theoretical derivation and the implementation of a foot-ground contact dynamics system for legged robot physical simulation. The physical simulation imposes high requirements on the contact dynamics for fidelity and accuracy. Thus, we explored the contact mechanism and developed a contact model. We theoretically analyzed the ground reaction force (GRF) by the kinetic equation with constraints first. Considering the limitation of the numerical analysis method, we propose a new contact model based on Hertz theory and clarify the fundamental differences between the model and reality. Due to these differences, we optimized some key parameters by the genetic algorithm. Then, we expanded the contact model to nonhorizontal and nonstationary surface contact. At last, we simulated several cases to present the usefulness of this contact model and analyzed the dynamic contact behavior. The contact force based on this model highly accords with the results of theoretical analysis and includes all phenomena that may occur during a locomotion cycle. [ABSTRACT FROM AUTHOR]

Machine learning for the diagnosis of rotary machines takes priority in generating a training data set through the machine's past data. The training data set uses features that have physical and statistical meaning of vibration signals. A training data is formed on the assumption that the normal condition of the facility is almost similar over time. However, many industrial power plants perform regular O/H (overhaul), and the vibration level of the machine's normal condition is likely to change depending on the O/H results. The vibration level is one of the important factors representing the condition change of rotating machines and is difficult to ignore easily. This paper is a study on a method that can be used for feature-based machine learning with training data formed from past data whose vibration level of the rotating machine has changed due to the influence of maintenance. Data acquisition was made through lab-scale defect simulation test devices, and experimental equipment was simulated before and after O/H with several faults that could occur in rotating machines. The signal named "delta signal" refers to a signal that sets each normal data as a reference signal, matches the fault signal through phase synchronization and resampling, and subtracts it to leave only a difference. The algorithms used in machine learning used genetic algorithm (GA) based feature selection and support vector machines (SVM) for learning and classification. According to the experiment, it was confirmed that in raw signal learning, the similarity by the learned condition (label) decreased due to the influence of maintenance, but the method using delta signal decreased the effect by maintenance, increasing the similarity within the same learned condition. [ABSTRACT FROM AUTHOR]

The main objective of the present article is to solve the problems of poor molding quality, large warpage, inadequate cooling effect and unsuitable selection of process parameters, in the injection molding process for passenger vehicle front-end plastic wing plate. The thickness and parting surface of the vehicle front-end fender were determined, the injection mold and its cooling system were designed. The relevant process parameters, affecting the product molding quality, were tested, according to orthogonal experimental approach, while their influence on the warpage was obtained, by analyzing the data. Finally, the BP neural network of warpage model was established and globally optimized using genetic algorithm. The optimal parameter combination of the injection molding process was derived as: melt temperature 236 °C, mold temperature 51 °C, cooling time 32 s, packing pressure 97 MPa and packing time 16 s. [ABSTRACT FROM AUTHOR]

The shape of the blade stacking line greatly influences the low-pressure axial-flow fan's operational efficiency, but there is still no fast and effective way to determine the stacking line's optimal shape. This paper presents an automatic optimization design procedure for the blade stacking line in the low-pressure axial-flow fan based on the FINE/Design3D™ platform. The procedure combines computational fluid dynamics (CFD), surrogate model method, and genetic algorithm (GA) to execute a secondary optimization on a composite skewed-swept rotor-only axial fan blade. The results show that the static efficiency and the static pressure rise of the optimized fan respectively increase by 3.76 % and 5.82 % without stall margin decrease. The blade shape variation in skew and sweep direction reduces the tip leakage flow loss, improves the blade loading distribution, and contributes to the efficiency increment. This research provides a useful reference for the blade stacking line's automatic optimization for the axial-flow fan. [ABSTRACT FROM AUTHOR]

This article introduces a rigid-flexible coupling calculation model based on a multi-platform integrated method to analyze and optimize the cab comfort of commercial vehicles. A 12 degree-of-freedom (DOF) vibration model is first derived by Lagrange's equations. Further, a refined cab with flexible body, bushing, and nonlinear damping is constructed and connected with the vibration model to form a novel dynamics model. Next, the coupled suspend parameters on cab vibration are explored with a Latin hypercube method, which concluded that the front suspend is the most significant component for cab comfort. The multi-objective genetic algorithm (MOGA) and multiverse optimizer (MVO) is then applied to obtain the optimal suspend parameters for ideal cab comfort. Finally, the optimization is verified via simulation and experiment, illustrating that the coupling calculation model and multi-platform integrated method can provide an effective approach to excavate the optimal solution for the optimization of high-complexity engineering problems. [ABSTRACT FROM AUTHOR]

Optimization with a surrogate model is a popular method used in design to avoid a high computational cost problem and is often encountered in computational fluid dynamics and experimental design. However, the optimization with a surrogate model is assigned to solve single-fidelity optimization problems. Thus, a multi-fidelity surrogate model is proposed to solve the multi-fidelity airfoil design problem by combining the data of the wind tunnel experiment and the aerodynamic evaluation using JavaFoil software. In the optimization method, an RBF/Kriging hybrid multi-fidelity surrogate model is used. A non-dominated sorting genetic algorithm II (NSGA-II) is selected to solve the airfoil design problem. The objective of the optimization problem is to minimize the aerodynamic drag and maximize the lift force. The results showed that the algorithm successfully found the UAV airfoil shape and the selected optimum airfoil shape had the error of the aerodynamic lift and drag less than 10 % when tested in the wind tunnel experiment. [ABSTRACT FROM AUTHOR]

A continuous-curvature parameterization method was coupled with the genetic algorithm and a RANS flow solver to optimize the cascade of VKI and Aachen turbine blades. The main advantage of the method is to generate blades with a continuous curvature distribution, which results in a smooth distribution of pressure and velocity on the blade surface. The geometry of the blade cascade was parameterized by 33 variables, and two objective functions were considered for the optimization. The first cost function was to reduce the total pressure loss with the constraints of mass flow rate, blade loading, and outlet flow angle. At the second cost function, the constraint of constant cross-sectional area was added to the previous constraints to preserve the structural strength of the turbine blade. The total pressure loss for the VKI blade decreased by 14.7 % and 10.6 % for the first and second objective functions, respectively. The total pressure loss for the Aachen blade was also reduced by 9.5 % and 7.5 % for the first and second objective functions, respectively. Due to the efficient geometry parameterization, the proposed method quickly converged to a high-efficiency blade at the early generations. The proposed method can be developed for optimizing the different blades of turbine, compressor, and airfoil types. [ABSTRACT FROM AUTHOR]

The effects of axial profile parameters on the main bearing performance of the engine were investigated through the numerical method based on elasto-hydrodynamic lubrication theory, average flow, and asperity contact model. Results show that quadratic profile significantly improves the bearing performance, and the influence of profile varies with its width-to-height ratio. The performance is most improved when the ratio is between 0.8 and 2. An artificial neural network fitting model was developed to predict bearing performance, and multiobjective optimum analyses were performed using genetic algorithm and particle swarm optimization. The optimization goals are average peak asperity contact pressure and average total friction loss. The obtained Pareto front roughly includes three groups, and solutions in group 1 achieve a balance of the two goals, with a width-to-height ratio of 1.5–2. Finally, bearing friction tests were conducted on four profiled bearings to verify the simulation model and optimization results. [ABSTRACT FROM AUTHOR]

In this work, a method for obtaining a set of optimized maximum wheel steering angles for a multiaxle vehicle with multiaxle steering is proposed. During low-speed operations, multiaxle vehicles must have a specified minimum turning radius to traverse crowded areas. The conventional geometrically correct steering configuration may not result in the desired turning radius due to constraints on the maximum steering angles. In such cases, large tire lateral slip angles may be introduced to decrease the turning radius. However, high tire lateral slip angles cause high tire wear. The proposed optimization process aims to achieve the minimum possible turning radius together with minimum tire lateral slip angles. Genetic algorithm is used in the optimization. A case study is conducted with an 8×8 all-wheel steering off-road vehicle. Results show that significant reductions in turning radius and lateral tire slip angles can be achieved. [ABSTRACT FROM AUTHOR]

The effect of design variables on the performance measures in the entire design space can be evaluated by global sensitivity analysis (GSA). Therefore, a high-dimensional engineering problem can be transformed to a reduced-dimensional problem by ascertaining important design variables according to GSA. Subsequently, the reduction of design variables helps save computational cost and the duration of design optimization. In this study, GSA was applied to determine the most influential design variable in the design of two reaction hydro turbine components - stay vane and casing. The global sensitivity index technique is selected among GSA methods because it can evaluate both the individual and interaction effects of design variables. Genetic aggregation surrogate models are used to reduce the number of computational fluid dynamics (CFD) analyses for GSA and design optimization. The responses from surrogate models are evaluated to obtain global sensitivity indices for performance measures. The important design variables are selected according to the indices. Using a multiobjective genetic algorithm (MOGA), design optimizations with the selected design variables are performed and the results are compared to the optimum designs with all design variables. The optimum designs using the selected design variables show comparable performances to those of the full-dimensional optimum designs. [ABSTRACT FROM AUTHOR]

Reliability assessment is the key to the reliability design of a complex system. This paper takes the lithium battery pole slice separators (LBPSS) as the research object and selects the superposition Weibull model as the initial model. Based on the graphic method, a pivot quantity is introduced to estimate the range of scale parameters. A genetic algorithm is used to optimize parameters in a given parameter interval. Compared with other models and techniques, using the superposition Weibull model, pivot quantity, and graphic method to evaluate the reliability of LBPSS is more reasonable, provides a reference for improving the reliability and maintainability of LBPSS, and lays a theoretical foundation for analyzing the failure mode and failure mechanism of a complex system. [ABSTRACT FROM AUTHOR]

For the firing accuracy problem of the moving tank with different speed and different road level, the modelling and optimization method of the projectile disturbances of the moving tank under the general driving conditions were studied. A high-dimensional multi-objective optimization method for the projectile disturbances of the moving tank considering the different driving conditions was proposed based on the sequential approximate optimization method. Therein, a projectile disturbance multi-disciplinary simulation model of the moving tank considering multiple nonlinear factors was built and the sliced Latin hypercube design method was introduced to apply the experimental design. On the basis, the BP neural network surrogate model of the projectile disturbances at muzzle-exit of the moving tank under the general driving conditions was established. With the genetic algorithm and the potential optimal criterion, the sample points were updated to improve the accuracy of the surrogate model during the optimization. The optimization results show that the projectile disturbances at muzzle-exit of the moving tank under the general driving conditions were reduced effectively, and the comprehensive firing accuracy of the moving tank was improved. The research can provide a reference for realizing the reasonable match between firing accuracy and mobility of the tank. [ABSTRACT FROM AUTHOR]

There are some advantages of parallel robots over serial ones, such as strong structure, big payload ratio with respect to the weight, and good dynamic response. The disadvantages of the parallel robots are the difficulty of solving the forward kinematics and the limited workspace. We apply the advantages of both neural networks and genetic algorithm, for the parallel robot design. With these two methods, we solve the problems of the maximum workspace and the forward kinematics. The design method we proposed in this paper uses the optimum algorithms such that the workspace volume of the robot is maximum. We successfully applied this method to the Stewart platform. [ABSTRACT FROM AUTHOR]

The oscillating flow is a problem occurring in the stay vane passage of hydro turbines. The improper shape of the stay vane causes wake formation and sheds large eddies in the trailing edge of the stay vane. The wake at the trailing edge of the stay vane produces pressure fluctuation in the stay vane passage, which leads to noise and vibration during the turbine operation. Therefore, the improper shape of the stay vane may cause flow oscillation in the stay vane passage. Thus, a proper shape design of the stay vane considering the oscillating flow is necessary to mitigate the flow instability. Consequently, experiment and CFD analyses showed that the initial stay vane (ISV) shape causes recirculation and pressure fluctuation. An optimum design methodology is adopted to improve the flow behavior around the stay vane. Optimization of the stay vane is also conducted by using two objective functions (turbine efficiency and flow uniformity) and 12 design variables. The optimal stay vane (OSV) shape is attained by a multi-objective genetic algorithm. Finally, the flow behavior in the stay vane passage with ISV and OSV is compared by experiment and CFD analyses. Thus, the flow instabilities are mitigated with OSV. The installation of OSV improved the flow angle distribution, secondary flow, and pressure fluctuation in the Francis hydro turbine. [ABSTRACT FROM AUTHOR]

To minimize the simulated performance error of gas turbines, traditional adaptive methods are mainly concerned with the tuning of cycle design point and component maps given by the manufacturer, usually ignoring the fact that performance at cycle design point may not match the field data due to the deviation between test-rig conditions and field conditions. In this paper, a new tuning scheme of the cycle reference point is proposed to minimize the simulated errors simultaneously at design point and off-design points. The scheme is composed of a backward iteration algorithm and a genetic algorithm. During the backward iteration, the field data at the maximum operating condition is selected to obtain the initial cycle reference point with several undetermined parameters. Further, the genetic algorithm is used to optimize the undetermined parameters. The accuracy of the proposed method was validated by the simulated performance of a PGT25+ gas turbine. [ABSTRACT FROM AUTHOR]

The identification of aero-engine dynamic parameters is fundamental to establishing accurate dynamic models, which has a great effect on the accuracy of model calculation. The accurate parameter range, however, is not easy to define in practical engineering applications. In many cases, it could only be constructed out of experience. In order to reduce the impact of initial parameter interval accuracy on the identification results, the adaptive parameter domain with multiple genetic algorithms is proposed to identify the aeroengine dynamic model parameters, through which the vibration modal parameter identification is studied under the condition of initial uncertainty of the parameters. The effectiveness of the adaptive parameter domain method is verified through the third-order model with severe modal coupling, which also proves the efficiency and rapidity of finding the correct value of parameters and, indicates that it would not be necessary to introduce an exact definition of the initial interval of parameters. All these suggest that the proposed multiple genetic algorithms of the adaptive parameter domain has a good reference value for engineering applications. [ABSTRACT FROM AUTHOR]

The electro-hydraulic robot has great driving torque and plays an irreplaceable role in manufacturing. In this paper, a trajectory planning method for a 6-DOF electro-hydraulic robot based on time-energy-jerk is proposed and realized. A multi-objective function of the electro-hydraulic robot about working efficiency, impact, stability and energy consumption factors was built, at the same time, the kinematic and dynamic constraint functions were also built. The multi-objective trajectory function of the robot was optimized by using the non-dominated neighborhood immune genetic algorithm, and the optimal position, velocity, acceleration and jerk planning curves of each joint were obtained. In the experiment, the flow output curves of the hydraulic system of joint 6 based on fuzzy-PID control strategy under no-load, 1.5 kg load, 4 kg load were obtained, and show that the velocity and position trajectory of the joint can be well controlled using the presented fuzzy PID control strategy. [ABSTRACT FROM AUTHOR]

This paper studies the influence of hinge point positioning of a steering cylinder on the pressure fluctuation of the steering system of a wheel loader. Affected by the hinge point position of the steering cylinder, the main causes of the pressure fluctuation are found to be the stroke difference and the force arm difference. Based on theoretical analysis, the hinge point positioning of the steering cylinder was optimized by using genetic algorithm (GA). The AME-sim simulation model of the steering system was built to verify the accuracy of the GA. The wheel loader was reformed according to the optimized articulated points position. Steering tests showed a significant drop in system pressure fluctuation, the energy consumed is also reduced. Further, the analysis shows that the force arm difference is the main factor causing the pressure fluctuation, providing the theoretical and experimental basis for the design of a steering system of wheel loaders. [ABSTRACT FROM AUTHOR]

R22 and R407C mass flow rates through straight and coiled adiabatic capillary tubes are analyzed with three ANN models, i.e., the feed forward network with back propagation (BP) algorithm, GA-BP (genetic algorithm optimized BP network) and PSO-BP (BP algorithm combined with particle swarm optimization). The modelled outputs by these ANN methods are compared with experimental data. The results showed that the predicted mass flow rates with the three models of BP, GA-BP and PSO-BP agree quite well with the experimental data with the mean relative error of 3.82 %, 3.14 % and 2.3 % for R22, and 3.17 %, 2.66 % and 2.46 % for R407C, respectively. PSO-BP network is then employed to predict the coiling effect of capillary tubes on the mass flow rate. It is shown that the mass flow rates with coiled diameter between 0.04 m and 0.6 m are about 4 %–13 % lower than that of the straight capillary tube. [ABSTRACT FROM AUTHOR]

The traditional fatigue design of accessories attached to a train body does not consider welding residual stress for fatigue life prediction of welded structures during the design stage, thereby resulting in insufficient fatigue strength for local welded parts. The power spectral density of random excitation caused by rail irregularity from American six-level spectrums was obtained in this paper. Through welding thermal elastoplastic theory combined with ellipsoid heat source model, a numerical simulation for welding residual stress production and distribution of a train buffer beam was conducted. In accordance with the coupling stress component, the fatigue performance of a train buffer beam under random loads was evaluated by the AA6082 welded joint S-N curve, whose lifetime did not satisfy design requirements. Then, the surrogate model of a train buffer beam for anti-fatigue optimization design was constructed based on the response surface method. The fatigue life of the optimized train buffer beam increased by two orders with the use of the genetic algorithm to find the optimal solution in global variables. [ABSTRACT FROM AUTHOR]

Given the serious work hardening phenomenon, high chemical activity, severe tool wear, and poor workpiece surface quality during the cutting of titanium alloys, placing micro-texture on the tool surface decreases the tool-chip contact area and friction and plays the role of wear resistance and friction reduction. In order to improve the cutting performance of cutting tools, the present study examins the effect of micro-texture parameters of a cemented carbide ball-end milling cutter on cutting force, tool wear, and workpiece surface roughness and establishes a regression analysis model via regression analysis. We consider the cutting performance of a cemented carbide micro-texture ball-end milling cutter as an evaluation standard. A genetic algorithm is applied to the multi-objective optimization of micro-texture parameters, and a regression analysis model of micro-texture design criteria is established. This provides a standardised reference to select standard micro-texture parameters in the design and preparation of a cemented carbide micro-texture ball-end milling cutter. [ABSTRACT FROM AUTHOR]

This paper proposes an automatic multi-objective optimization tool for the optimum design of a cracked pressure vessel with composite coating based on the non-dominated sorting genetic algorithm (NSGA-II) and maximum tangential strain energy density (MTSED) criterion. The proposed tool takes into consideration all the important parameters of the optimization of composite materials such as the angle, thickness, number, and material of each layer, as well as the effective parameters of fracture mechanics such as the orientation, length, and aspect ratio of cracks in the metallic liner of vessels. The optimization processes have been conducted for 3D longitudinal, circumferential and inclined semi-elliptical cracks in the sample composite pressure vessel and optimal stacking sequences and Pareto's front for each kind of crack are presented. The in-plane and out of plane fracture envelop predicted by the MTSED criterion before and after optimization under mixed mode I/II/III have been investigated. The proposed tool is validated against collected results from existing literature and these results have largely agreed with each other. [ABSTRACT FROM AUTHOR]

Numerical analysis of intake and exhaust systems in diesel engines has been used to reduce harmful emissions from them. An unsteady gas flow analysis in 3D format of the intake and exhaust systems was inefficient due to the need for a high-resolution workstation and enormous time for analysis. Therefore, a 1D-3D coupling algorithm was developed to resolve this issue. The algorithm was based on the method of characteristics coded in the C language for ID, and utilized ANSYS FLUENT R15.0, a commercial computation fluid dynamics program for 3D. The 1D-3D coupling method employed a User-Defined Function to share the computation between ID and 3D. Numerical analyses of unsteady gas flow in a pipe system were performed under four conditions including ID, 3D, 1D-3D and 3D-1D coupling. As a result, the characteristics of gas flow including pressure and velocity distributions were similar and the calculation time of 1D-3D coupling model was 11.46 times faster than in 3D. Therefore, the proposed algorithm was strongly expected to enhance the results and shorten the calculation time. [ABSTRACT FROM AUTHOR]

This study presents the optimum design of a ported shroud of a compressor for a hydrogen reciprocating engine of a high-altitude unmanned aerial vehicle. The design aims to expand operating range and improve efficiency by replacing only the casing of the turbo-charger compressor while maintaining the current engine system. Four design variables were selected to determine the slit shape of the ported shroud. Factors, such as efficiency, pressure ratio, and stall mass flow rate at the maximum speed and cruise condition, were defined as objective functions and constraints. A neural network model was constructed, and a multi-point multi-objective optimization design was achieved using a genetic algorithm. k-means clustering was applied to the optimal solutions. Results showed design A with a 4% reduction in stall mass flow rate and design B with a 2% improvement in efficiency under the cruise condition. The performance of each design was confirmed from off-design analysis. The cause of the decrease in stall mass flow rate in design A was investigated by observing flow recirculation from the slit to compressor inlet. [ABSTRACT FROM AUTHOR]